PRESIDENTIAL COMMISSION ON THE
HUMAN IMMUNODEFICIENCY VIRUS EPIDEMIC

HEARING ON INCIDENCE AND PREVALENCE

Department of Health and Human Services
First Floor Auditorium
330 Independence Avenue, S.W.
Washington, D.C.

Friday, December 11, 1987

COMMISSION MEMBERS PRESENT:
ADMIRAL JAMES D. WATKINS (RETIRED), CHAIRMAN
JOHN J. CREEDON
THERESA L. CRENSHAW, M.D.
RICHARD M. Devos
KRISTINE M. GEBBIE, R.N., M.N.
BURTON JAMES LEE III, M.D.
BENY J. PRIMM, M.D.
PENNY PULLEN
CORY SERVAAS, M.D.
WILLIAM B. WALSH, M.D.

POLLY L. GAULT, EXECUTIVE DIRECTOR

 

 
 

I-N-D-E-X

WELCOME

Adniral Watkins

PANEL THRE

Presentation by Dr. Sheldon Landesman
State University of New York
Health Science Center

POPULATIONS AT RISK

Presentation by Dr. Stephen Joseph
New York City Commissioner of Health

Presentation by Dr. James Wiley
University of California Berkeley
School of Public Health

PANEL FOUR

COMMUNITY BASED ORGANIZATIONS PERSPECTIVE

Presentation by Mr. Don Edwards
National Minority AIDS Council

Presentation by Dr. Jane Delgado
National Coalition of Hispanic Health
and Human Services Organizations

Presentation by Mr. Jeff Levi
National Gay an Lesbian Task Force

PANEL FIVE

STATISTICIANS/MATHEMATICAL MODELING

Presentation by Dr. Victor DeGruttola
Harvard University, Department of Biostatistics

Presentation by Mr. Michael J. Cowell
UNUM Life Insurance Company

Presentation by Mr. Anthony Pascal
RAND Corporation

Presentation by Mr. Anthony Robertson
Research Testing and Development Corporation

ADJOURNMENT

APPENDIX

Submitted Testimonies

PAGE

164

166

212

235

243

250

262

269

278

280

311

 

 
P-R-0-C-E-E-D-I-N-G<S

[9:04 a.m. ]
OPENING

MS. GAULT: Good morning. Ladies and gentleman,
members of the President's Commission, my name is Polly Gault. I
serve as the designated federal official. And, in that capacity,
it is my privilege to declare this meeting open. Chairman
Watkins.

WELCOME

ADMIRAL WATKINS: Good morning. I am pleased to again
have this opportunity to welcome my fellow Commissioners, our
panelists, and the audience to the second day of a two day
meeting of the Presidential Commission on the HIV epidemic.
Yesterday we were privileged to receive some very important and
thoughtful insights from a variety of experts who shared with us
their views on incidence and prevalence of HIV infection in the
United States, the focus of these two days of hearings. As I
noted at the beginning of those proceedings, the Commission is
committed to moving ahead in addressing several issues we feel
demand early consideration.

In our preliminary report that was submitted to the
President on December 2, we indicated that the incidence and
prevalence of HIV infection is one of our first priorities. Two
of our Commissioners, Mr. Richard DeVos and Mr. John Creedon,
have taken a leadership role in putting together these panels to
enable us to gain a better understanding of many of the factors
related to the spread of the disease.

Yesterday we heard testimony on the staging of the
disease and the quality of the epidemiological data collected at
various stages, as well as a cross section of perspectives from
the public health community. Today we hope to further focus our
deliberations and gain some additional knowledge about the
behavior and risk factors of individuals who have become infected
with HIV, and ways in which positive intervention might alter the
course of the infection in those at risk.

We also intend to carefully examine the mathematical
methodologies and tools designed to provide accurate projections
of epidemiological data. We established some effective operating
procedures at yesterday's sessions that we will follow today, and
I feel it would be appropriate to continue in that manner.
Without objection from any of the Commissioners, I will now turn
over the Chair to Mr. John Creedon, who will act as Chairman for
today's hearings. Mr. Creedon?

164

 

 

 

 
 

 

 

MR. CREEDON: Thank you, Mr. Chairman. Yesterday and
today we are trying to determine the answer to a number of fairly
basic questions, the first of which is how many people have the
virus now. There have been different estimates made about that,
there have been different tests that have been undertaken, there
are some that have been planned. But one of the key questions
for baseline purposes in trying to determine where we stand is
how many people have the virus now. In that regard, one of the
questions is how reliable are the tests that are administered to
determine whether or not the virus is present. We received
testimony yesterday that suggested a very high degree of
reliability of the tests if they are properly administered, and
if the laboratory is a good laboratory. I thought that the
testimony yesterday was quite good on that subject.

With respect to how many people have the virus, I
thought there were different conclusions that could be drawn. If
you look solely to the tests that have been administered by the
military, which involves a certain of group of people -- people
who apply to the military, or people who are already in the
military -- that particular group suggests a fairly low incidence
of the virus.

If you look at other groups, especially the high risk
groups, the homosexual groups or the IV drug user group, a much
higher percentage of those groups. One of the difficulties is
how do you get a representative group and test that .
representative group. If you have any suggestions or comments
with respect to that subject, we would very much appreciate it.
Another question that we looked at, and are trying to get the
answer to is after someone incurs the virus, how long does it
take before there is some manifestation of the disease that
requires medical treatment.

With respect to that question, Dr. Curran from CDC had
some charts which show a fairly good factual basis for concluding
that something on the area of 30 percent of the people have
gotten some manifestation of the disease in a specified period of
time; but that going beyond the data that they now have, they
don't know what percentage will get the virus.It could go
straight up, as the line has been going up, or it could start to
flatten. Again, if you have any judgment or estimate on those
subjects, I think it would be very helpful. We are interested in
knowing more about the vulnerable groups, the high risk groups.
What are the characteristics of those groups that result in their
being vulnerable? What is your judgment as to whether the
disease is likely to spread to the heterosexual community, other
than those who are involved with the IV drug users, and so forth?

One of the questions we were particularly interested in
yesterday was do we need, as a Presidential Commission, to

165

 
 

recommend studies other ,than those that are presently planned.

We had a group of doctors yesterday who said that if you could do
a country-wide study, that would be a good way of determining the
incidence of the disease; but they recognize that there were
obstacles in doing such a study, in that people might not
cooperate unless they were assured of anonymity, et cetera, et
cetera.

Any views that you have on that subject would also be
appropriate, and especially your views on the reliability of the
data that we have as a basis for our trying to determine and make
recommendations to the President. I think we recognize that this
is a very complicated area; that, to a certain extent, it is
fluid. It may be changing from month to month or year to year;
that there are conflicts of opinion on what the situation is.

We had testimony from Dr. Langmuir yesterday, who used
to be the Chief Epidemiologist at CDC. He is retired, and has
done extensive studies over many years -- 50 years -- on the
question of epidemiology. It is his judgment that, instead of
the line going up as it has been going up, that it is going to
flatten and curve.

These are the kinds of differences of opinion that we
are faced with at this juncture. We would appreciate any light
that you can shed on them. Our first witness this morning is Dr.
Sheldon Landesman, from the State University of New York's
Health Science Center. Dr. Landesman?

PRESENTATION BY DR. SHELDON LANDESMAN

DR. LANDESMAN: Thank you. It is a pleasure to be
here. I hope the information that I convey here will be of some
use to the committee. Specifically I did not have my prepared
testimony directly related to some of the questions that you had,
but please feel free to ask about them as I go through it.

My background is that of an infectious disease person
in the area of epidemiology, working in a large, inner city
hospital in New York City, which has on the average of 50 to 60
AIDS patients in house at any one time. We deal with drug
addicts, sexual partners of drug addicts, women who are infected,
children who are infected; and largely the problems of the AIDS
epidemic in the urban, inner city poor.

What I had planned to talk about here was the issue of
heterosexual transmission of HIV. My purpose is to try to paint
a somewhat different picture of the issue of heterosexual
transmission, and clarify some of the concepts and the terms and
the magnitude of this problem. Because in part much of the
confusion has to do with the terminology that is used around
this issue. I have about seven or eight points, and a few brief

166

 

 
 

 

recommendations to make on it. Let me start, then, with point
number one, which is the concept of the spread of HIV through
different phases of the disease, or different phases of spread.
We can conceive of the human immunodeficiency virus as being
spread in three separate phases: primary, secondary, and
tertiary.

In phase one, we have the introduction and spread of
the virus into Gay men and intravenous drug users. This is the
primary phase of spread with Gay men and drug users -- and here I
mean intravenous drug users, past or current -- constituting the
primarily infected populations. The secondary phase of spread is
spread from the infected drug users and bisexual men into their
non-drug using female partners. This secondary phase is
characterized by heterosexual spread of the virus from a
population of infected men to their non-drug using female
partners.

Although female-to-male spread of HIV will also occur
in this phase, it is less likely to do so, simply because the
reservoir of infection is largely male. The tertiary phase, or
"spread out of the risk groups," occurs when there are a
sufficient number of non-drug using infected females to infect or
feed the virus back to non- drug using heterosexual males. These
men can obviously, then, infect more women and then the process
could become self-sustaining, which is a question that has
occupied many people.

Confusion arises because of the inadvertent blurring
together of the secondary heterosexual phase of spread with the
tertiary phase of the epidemic. What the population at large,
and many governmental or public officials mean when they refer to
heterosexual transmission of HIV, is spread of the virus out of
the risk groups, or the tertiary phase of spread. Currently
there is little hard evidence to indicate that tertiary spread of
HIV is occurring. However, the secondary stage of heterosexual
transmission, spread from drug users -- and, to a lesser extent
~~ bisexual men into their non-drug using female partners is
occurring in certain areas of the country in certain populations
at a significant rate.

Point number two. There already exists in the United
States a large pool of infected heterosexual persons, and these
people will also be a source of spread to other heterosexual
persons. Cases of AIDS and HIV infection are classified by risk
group: Gay men, drug users, et cetera. This classification
identifies how persons have acquired the virus. If we look at
these risk groups carefully, we see that approximately 25 percent
of all persons in risk groups are themselves heterosexual, the
single largest group being current or former drug users. The
point that I wish to make here is that when we look at persons
with HIV infection, we must not only look at how they acquire the

167

 

 
 

 

virus, but also how they spread the virus. The potential for
heterosexual transmission is minimized by focusing on the
mechanism by which drug users acquire their infection:
generally, the sharing of infected or contaminated needles. We
must remember that most drug users are male, are heterosexual,
and can thus spread the virus two ways: through heterosexual
contact, and by the sharing of needles.

The recent CDC report to the Domestic Policy Council,
done in conjunction with the National Institute of Drug Abuse,
estimates that there are 900,000 to 1.4 million infected
persons. At least 25 percent of these persons are heterosexual.
Thus, the pool of infected persons who are heterosexual and
capable of spreading the virus heterosexually ranges from
225,000 to 350,000 people. This is a large number of persons.
The majority of these infected persons, again, are drug addicts
and male. Their partners are generally non-drug using females.
Thus, the infected male addict will be the principal vector or
principal source for the spread of HIV into non-drug using
heterosexual females.

We are thus currently faced with a large reservoir,
equal or exceeding a quarter of a million heterosexual
individuals, who are chronically infected with this virus. The
impact of this population on future generations of women and
children is just starting to be felt.

Point three. Current estimates are that somewhere
between 30,000 and 115,000 persons have already been infected
through heterosexual spread. A natural consequence of the large
reservoir of infected men is the infection of women who live in
close sexual proximity to these men, and who interact sexually
with them. This brings us to the problem of HIV infection in
women and children. In our own hospital, Kings County Hospital
in Brooklyn, New York, two percent -- or one in 50 <= of all
women who give birth are infected. Approximately half of these
women are infected via heterosexual activity, the remainder being
drug users themselves.

At other hospitals in New York City, the rate is even
higher. At Grady Hospital in Atlanta, Georgia, more than one
percent of the women who deliver are infected. In a recent study
in Massachusetts, the rate of infection among women who recently
delivered range from one in 1,000 in the rural areas to nearly
one in 100 in the inner city areas. In San Juan, Puerto Rico,
1.7 percent of the women coming in for prenatal care were
infected. In Alameda County, California, 0.5 percent of women
tested positive. Other studies of a revealing nature include a
nearly one percent rate of positivity in a Baltimore family
planning clinic, and a .7 percent rate of positivity in a
prenatal clinic in Jacksonville, Florida.

168

 

 
 

 

These numbers represent a mixed population of women,
some of whom are infected via drugs, some of whom were sex
partners of infected men. Infected drug using women will, over
time, comprise a decreasing proportion of the total number of
infected women, while the number of women infected through
heterosexual transmission will gradually and inexorably increase
as a consequence of the sexual activity of the large reservoir of

infected men.

Estimates derived from the CDC document indicate that
the number of persons infected through heterosexual contact is
already significant. These estimates are in the range of at
least 30,000 to 115,000 persons. This number thus represents the
current extent of the secondary phase of HIV spread, and defines
the current magnitude of the heterosexual transmission problem.

Point four. The terminology that we use to classify
people with HIV infection masks the potential for heterosexual
transmission or perception of heterosexual transmission.

The existence of more than a quarter of a million
infected heterosexual persons, and 30,000 to 115,000 persons
infected through heterosexual contact is partially hidden by the
terminology we use to describe the epidemic. For example, it is
often stated that HIV disease is not spreading outside of the
risk groups. While this statement is technically correct, it
contains a certain element of circular reasoning. Specifically,
persons who are heterosexually active with another risk group
member ~- such as a bisexual man or a drug user -- and who then
acquire the disease, are themselves defined as being in a risk
group. Thus, partners of drug users, female sex partners of
bisexual men, et cetera, constitute a risk group.

Our terminology then hides the concept of heterosexual
transmission. By classifying sex partners of bisexual men or
drug users as a risk group, we count these cases that occur in
this manner, state that these persons are in a risk group, and
then assure ourselves that the disease is still staying within
our self-defined risk groups. This may calm the population but,
in fact, does little to slow the spread of heterosexual
transmission of the virus.

Point five. The consequences of heterosexual
transmission of HIV will fall most heavily on the poor. The
majority of the drug using population reside in the inner city
areas. This population is largely composed of Black and Hispanic
persons. Thus, the problem with heterosexual transmission of HIV
is largely -- but not exclusively -- a problem of the urban
minority populations. Inner city urban minority women who live
in close social and sexual contact with infected male drug users
are especially at risk for acquiring the disease.

169

 

 
 

| . Infection in this population is a terrifying addition
to the equally depressing, already existing conditions of
poverty, unemployment, unemployment discrimination, and drug
abuse. The threat to this population is large. As noted
earlier, 30,000 to 115,000 persons are estimated to have been
infected through heterosexual transmission -- the largest number
clearly being people of color. The rates of seropositivity in
pregnant women quoted earlier give stark evidence to this fact.
Most of the studies cited earlier are from inner city populations
composed almost exclusively of people of color. Blacks and
Hispanics clearly represent a disproportionate number of AIDS
cases. Excluding homosexual men, the likelihood of acquiring
AIDS if you are Black is 12 to 13 times that of a White person,
while 80 percent of all the children born with AIDS come from
minority populations.

Point six. Heterosexually transmitted HIV is a disease
of the family. Preeminently among diseases, HIV infection within
the heterosexual community is a family affair. The usual
scenario is that of an ailing or dead infected male drug user who
has infected his non-drug using wife or girl friend. The wife or
girl friend may have already given birth to an infected child, or
may give birth to one in the future. Often the infected women is
already clinically ill, and usually has two to three young
children at home.

Thus, we are faced with a seriously ill male, largely
unable to care for himself, an infected or ill partner with young
children at home who -- if the children are lucky -- might be
free of HIV infection, but still suffer the serious effects of
having ill or absent parents. All of this occurs within an
environment and social circumstance of poverty, general poor
health, inadequate housing, a fragile family structure, and the:
near-total absence of significant resources or support systems.

It cannot be emphasized too strongly that HIV in this
context destroys families. The children caught in this situation
lose on two counts. Either they are born infected.

Approximately one-third to two-thirds of children born to
infected mothers will themselves become infected, or the
infected parents will become ill, and the children often end up
in foster care institutions as boarder babies in hospitals or
some other secondary home setting.

Added to all the above is the misery and the pain
associated with being an HIV-infected child. The Coolfont report
published in 1986 estimated that there would be 3000 cases of
pediatric AIDS in 1991. Since there are three to four
HIV-infected infants for every case of AIDS in children, we can
anticipate 10,000 children who have acquired the disease through
perinatal transmission. The problem of perinatal or pediatric
AIDS, already severe, will only grow with time. The reservoir of

\

170

 

 
 

 

infected heterosexual males and the fact that 1 to 3 percent of
pregnant women in many inner city locations are currently
infected tells us that pediatric and perinatally acquired HIV
disease will be an enormous problem in the upcoming decade.

Point Seven: Reaching the communities most affected
by heterosexual transmission will not be easy. Care, tact,
boldness, and a sense of urgency must imbue our attempts to do
so. The minority communities are justifiably concerned about
their being labeled and stigmatized as another risk group. They
fear that any campaign directed towards HIV counseling and
testing could turn into a eugenics program with forced
sterilization or coerced abortion. Both of these fears have a
strong basis in reality. The discrimination and stigmatization
surrounding populations at risk for HIV disease is well known.

Less well known but well documented is the history of
forced sterilization imposed upon minority populations in earlier
years in this country. Further, these communities are concerned
that campaigns directed towards condom use or sexual abstinence,
which are designed to slow transmission, represent subtle and not
so subtle campaigns of racial genocide or denials of procreative

rights.

In approaching the problem of heterosexual
transmission of HIV infection in the ghetto areas of our country,
understanding and boldness will be required by all concerned.
The political elites and governmental authorities who allocate
money and set policy for this problem must understand that the
black and Hispanic populations will approach any governmental
solution with caution, if not outright hostility and suspicion.
Patience will be required. The social and political leaders of
these communities must not underestimate the threat that HIV
poses to their constituencies, nor must they shy away from
meeting their responsibility.

Once the level of HIV infection in females reaches a
significant, but as yet undefined level, tertiary spread of the
virus can occur. We already have 1 to 3 percent of females of
childbearing age being infected in certain cities. If the level
continues to rise, the consequences are dire. If HIV infection
progresses any further in these areas, the issue of altering
sexual behavior or use of condoms will not become one of racial
genocide; rather it will become one of racial survival.

Briefly, I would just like to make three or four very
brief recommendations based upon some of this information. One
on the provision of care: Additional resources must be targeted
toward the treatment of drug users. This must be our first
priority. We can never hope to stop the heterosexual spread of
HIV until we've slowed transmission among and by drug users.
Other speakers on other panels will more directly address drug

171

 

 
 

 

users in greater detail, but I cannot stress the importance of
that issue around the issue of heterosexual transmission. I
can't stress it too much. One point to remember within the
context of resource allocation for drug users is that they are
not popular persons in this country or in our society.
Nonetheless, we treat and care for them because our societal
ethic demands that we care for all of our citizens. Further, it
is critical to understand that we must provide resources for the
drug users, if for no other reason than our own narrow
self-interest as we attempt to stop heterosexual transmission of
HIV.

Point Two: Regional comprehensive care programs for
women and children infected with HIV should be created. These
programs would have the following purposes: One, to educate the
women at risk. Two, to provide a linkage between the
community-based support groups, and I stress we must generate
more community- based support groups -- provide a linkage between
the community groups and the hospital-based programs, thereby
lessening the fragmentation of health care that is so common in
the inner city areas. Three, such programs would help to
stabilize the fragile family structure which is under severe
stress when HIV disease enters the population. And four would be
an attempt to decrease the number of homeless infants or boarder
babies and children born to HIV- infected mothers, which will
become a problem of increasing severity. The only way to do this
is to develop an integrated comprehensive care, multidisciplinary
program around the issue of women and children.

Briefly on the point of education, culturally
appropriate educational programs should be targeted to the female
populations at highest risk of HIV disease, including the
widespread availability of voluntary HIV counseling and testing.
Such programs must be accompanied by additional support
services. You cannot tell a young woman she is pregnant and
infected and then just have her leave your office.

Campaigns advocating increased condom use are useful,
but a broader approach is needed, especially in the inner city
areas. Women must be made aware of the high risk attendant to
sexual activity, particularly with a person who is a current or
former addict. A message of sexual selectivity must be
incorporated into educational programs. The recent debate on
condom promotion versus promoting abstinence is a false issue.
It is possible to steer a middle course.

Four, educational campaigns and grants should be
subject to critical analysis as to their effect and impact.
Exhortation, history tells us, rarely changes behavior.
Behavioral scientists, including anthropologists and
sociologists, should be drawn into the creation and evaluation of
educational programs. We must learn if a message sent is a

172

 

 
 

 

message received, and if received, if it's acted on. And five,
which is an area which I must admit I have some of my own
interest in, is that more information is clearly needed about the
issues of heterosexual transmission, its incidence, prevalence,
causative factors, behavioral and biological, and finally
prevention. Additional behavioral, biological, and
epidemiological research in this area should be done.

As an example, the studies now planned on the
seroprevalence of HIV infection in women should be continued.
The currently planned national study for HIV infection of new
mothers, done by testing a drop of blood taken from their newborn
infants, uses a relatively unbiased method to give a fairly good
picture of HIV infection in women of childbearing age. This type
of study would be very useful for public health management and
HIV surveillance in the heterosexual population. Thank you for
the opportunity. I'll be glad to answer any questions.

MR. CREEDON: Thank you, Dr. Landesman. That certainly
is a clear and sobering and somewhat depressing report. From
what you indicate, it seems to be clear that there is
heterosexual transmission from infected males to non- infected
females, and especially in the IV drug user situation. There
have been articles written and which have come to the attention
of some of the Commission members at least which suggest -- and
this goes back to your very early point about how do you classify
groups -- that transmission heterosexually is much more difficult
than transmission homosexually, because of the physical
characteristics of the differences between those forms of
intercourse. And I don't know if you have a view on that, but
certainly from the standpoint of the Commission and trying to
evaluate the likely size of the group that society is going to
have to contend with between now and the year 2000, one of the
questions is whether the disease is likely to spread outside of
the partners of the IV drug user and into the heterosexual
community generally. And I wonder if you have any view on that
subject? |

DR. LANDESMAN: Two points here. One, the issue of
whether HIV is spread more readily through anal/genital sex
rather than vaginal intercourse, I think by and large is a false
issue. If there is a difference in the rate of transmission
between anal/genital and vaginal intercourse, it is probably not
one that is orders of magnitude difference. We're not talking
about something that is 100 times more efficient or 1000 times
more efficient one way or the other. If one would say that the
risk of anal/genital intercourse, just to take some theoretical
numbers is 1 in 100, and the risk of transmission through vaginal
intercourse is 1 in 200, from a public health point of view, that
largely doesn't make a significant difference in terms of where
we are going.

173

 

 
 

Overall, the partner studies have indicated that the
long-term sexual partners of HIV-infected persons, the long- term

_ heterosexual partners of HIV-infected persons, run roughly a 20

 

to 40 percent chance of getting infected from their infected
partner. Those numbers probably will not reach a greater degree
of clarity over time, given the variables that we have. But we
do have to remember that it is heterosexually transmitted.
Whether it’s one in 200 or one in 250 is largely irrelevant.

Concerning the issue of the spread of the virus outside
of the risk groups, the point I tried to make was that I think we
to some extent delude ourselves if we keep talking about the
issue of risk groups, that ultimately everybody who is infected
sexually will, if you were to try and study them carefully, be in
a risk group, because obviously their partner will have had to
have been infected. And seeing that we define sexual activity
with an infected partner as putting you in a risk group, every
person who would then get infected would have had to have sexual
activity with an infected partner.

It is probably true that the majority of heterosexual
transmission at this point in time will stay within this
secondary phase of transmission, from drug-using persons or
bisexual men into their sexual partners. We will not get to the
level of tertiary transmission or outside of that until we build
up a sufficient reservoir of people who are infected through this
secondary method of transmission.

So I think the purpose and the goal in terms of
preventive efforts at this point in time is to try to slow down
heterosexual transmission from what are now called the primary
risk group people into their non~-drug-using partners. That would
be the preventive strategy at this point in time, because it is
only when there is a sufficient reservoir of infected women who
are, let's say, non-drug-users, who then start to feed it back to
non-drug-using men, that the process could then become a tertiary
or self-sustaining process.

I don't think we have good data on that at the present
time. I don't see anything which indicates that is occurring at
a significant rate at this time, although my great concern is for
the urban inner city populations where you are building up a
large reservoir of infected women, and even if the rate of
transmission from females to males is relatively low, once the
reservoir of infected --

MR. CREEDON: Is that the fact, in your opinion?

DR. LANDESMAN: Well, again, I don't think we're
talking about orders of magnitude difference here. It is
probably true. that receptive intercourse, receptive intercourse
in males and receptive intercourse in ordinary, everyday

/

174

 

 
 

 

heterosexual vaginal intercourse, receptive intercourse is
probably a more efficient means of transmitting the virus. But
we're not talking about orders of magnitude difference here.

Would I sleep with an infected female? No. Is it
safer than sleeping with an infected male? Marginally. Again, I
think we are ending up splitting hairs around the issues of the
efficiency of transfer because we're not dealing with orders of
magnitude difference.

It's clear that the virus can be spread bisexually.
It's clear that it is a sexually transmitted disease. And the
critical question, seeing as we're dealing with tens of thousands
of infected people, is whether we are going to ever develop a
large enough reservoir of infected females largely to then start
to feed it back into non-drug-using men.

In places like New York City, especially in certain
areas of the city, that's a concern that could happen, as it is
in other areas of the country. I think our goal from a
preventive strategy is to try to limit or curb the number of
women who become infected from their partners, because that's the
way to prevent heterosexual transmission from, quote, breaking
out into the general population, unquote.

MR. CREEDON: Dr. Landesman, one of the points you made
was that you thought there should be further research and study,
specifically in the area of heterosexuals. Is that being done
adequately? Do you think plans are afoot now to do that, or do
you have any specific recommendations as to exactly what we
might recommend in that area?

DR. LANDESMAN: Well, there are a variety of
initiatives now going on by the Public Health Service. I'm not
privy to all of them, but I know that there are a significant
number of studies.

MR. CREEDON: This is Dr. Axelrod or nationwide?

DR. LANDESMAN: No, no. This is NIH. In other words,
studies on the issue. I think clearly where more information is
needed are two separate areas. One is going to be in the area of
education. I mean, we really know -- I mean, none of the really
sophisticated educational methodologies or processes have been
applied to the issue of education in HIV disease, particularly in
dealing with the populations most affected. There's very little
that I have seen that's been built into educational programs
which evaluate how effective they are, and it seems to me that
all of the research has to have a sort of serve-or-follow-up
mechanism of you do it, you evaluate it, and then you adjust what
you do to make it more efficient. And I think that has to be a
critical part. We have to learn as we educate. We obviously

175

 

 
 

 

can't wait, but every educational program that's done should
contain an evaluation segment in them, so we can learn something
from it and then fine tune it.

The second thing, which actually was pointed up by a
recent conference, I think, in Colorado, the Kinsey conference
that just concluded, was that we know little, if anything, about
the sexuality of the groups that are most at risk for
heterosexual transmission. I mean, the CDC document to the
Domestic Policy Council, the sexual study most commonly cited was
the Kinsey Report of 40 years ago, and that's basically all we
have.

We know little about the sexuality of white, middle-
class teenagers and adolescents and young adults. We know
nothing about the sexual culture and mores of the inner city
populations. I mean, literally -- literally zero about that
population. And if we're going to design effective campaigns
around the issue of sexuality in that population, we'd better
learn something about that population as fast as we can.

MR. CREEDON: Do you have any sense about the bisexual
problem, how big that is?

DR. LANDESMAN: I have no real hard data on it. I
think Professor Wiley may have more information on that than I
do. That's not an area that I can say anything on.

MR. CREEDON: Other questions? Yes?

DR. PRIMM: Dr. Landesman, I want to particularly
commend you for certainly pointing out that community-based
organizations certainly need strengthening and to play a greater
role and for pointing out the fact that in heterosexual spread,
this is a family disease. I've talked about that, and I think it
is so very important. And the emphasis that you put on
expanding services certainly for the intravenous drug-using
population, I think that's commendable, and I strongly support
what you have said.

The other thing, my question is, it's been reported in
the literature that the effectiveness of transmission of the
virus is secondary to the pathogenesis of the virus itself, the
ambivalence of the virus, for example, and that in females the
virus has not been present that long as to become so virulent
that the effectiveness of transmission is there as it is fron,
say, male to male or male to female. I would like for you to
comment on that, if you would.

DR. LANDESMAN: I'm not aware of anything which
suggests that the residence of the virus in females or the
duration of the residence of the virus in females somehow affects

176

 

 
 

 

their ability over time to transmit it to males. The evidence
from Africa and Haiti do not seem to suggest that that's the case
at all. So my answer is, I'm not aware if that, in fact, would

be a true.

DR. PRIMM: Have you heard that? Have you heard that
before?

DR. LANDESMAN: I've heard something similar to that.
One point in general is, there are many discussions that go on
around the issue of whether anal intercourse is more efficient
than vaginal intercourse, whether females can transmit it more or
less efficiently to men than men transmit it to women.

The point I wish to make is, I think many of those
questions are by and large minor in nature. I think there is a
sufficient amount of data to indicate that we're not dealing with
major orders of magnitude differences in these things, and that
those should not be the principal questions that we should ask,
but we should focus more, we should accept this as a sexually
transmitted disease bidirectionally and focus more on issues of
prevention of the epidemic.

MR. CREEDON: Dr. Crenshaw?

DR. CRENSHAW: I want to first compliment you on your
thoughtful, clear, and exceedingly concise reflection of the
facts. I think I've rarely heard them so well expressed.

DR. LANDESMAN: Thank you.

DR. CRENSHAW: I want to ask you a very rudimentary
question that's exceedingly frustrating to me. And put aside for
a moment all the complex mathematical medels that are being dealt
with that I have some difficulty following.

You indicated that 225,000 to 350,000 incidents of
infection already in the heterosexual population, and CDC puts
those figures lower. And my point and my question is, even if
they were a fraction of CDC's and your consensus, given the fact
that a mere five years ago, we had 183 cases in the homosexual
community, and today we have such a magnitude there that in major
cities, the very gay community is endangered. My question is why
in this day and age are we even raising the question: Is this a
threat in the heterosexual community? It seems to me that common
sense, not mathematical manipulations, tell us loud and clear
that it is definitely one. It already is one, and it's not a
question that it's going to be one. So tell me, if you can,
because I can't answer this question, why we are raising the
question at this late day and age?

177

 

 
 

 

DR. LANDESMAN: Two points. What I said in the
testimony was that there appear to be at least a quarter of a
million persons who are infected who are heterosexual and thus
capable of spreading the virus heterosexually. I think the cpc
probably agrees with that, with that number. That's HIV
infection, not cases of heterosexual spread. As to why we are
debating the issue, that's more of a political and societal and
communications question than it is, I think, a scientific one.
But I can give you the benefits of my thoughts as I sit in
Brooklyn. And that is that the issue of heterosexual
transmission is a very politically laden topic, and many
thoughtful people on this issue have attempted to walk a very
thin and careful line, which is an attempt to develop concern in
the public without developing panic in the public, and that's a
very difficult line to walk. On one hand, you want to generate
both preventive measures, concern, and development of an
allocation of resources. On the other hand, there is fear that
you can cause panic with widespread discrimination and isolation
of the people infected.

So it becomes a very delicate political and
communications problem of how you tell the public there is a
problem, how the problem is, in fact, really not one epidemic,
but a variety of sub-epidemics in different areas, without
panicking them, and it is a very difficult problem to deal with.
It's perhaps the most single difficult one, and much of our
difficulty in concepts comes out of people going back and forth
between those two things, urgency and we have to do something on
one side versus fear of arousing inappropriate action on the
other side.

DR. CRENSHAW: I think those points are well taken, and
the follow-up question that you've obviously already given a lot
of reflection to this is, given the danger of not facing reality
due to all of these obstacles along the way and sensitive points,
do you have any recommendations on how to get beyond this
combination of denial systems and political maneuverings so that
we can confront reality more effectively and more efficiently
without engendering the panic and the negative consequences that
people fear?

DR. LANDESMAN: Now we're getting into philosophy. My
sense of how to deal with the debate in terms of its public
nature is that we have to try to stop separating out the
infected from the uninfected, and that is what much of the
terminology of the risk groups has done. It has placed in the
minds of the public the concept that they are they, the risk
groups, the infected, versus us, the noninfected. What happens
is that when the general public or "the general populace" thinks
they are at risk, then they are going to go to their state and
federal government and demand severe and what they would think
would be rapid solutions to protect themselves, which may

x

178

 

 
 

 

involve coercive state powers. That's the wrong way to deal with
this epidemic from a psychological point of view. I think in
terms of psychology or how we talk to the public at large, we
have to change our frame of reference and put this as a societal
problem where all of us are in it together. All of us have to be
willing to give up a little bit in order to protect the group as
a whole.

As an example of how the terminology ought to be
changed, is the question, which I oftentimes hear in the State
of New York asked by news media of a public official, is the
epidemic spreading to the general population? The officials say
with the best of intentions because they don't want to frighten
the population, they say no, it's not spreading to the general
population. It's still within the risk groups.

I would say that both the question and the answer are
incorrect. The question is incorrect because the persons
infected are part of the general population. Unless the person
answering the question points that out, the concept that the
public takes away from that question is, well, there's me in the
general population uninfected, and there's them over there who
may give us the disease. It fosters that divisiveness, that
separation in our society rather than bringing us together as a
societal group.

Then when the public official says, no, it's still in
the risk group, again with the best of intentions because we
don't want to frighten people into coercive and severe actions,
that calms the public for the moment and lets them think they're
safe, but seeing as the biology of the virus has its own
imperatives, it has news reports and scientific reports come out
showing some evidence of heterosexual spread or spread to the
general public, the public, having heard once that they are safe
and now hearing that they are not, lose faith in their public
officials and tend to demand increasingly severe or radical
actions from then.

So our job in part is really one of communication, of
making this for the public at large to see this as a societal
problem and that we're really all in this boat together, and that
it isn't a matter of AIDS being in the general population. With
1.4 million, 900,000 to 1.4 million people infected, AIDS already
is in the general population. Because it's not our neighbor, our
sexual partner or the person down the street doesn't mean it's
not in the general population.

DR. CRENSHAW: Thank you for your point. I have
noticed that the AIDS virus is a very democratic virus, and I
think the sooner we start treating it that way, the better we'll
be able to contend with it.

179

 

 
 

MR. CREEDON: Dr. Walsh.

 

DR. WALSH: I think, Dr. Landesman, the points that you
make are well taken. The thing that I wondered about was what we
have seen in this disease, as you have outlined, it went first
from a homosexual disease to a minority disease and, of course, I
think it eventually will go to what is referred as the middle
class white population. We're reading about and getting
increasing reports now of the middle class white population of
the senior high school level, college freshmen and so on levels.
Is there any significant data yet collected from that group as to
how much of an impact this is making now in that group?

DR. LANDESMAN: I'm not aware of any significant
reliable body of data around HIV disease in adolescents or young
college students. Others maybe, but I haven't seen any good data
in that area. My own sense is that the impact of this virus,
while present in, so to speak, the white middle class
population, will never be as large and as serious a problem as it
will be in other populations in the inner cities.

I think a point that I really wish to repeat is that
the focus of the preventive measures should not be in order to
prevent its spread into the white middle class population. The
fact that it already infects large numbers of persons who happen
to be either poor, minority, depressed or disadvantaged persons
is reason enough, given the numbers of people we already have and
the enormous scope of the problem to drive our efforts.

I think in addressing the public, the problem must not
always be put as one of we're going to protect that sort of
middle class populace. I think we have to change our conception
and our society more than one of just -- this may be an
impossible idea -- but more than just one of protecting
ourselves. But our society is threatened by this thing as well,
and our next door neighbor or the guy a couple blocks away is
threatened, and that's reason enough to put forth the resources.
But in answer to your question specifically, I know of no
reliable data in that population. I would think that we need an
enormous amount of it, especially around the issue of adolescent
homosexuality, because the other group that we now have to look
at is the issue of young men entering the gay lifestyle and how
we can prevent them from reproducing the epidemic that occured in
25 to 44 year aged gay men. That's where preventive measures
have to be focused, especially in the younger age groups.

DR. WALSH: I don't disagree with the need for degrees
of concentration on certain populations, but I'm thinking in
terms of one of the things with which we are grappling is this
so-called political or societal problem of saying it's their
disease. I just have the feeling that one of the reasons we're
not getting the objectives that you would like to get is that so

180

 

 

 
 

 

much of the middle class white population thinks they are above
the threat. I just think we may be surprised. If one is to
believe what we read, when you read of two, three and four
seropositive students in a relatively small class of middle class
whites, I think that there is something below the iceberg that
would be of benefit to increasing the efforts towards the
so-called high risk population. I share completely your
sentiment that there is no such thing, to my mind, in AIDS as a
high risk population. It should be a general thing.

Now, I have one other question, and that is this. Do
you have any comparable data available to you on heterosexual
spread in Western Europe primarily because of their much closer
relationships to their minority populations which come
significantly from Africa where it is a generally heterosexual

disease?

DR. LANDESMAN: I have no hard data on numbers. I
know that in areas of the Netherlands and Amsterdam, there's
great concern about it around the issues of drug users. In fact,
there are large numbers of drug users in several major European
countries who are infected, but I haven't got any hard data. I
don't know of data that really exists. There are probably
ongoing studies around the issue of heterosexual transmission in
those countries.

DR. WALSH: Do you have any adequate explanation or
knowledge of why, for example, in Sweden, which has been
generally accepted as a pretty liberal nation sexually, why the
instance of sexually transmitted disease in general is so much
lower there than, say, in Amsterdam and so on and in the rest of
Western Europe? Because they must be doing something in their
education process that we are not doing, and I think we could
learn a great deal from them. Do you have any idea on that?

DR. LANDESMAN: No, I'm not aware of the public health
programs in Sweden around the issue of prevention of
transmissible diseases.

MR. CREEDON: Dr. SerVaas.

DR. SERVAAS: I want to repeat the compliments from the
other panel members about your good report. I have a question.
If a woman is infected on Saturday night, can she give AIDS to a
different sex partner the following night? I am asked this
question, and I would like your opinion on that.

DR. LANDESMAN: If she's infected on Saturday night,
can she give it to somebody on Sunday night? Actually, we don't
know the answer to that question. The issue of how infective a
person is during different phases of the disease is one that
really is still being explored. Can she give it to somebody six

181

 

 
 

months later? Probably yes. Can she actually give it to

' " somebody the next night? I really don't know the answer to that
question because it may take time for the virus to become part of
her cells and multiply enough. But if you ask me the next night,
I honestly don't know. I would probably actually doubt it. I
mean, do you ask the same question if somebody gets syphilis on
Sunday can they give it to somebody on Monday morning. My
suspicion would be simply because there is a biological time lag
there that's required for multiplication in the person that's
infected before they become infectious to others. One day to
the next, no, but one month to the next or six months later, it
may be yes.

DR. SERVAAS: CDC was telling us if someone comes to
you and wants to be tested and they think they had an exposure
this week, you tell them to wait three months. This is what cDCc
said. Then we would probably be sure the antibodies would be up.
How long do you tell people to wait when they think they've had
an exposure before they are tested?

DR. LANDESMAN: Generally what we do if somebody comes
to us and says they've had an exposure and want to be tested is
that we offer to test them right then and there first to document
that they are negative at that point in time. It also serves a
psychological purpose of thinking that at least something is
being done rather than just sending them out and waiting. Most
of ours come around the issue of needle sticks rather than sexual
transmission, and then we tell them to come back at three and six
months to have a repeat test.

MR. CREEDON: Dr. Gebbie.

MS. GEBBIE: I want to go back to a couple of things
you said about the data early on. If I heard you correctly, one
of the useful things that might be done about the data which are
now coming into state and local and federal public health
agencies would be to categorize two different ways. In addition
to our current categorization which shows how a person became
infected, to routinely categorize and publish by the method by
which that person could spread the disease to others; that would
be a healthy conceptualization to look at how this disease is
spread. Is that accurate?

DR. LANDESMAN: Yes. I think the concept of how large
the heterosexual infected pool is is an important piece of
information that we should have and will give us a concept at
least of our starting point for heterosexual transmission. I
think one or two people from the CDC may have actually started to
do that, but the critical issue around heterosexual transmission
is how large is the pool of people capable of transmitting the
virus. Up until now, that hasn't been counted.

 

182

 

 

 
 

 

MS. GEBBIE: The other question is part of what we're
looking at here is what numbers can we use to base our
recommendations and predictions on. I gathered from your
presentation that you are basically comfortable using. the current
cpc estimates for numbers of people infected. If not, could you
clarify it?

DR. LANDESMAN: I am comfortable with the basic cDC
numbers. I would certainly grant you that they probably are not
perfect, but within the ranges that they have quoted, I am
perfectly comfortable with them. One thing I think we have
sometimes lost sight of in this epidemic as to whether the number
is 1.4 million, 900,000 or even 700,000, which is below the
number that the CDC cited, by any stretch of the imagination it's
an extraordinarily large number considering the other responses
we've had to other public health problems in this country,
whether it be toxic shock, Legionnaire's disease, lime arthritis
or half a dozen others. We're dealing with hundreds of
thousands to millions already with this thing, and it's reached
an order of magnitude beyond which we haven't had from any other
infectious disease that I can remember. In fact, it probably
exceeds virtually all the other major reportable diseases by CDC
already.

MR. CREEDON: Mr. DeVos.
MR. DeVOS: No questions.
MR. CREEDON: Ms. Pullen.

MS. PULLEN: In your paper that is in the Journal of
the American Medical Association of November 20th, you mentioned
the relevance of the prevalence rate in the hospital where you
did your tests as compared to the incidence of perinatally
contracted herpes and neural tube defects. Would you discuss
that for a moment to put that into our record, please?

DR. LANDESMAN: The point we made in the paper where we
documented a 2 percent rate of positivity among women who
delivered or 1 in 50 was that, compared to other diseases which
are perinatally transmitted or other diseases which we offer to
test women for, this disease probably in our population
constitutes cumulatively the single biggest problem that these
women face; certainly several orders of magnitude more common
than the issue of herpes transmission, than the issue of neural
tube defects, than the issue of congenital rubella, than the
issue of congenital toxoplasmosis. So as a comparative health
problem, that was the purpose in there, that in the populations
that we serve, perinatal transmitted disease or the risk to the
fetus is probably cumulatively a greater risk than almost any
other perinatal infectious disease that these women are at risk
for their fetuses.

183

 

 
 

MS. PULLEN: These other diseases are routinely
offered for testing?

DR. LANDESMAN: Routinely offered, yes, ma'an.

MS. PULLEN: The tests are routinely offered. Thank
you.

MR. CREEDON: Chairman Watkins.

ADMIRAL WATKINS: I had a little different view of what
you said about the CDC information that is now packaged in
certain ways, particularly be the heterosexual base, the pool
that you talked about. Are you saying that the way the data is
now presented gives a false sense of security about this
so-called spread into the heterosexual community? If so, how
would you change the architecture of the epidemiological data, or
how would you seek information that would give you the kind of
feeling or give the American people the kind of feeling that we
are clear on this subject of heterosexual spead?

DR. LANDESMAN: The data as it is listed, my.point was
not so much as to how the data was listed but how we then
interpret it. The point I wished to make was that the problem is
the terminology of risk groups, which is a useful terminology for
a variety of reasons. But then we have to expand that
terminology as we discuss the issue, because what you hear is a
discussion of nobody getting the disease beyond the risk groups.
But then we have defined as being a risk group a person who is
sexually active with an infected person. If you start with the
assumption that the only way you can get the disease sexually is
by being sexually active with an infected person, then everybody
who acquires the disease sexually must be in a risk group.

Now, that was the example I gave of the circular
reasoning, and I don't wish in any way to criticize the CDC for
Classifying people in risk groups. My concern is the issue of
how we then use the information to say that persons are not in
risk groups. Eventually everybody who acquires it sexually, if
we do enough investigation, will have been found to have been in
a risk group of some sort. At the present time, it's a matter of
interpretation of that data and how we tell the population about
it. My concern is, again, as much a societal concern as an
epidemiological one. When we go and we keep telling the members
of the public not to worry, it's still in the risk groups, it
continues to foster this them versus us, this risk group versus
nonrisk group mentality and makes it infinitely more difficult to
deal with the problem on a broader societal level. We are
telling people it's not their problem by telling them it's --

184

 
 

ADMIRAL WATKINS: I get the point, Dr. Landesman --
and I agree with you 100 percent. What I'm trying to say is
sometimes the bureaucratic packaging of information flow in
itself can be modulated to give us a better feeling along the
lines you suggest. So while technically you can understand it
and you can begin to translate from the data to something more
logical, it's very difficult for others unless you package it
that way to begin with. So I'm just saying, do you have a
recommended change in the architecture of the epidemiological
report that we could make a recommendation to CDC saying we think
it's very important you present this in this way in your data so
that we avoid the very thing that you're proposing here, which I
think is commendable. So I'm trying to get you to say if you
want to go after the pool that you talked about -- you say you
don't have enough knowledge about that pool and we should be more
sensitive to that, then maybe there's a way to focus surveys.
They're going into 30 major metropolitan areas here in the very
near future to collect survey data.

Are their surveys going to ask the right questions and
give you the data? Is there another way of packaging it so that
it could improve the perception in the nation that this is what I
would call in the Navy an all-hands evolution.

DR. LANDESMAN: I guess two points. One is, as
suggested by Ms. Gebbie and as alluded to earlier, the idea of
starting to talk about the number of persons who are heterosexual
and infected, I think, will give a different view of the problem
in terms of its scope. When we say that there are a quarter
million or more persons in the country who are capable of
spreading the virus heterosexually, yes, they do have certain
demographic characteristics, and that changes the nature of the
perception of the problem.

That would be one suggestion to make. Whether
suggestions could be made around the classifying of data as to
secondary and tertiary spread, I think some steps are already
being initiated along those lines in terms of persons who are
classified as having no identified risk. The people who are
classified as having no identified risk, if appropriately and
intensively interviewed and if after the interview they still are
listed as having no identified risk, the presumption there is
that they were infected through what I called earlier tertiary
spread or transmission from an infected person whom they were not
aware was infected. How we're going to expand or change that
terminology beyond what I've already said, I don't yet have an
answer to. Again, it's part of this delicate problem of trying
to develop real statistics and real concern and urgency without
panicking the public into severe action.

185

 
 

ADMIRAL WATKINS: I think you've opened up a new issue,
to the Commission in general terms today and I think it has been
very valuable, extremely illuminating. I'm just trying to
follow up and get very specific about it. Because if we're
trying to change perceptions and get more realistic without
generating panic, the question is, is the data base adequately
aggregated today in that area, and can that data be obtained by
packaging it in different ways to enhance the national and global
view of this disease in a more realistic way. If you have a
recommendation, we would like to receive it on the Commission.

DR. LANDESMAN: I can't give you a specific one at this
point beyond what I've said. If there's any other information,
I'll be glad to try to communicate it if I can think of it.

ADMIRAL WATKINS: Only one more point. You talked
about obstacles to achieving several recommendations in other
areas that we're not focusing on today. I don't want to ask any
questions in those areas, but I'd like you to come back and tell
us in the areas of care, of education and any specific
suggestions about community-based organizations and health care
organizations, how could they better work together.

We would like to receive information, what you're
really talking about on those issues so we can better prepare for
the follow-on hearings in those areas. Care hearings, for
example, are coming up in January. So in New York in
particular, it would be very valuable for us to know what
obstacles you see among the poor, among the affected individuals
in the area of care, the continuityfrom hospital to nonhospital
care. What are the obstacles, bureaucratic obstacles if you
will? We can't open the residences because we don't have enough
fire exits. That is an issue probably within the local law, but
perhaps those kinds of things would also be very valuable to us,
and obviously you have some feelings in that area. It would be
helpful if you would give us what you see are the obstacles to
getting on with those programs that you feel are so necessary,
either in the education or care field.

DR. LANDESMAN: I can actually tell you, I allude to
many, many obstacles, but one of the actually real live obstacles
among many which we came face to face with a couple weeks ago as
an example was a bureaucratic one which really was simply part of
the mechanism of government. We had down at our medical center
people from five different state agencies around the issue of
setting up a comprehensive care center for infected women and
children. People from the Mental Health Department, from the
Department of Health, from the AIDS Institute, from the social
service agencies, from child and adolescent medicine.

As we put together a package to show them how we would
do such a program which combined community and hospital- based

186

 

 
 

 

services, it was clear that even if they had wanted to help us,
and they did -- they were very supportive -- the bureaucratic
mechanisms cut across different state agencies, and it's so
difficult to try to bring all the state agencies and all the
polices and procedures and laws that are already in effect on
line to create an integrated program that, in fact, is very,
very, very difficult to do so. I think Dr. Joseph can probably
tell you more about that in terms of trying to coordinate care in
New York City across a variety of city agencies as well. It's
genuinely a difficult problem, even with the best of intentions
and the most intensive pull.

ADMIRAL WATKINS: Would you be willing to give us a
follow-up report from your perspective on these areas that you
talked about outside the prevalence discussion this morning?

DR. LANDESMAN: Yes, sir. -
ADMIRAL WATKINS: Thank you.

MR. CREEDON: Our next witness is Dr. Stephen Joseph,
who is the Commissioner of Health from the City of New York. We
heard yesterday from Dr. Axelrod, who is the Commissioner for the
State of New York, as Dr. Joseph is for the city itself.

PRESENTATION BY DR. STEPHEN JOSEPH

DR. JOSEPH: Thank you, Mr. Creedon and members of the
Commission. I certainly appreciate this opportunity to testify
before the Commission. I'll discuss the prevalence and
prevention of the disease with an emphasis on the growing list of
associated policy issues. I presented you some background
material a few days ago, and what I'd like to draw out of that is
that we are once again, as we seem to be about every six months,
at a critical juncture in the epidemic. I think it's important
to remember that because of the biology of the virus, the public
policy decisions we make today determine the state of the
epidemic and its impact on the nation five, seven and even ten
years from now.

The flip side of that coin is, of course, that any
delay today in appropriate public policy actions means that we
will suffer increasingly from the effects of that inaction
throughout the next decade and not just in the next year. Of
all the charges before you, perhaps none is more important than
your ability to stimulate the courageous, farsighted,
comprehensive national strategy that we must have and that we
currently do not have. Nowhere in North America is the tragedy
of AIDS more starkly felt than in my city. We have well over
12,000 people who have been diagnosed now with CDC-defined AIDS.
That's 27 percent of the national total, and we are up past 6,700
deaths. To put that in perspective, when I became commissioner a

187

 

 
 

little over a year and a half ago, we had something like 6,000
diagnosed cases of AIDS in the city, between 6,000 and 7,000.

We have as many as 100,000 additional people suffering
from AIDS-related illness not classified as the reportable
syndrome AIDS. In New York City today, AIDS is the current
leading cause of death among all men age 25 to 44, and I think
very ominously for the future, it is now the leading cause of
death among all women age 25 to 34 in New York City. We believe
we have about 400,000 infected people in New York City,
including some 250,000 gay and bisexual men and 50 to 60 percent
of the city's estimated 200,000 IV drug users. Over the past
three years in New York City, we have strong reason to believe
that the number of deaths among IV drug abusers due to
AIDS-related illness that did not meet the strict CDC definition
of AIDS is actually more than 150 percent higher than we had
previously counted in our surveillance.

MR. CREEDON: Dr. Joseph, where does the 400,000 figure
come from?

DR. JOSEPH: The 400,000 figure is derived from our
samples within gay and bisexual men, samples within IV drug users
extrapolated to that population and our estimates of the infected
sex partners of those groups plus other smaller groups such as
hemophiliacs, et cetera. I'd be happy in the discussion period,
I have current results of about six sero surveys, seroprevalence
surveys, that we just have coming off the line in different
groups that I think will shed some light on the first question
you asked at the beginning of the hearing.

MR. CREEDON: I think that's an extremely important
number, because if you have 400,000 in the City of New York
alone, then --

DR. JOSEPH: Mr. Creedon, I can assure you. We may not
have 400,001. We have somewhere between 300,000 and 500,000
infected people in New York City. I know that as surely as I
know anything else I say this morning.

MR. CREEDON: Thank you.

DR. JOSEPH: Again, as Dr. Landesman mentioned, the
exact number is almost irrelevant. That order of magnitude is a
certainty in the City of New York. The important point that I'm
trying to make at this juncture is, however, that what we have
found in just recent weeks is that there has been a significant
undercounting of the impact of this virus on IV drug users in the
city because of the way the definitions have been set up;
appropriate, but the way they've been set up. We probably have
undercounted severe disease and death among IV drug users by as
much as 150 percent. If we recalculate from that basis rather

188

 

 
 

 

than the former basis, what actually happens in New York City is
that the future that Dr. Landesman talked about is the present
and has been the past for some time, in that we have more severe
illness and death among IV drug users from this virus than we
have among gay and bisexual men from this virus.

The percentages that we talked about previously, about
55 percent or so of disease and death among gay and bisexual men
and 30-some odd percent among IV drug users, we now think need to
be flipped over, and about 55 percent of all the disease and
death so far is really among IV drug users and a lesser percent
among gay and bisexual men.

MR. CREEDON: Excuse me. Dr. Landesman, you're not
leaving, I hope, are you?

DR. LANDESMAN: No.
(Laughter. ]

MR. CREEDON: You're welcome to sit right up there
where you were before.

DR. JOSEPH: Against that background, there are four
central points that I'd like to make about the epidemic in New
York City. First, the numbers of people involved and the costs
associated with the epidemic are going to continue to increase
for the foreseeable future. By the end of 1991, our best current
estimates -- and, again, I'm quite certain of these projections
within orders of magnitude, Mr. Creedon -- Our best current
estimates project that over 60,000 people will have developed
full-blown AIDS in New York City, and 40,000 of them will be
dead.

To put that in perspective, we will diagnose more new
cases of AIDS in New York City in the year 1991 than have been
diagnosed in New York City from the beginning of the epidemic
through the end of 1986. Now, those projections include the
broad CDC definition that I'm sure you've heard about yesterday,
but they do not include the undercounting of IV drug abuser
disease and death that I spoke of a few moments ago.

Second, IV drug abuse is the main route of HIV
infection among addicts, women and children through needle
sharing, sexual intercourse, or from infected mother to child.
We have had over 1,300 cases of AIDS in women in New York City.
Eighty percent of them have been IV drug abusers or their sex
partners. We've had over 240 cases of AIDS in children in New
York City. Most of them were infected by their mothers, 80
percent of whom were IV drug abusers or their sex partners.
Young women represent 40 percent of the 36 cases of AIDS among
teenagers -- You asked that question a few moments ago --

189

 

 
 

compared with only 11 percent of cases in all age range. Half
of those young women with AIDS were either IV drug users
themselves or their sex partners were IV drug users.

HIV infection in heterosexuals is primarily linked to
IV drug users or, to a much lesser degree, bisexual men. We have
little current evidence for secondary or tertiary spread among
heterosexuals. I'm using those terms somewhat differently than
Dr. Landesman did. He would say tertiary spread for what I've
said there. Currently, we have little evidence of that spread;
however, the large existing pool of infected people in New York
City increases the probability of HIV spread through
heterosexual sex, particularly for sex partners of IV drug users
who are mostly women and mostly minority women. I'll give you
the results of those studies a bit later.

MR. CREEDON: Do you have any judgment, Dr. Joseph, on
why there is not this tertiary spread or secondary spread?

DR. JOSEPH: Yes, I do. Well, let me give it to you
now. I believe that what this epidemic really is best viewed as
is a series of telescoping phases. They are not separate phases.
They are not distinct. They are telescoping. At the very early
part of the epidemic, most infection and transmission was
concentrated in gay and bisexual men, but that telescoped very
early with transmission and infection among and to IV drug
users. Somewhat later, the next overlapping phase was the phase
that we really are in now, and that is the phase of the
transmission of virus from IV drug users, mostly men, and
bisexual men. We think they're about 4 to 1, IV drug users,
bisexual men, in terms of those who have infected their
heterosexual partners. That's the phase of the epidemic we're in
currently.

We're just beginning to see the tip of what may be the
fourth stage of the epidemic, fourth phase of the epidemic, which
is an increasing number of what Dr. Landesman referred to as
tertiary spread; that is, heterosexual transmission from people
who are not themselves IV drug users or bisexual men on to other
heterosexual partners. Now, I think most of us believe, most of
us in the public health business, believe that that fourth
overlapping phase will not be the kind of explosive outbreak that
was predicted in the media six to nine months ago, but we also
know that we are going to see steady growth in absolute numbers
in that phase, and I think most of us believe -- again, as Dr.
Landesman said very well a few moments ago -- that fourth stage,
if you will, of the epidemic, will be heavily concentrated in
particular population groups, in particular geographic areas.
What happens after that in what someone will sit here a couple of
years from now and talk about as a fifth stage I think is much
more conjecture. I'11 come back to this.

190

 

 

 
 

 

MR. CREEDON: You're not suggesting that this
Commission is going to still be here five years from now?

DR. JOSEPH: Well, I don't know.

[Laughter. ]

DR. JOSEPH: I'11 come back later, if you will, in
response to your questions and talk about why I think that's so
from the seroprevalence data that we have coming in from the
city.

MR. CREEDON: Thank you. ©

DR. JOSEPH: My third point, because IV drug users
concentrated in areas of poverty, AIDs is hitting the city's
minorities, both men and women, especially hard. Thirty percent
of our AIDS cases in New York City have been among blacks, 24
percent among Hispanics. Eighty-six percent of male IV drug
users with AIDS, 91 percent of mothers of children with AIDS have
been black or Hispanic. Of the men with AIDS in New York City
whose risk is sex with other men, 34 percent are black or
Hispanic. Those previous three points converge in my fourth
point. The key to the future of the epidemic is the IV drug
connection. Future projections turn on the spread of infection
among those most at risk through IV drug abuse; addicts
themselves, women who are their sex partners, children and all
those folded into a perspective of minority group status.

Breaking the link between those groups and IV drug use
is the most critical imperative. Although AIDS in New York City
presents problems of a magnitude not seen elsewhere in the
country, AIDS elsewhere undoubtedly shows a somewhat different
face, particularly in less densely populated areas of current low
prevalence. Every city and state across the country will
struggle with issues we've already faced for several years as
AIDS spreads across the country from high prevalence areas.

What, therefore, must a national prevention and
treatment program consist of? Simply put, we need resources for
understanding the causes and patterns of infection, controlling
the spread of HIV and providing those sick from the disease with
appropriate levels of medical and mental health care and social
services support. On any day this month in New York City, over
1,300 people with AIDS or AIDS-related illness occupy New York
City hospital beds, including about 450 of them in municipal
hospitals. The number of people hospitalized in the city is
expected to reach 2,500 beds per day by 1991.

Growing numbers of infected asymptomatic people will
need emotional support, clinical monitoring and possibly
hopefully preventive treatment as new therapies become available.

191

 

 

 
 

 

People will undergo enormous stress as thousands of AIDS deaths
wreak havoc on families and communities. The city will need to
provide home care services for three or four times the 216
persons who receive such services in 1987; for three times the
number of homeless persons with AIDS as were served in 1987;
we'll have a need for housing-related services such as supported
living, scattered site housing, SRO hotel housing and rent
support.

In the unlikely event of a development of a vaccine or
effective treatment in the near future, resource demands will
still intensify, of course, as you know, as the numbers of people
with AIDS grow. In New York City in Fiscal '88, spending for all
AIDS programs will be over $389 million, of which approximately
$100 million is city funds. This is up from $278 million, $74
million of which is city tax levy, in just the last Fiscal Year
1987. We've been fortunate in the city to have the vigorous
support of Mayor Koch for our program initiatives, but analogous
leadership has been sadly lacking at the federal level, where
much of the future funding must originate if New York is to build
effective prevention programs and meet service needs.

A significant federal commitment is too long overdue.
Your recommendations for an effective comprehensive AIDS program
must include the following. First, we need nothing less than a
bold and effective national prevention strategy. It must be
directed at slowing the spread of the virus among heterosexuals
as well as reducing the toll among homosexual men and IV drug
users and their sex partners and children. Implementing a
national prevention strategy is already within our reach. It
includes only three major elements: One, a massive national
public health education program which would consist of an outer
shell of explicit information to the general public and an inner
shell of targeted education for people practicing high risk
behavior. Explicit culturally relevant education is the most
effective way to reduce AIDS transmission. We must not be
deluded into believing that simply moralizing about values can
replace a frank approach to the uncomfortable issue surrounding
AIDS such as the urgent need for age appropriate and effective
sex and drug education in schools.

The second element of a national prevention strategy is
the rapid expansion of voluntary, confidential risk reduction
counseling and HIV antibody testing into every public and private
clinical facility, including physicians' offices, outpatient
departments, sexually transmitted disease clinics, family
planning and abortion clinics and anonymous test sites. I
believe these are now proven to help increase people's awareness
of their risk and encourage them to change their behavior.

Finally and perhaps most importantly, we must have
major efforts to curtail AIDS transmission via the IV drug user.

192

 

 
 

 

The national war on drugs is so far an abysmal failure. There
will be no slowing the spread of the AIDS virus or preventing its
seepage among heterosexuals without a meaningful war on drugs.
Efforts must range from interdiction at the international level
to law enforcement at all levels to more education programs,
increased and liberalized methadone maintenance, and rapid and
massive detoxification programs, plus, in my view, the
availability of clean needle exchange. Mayor Koch will speak
about these issues in greater detail when he testifies before the
Commission next week. Without a vigorous comprehensive national
prevention program, in my view, within the next 18 to 24 months,
we will fall behind in the epidemic among heterosexuals as we
have fallen behind with gay men and IV drug users. Seizing the
opportunity now would save large numbers of lives and
substantially reduce the enormous burdens and costs of the
epidemic well into the next decade. We need increased resources
for clinical care and research that do not come from other
projects. The absurd example is the administration's proposal to
transfer funds from sexually transmitted disease projects into
AIDS projects.

Thirdly, and this is very important, we need federal
legislation to protect against unauthorized or inappropriate
disclosures of confidential health records. This confidentiality
protection must also protect the physician's ethical and
professional duty to warn a patient's known contacts of the
substantial risk of infection when the patient refuses to do so
without deterring counseling and testing for those who seek these
services. Both the Medicare individual reimbursement cap for
home care and the 80/20 rule for hospice care are inadequate to
meet the needs of AIDS patients and do need to be changed through
legislation.

Finally, a national AIDS program demands a moonshot
approach to federal resource commitment guided by aggressive,
articulate and visible leadership from the highest levels of
federal government. Except for Surgeon General Koop, that
leadership has been lacking. The result has crippled control of
AIDS in this country. The federal administration has
continually abrogated its responsibility to take effective AIDs
prevention action. What it has done is to either run counter to
the best advice of the nation's public health authorities or has
been divisive and diversionary, the sum and substance of the
efforts only skirmishes that divert our attention from where the
real battle against AIDS must be fought.

Congressional adoption of three shortsighted
counterproductive amendments to federal legislation proposed by
Senator Jesse Helms exemplifies such wrong directiveness. I
provided for you a document that the mayor and I provided to all
the Senators yesterday commenting on 14 further amendments that
the Senator is about to offer to the continuing resolution.

193

 

 
 

 

i.

¢
" het ' | 7
wt gf u i aan

,
{ r

Senator Helms' amendment to the Health and Human
Services appropriation bill prohibiting the use of CDC funds for
"education, information or prevention materials and activities
that promote or encourage, directly or indirectly, homosexual
sexual activities" was an ill-conceived and deplorable example of
exactly what we do not need in this time of plague.

The Senate, unfortunately, overwhelmingly adopted this
amendment. Though the House directed its members in the
House-Senate conference to accept the Helms provisions, final
conference agreement is pending, and we hope that will go the
other way. The battle to slow the transmission of AIDS
infection will be won only through a frank and unified approach
to the uncomfortable issues surrounding the disease, not through
confusing vigorous, aggressive leadership and programming with a
coercive or punitive stance. Such quick fix solutions as Senator
Helms' amendments would neither efficiently use resources nor
involve people in our prevention programs.

Clearly, AIDS is forcing us to choose who and what we
wish to be as a society; whether we opt for the mean spiritedness
that burns down our own houses or the resolute spirit that builds
them up in times of adversity. Let me close my remarks with this
final observation. In my conversations with public health
officials from other countries, I have been struck by the almost
uniform perception among them that the United States has done
virtually nothing to institute a national effort against AIDS.

Despite protestations to the contrary and despite the
best efforts of the CDC and the NIH in their respective fields,
the truth is that there is no national AIDS effort. From the
administration, we have gotten only waffling, wavering and
failure of nerve. What we desperately need today are leadership,
vision and courage. Your Commission must seize the opportunity
to incorporate these into a national AIDS program that saves
lives and relieves suffering. Our time to do this is running
out. I'd be happy to answer your questions.

MR. CREEDON: Thank you, Dr. Joseph. Dr. Lee.

DR. LEE: I defer.

MR. CREEDON: Penny.

MS. PULLEN: No.

MR. CREEDON: Ms. Gebbie.

MS. GEBBIE: I will give you a couple of questions ana
then ask you to lump your answer together. You indicated you

have some current data on your seroprevalence surveys in New

194

 

 
 

 

York. I think we would appreciate hearing at least a brief
summary of those and perhaps seeing them in writing at some later
point.

Second, related, a major thrust of these two days is to
identify which numbers about current infection in the United
States provide the best base for our calculations, predictions,
and policies. Given what you know about New York City, how
comfortable are you with this Commission using something like
CDC's calculations? If not those, where do we look? What
magnitudes of error do they include? Is the family of surveys
they are proposing the right direction to round out those
numbers? If not, what different things should we do?

DR. JOSEPH: Let me give you some current
seroprevalence material that we have. Many of these are
preliminary, but I think we're fairly confident of the direction
they point. First, we're about finished with the major
seroprevalence survey in reproductive age women that you heard is
beginning in other places. The numbers that come in from, I
believe, its 48 sites we're sampling around the city, hospitals
and outpatient departments, the percentage of reproductive age
women infected in New York City is in the range that Dr.
Landesman mentioned. We believe it will turn out on final count
in New York City somewhere between 1.5 and 2 percent.

I would say as an example of why orders of magnitude
are not at issue, our 400,000 that you questioned about earlier
was based on an estimate two years ago that 3 percent of women
would be found positive. . Now we're finding somewhere between 1.5
and 2 percent currently positive. That may change that order of
magnitude -- that may change that estimate of people infected
from 400,000 down to 350,000 or 450,000 down to 400,000, but it
really makes no difference. So we're looking currently in the
city at about 1.5 percent of all reproductive age females are
infected. We have a most disturbing study that has looked at the
prevalence of HIV seropositivity in patients coming into one of
our sexually transmitted disease clinics. Again, you can quarrel
about the methodology of the selectivity of the population, but
we found that in persons who are visiting this clinic, this is a
minority poverty community, the rate of current positivity in
people coming to the sexually transmitted disease clinic is 7
percent, 8 percent males and 6 percent females.

If you look only at those people who have genital ulcer
disease -- and I believe, Dr. Walsh, that that's the key of the
dilemma of the African pattern and the North American pattern.

If you look at people who come in with genital ulcer disease to
that STD clinic, 20 percent of them are currently seropositive in
the City of New York, 21 percent of males and 19 percent of
females. We began just about two months ago doing HIV antibody
testing in our prisons in New York City,in New York City jails.

195

 

 
 

We do that on a voluntary, medically confidential basis as an
adjunct of counseling. We have only very small numbers so far,
but of the very small numbers that we have so far of people who
have volunteered for and gone through that testing, half have
been seropositive.

I think if you look at those numbers and the
populations that we know in the city of people who practice high
risk behavior, we can be very, very confident that the order of
magnitude of numbers that we are using, which are really right in
tandem with the CDC numbers though developed by different
methods, are an accurate order of magnitude estimation. I see no
reason for this Commission to doubt that range that has been
given by the CDC or the projections that follow from it.

Let me say one last thing about that. Remember that
when we're counting cases of AIDS, we're looking at a snapshot in
time of something that happened five years ago. As I'm sure most
witnesses before the Commission have stressed, what we
desperately need is a snapshot in current time that can only come
from the kind of seroprevalence infection studies that I've just
cited preliminary results fron.

My answer to your question, Admiral Watkins, is that
one of the most important things that could be done by the cpc
now is to design and’ put in place a series of what they call
sentinel surveys. I think the 30 metropolitan areas are both a
vision of current time and a way to have sentinel outposts to
tell us what is happening before it gets actually down upon us.

I think that in a wide variety of demographic and
economic and racial and ethnic and geographic settings, having
sentinel outpost surveys is going to be very important, because
there are twists and turns of the epidemic that we really cannot
predict. Behind the confidence of the numbers that Sheldon
Landesman and I give you this morning lies some uncertainty about
that fourth overlapping phase and certainly about the fifth phase
of the epidemic. I think if we have some pickets out to let us
know what's happening as early as possible, that will be more
important than arguing currently about the exact details of the
numbers themselves. I think that answers your question.

MS. GEBBIE: Thank you very much,

MR. CREEDON: Dr. SerVaas.

DR. SERVAAS: I wonder. You mentioned your school,
teaching AIDS in school. Would you be able to send to the
Commission the materials? You must have done a lot of work and

are way ahead of the rest of us on how to teach AIDS in school.
Do you have that?

196

 
 

 

DR. JOSEPH: I hope, as Dr. Axelrod mentioned
yesterday, the New York State Board of Regents mandated a
curriculum. We have a curriculum in the schools in New York
City. I'd be happy to send you all the material we have, which
is a videotape as well as a printed curriculum. I would
underscore the point, I think the bottom line is that this has
got to be explicit, direct, frank discussion at an age
appropriate level, and what is appropriate in the first grade is
not what's appropriate in the fifth grade. My answer to Dr.
Walsh's previous question around STDs in Scandinavia, my view is
because they do have frank, explicit repeated sex and family
life education and access to comprehensive health services,
including family planning and contraceptive services for
teenagers and education beginning in the earliest grades of
school. That is probably the single most important factor in
their STD epidemiology and their teenage pregnancy epidemiology
as well.

DR. WALSH: They've had it for years.
DR. JOSEPH: Yes.

DR. SERVAAS: My other question, Dr. Joseph, is when
you tested in your sexually transmitted disease clinics, how
successful were you in counseling these women? Are they told
then that they are positive? Are you able to keep them from
getting pregnant?

DR. JOSEPH: These are men and women. Our position in
New York is that it is critical to maintain the voluntary nature
of testing. It's critical to maintain this information medically
confidential; that the general statutes of medical
confidentiality do not suffice in that area, and that's why I
think in addition to the legislation we hope to put through the
state this year, we need federal model statutes.

I think there's a direct analogy to the 1964 Civil
Rights Act and its impact on individual legislation in the states
and our need for confidentiality protection and the duty to warn
protections at the federal level that can be modeled to the
states. So that in our anonymous test sites, we have four now --
it will be five this year, city funded, four state funded -- in
our sexually transmitted disease clinics, in the hospitals and
physicians' offices where testing is growing but still not
aggressively pursued enough, the bottom line principles are
voluntariness in testing, confidentiality of that information,
and testing not seen as an end in itself. Testing is no end in
itself. Testing must been as an adjunct to a counseling and
educational process that motivates behavior.

197

 
 

 

wo

" ia!
t

1

-. DR. SERVAAS: But my question was how do you counsel
them after you've tested them in this sexual disease clinic? Do
‘you have sufficient physicians to do this?

DR. JOSEPH: Well, no public official will ever come
anywhere and sit down and say we have sufficient physicians or
resources. We're all desperately behind in this epidemic. I
gave you the figures of what we're spending in New York City,
$389 million in this fiscal year and $100 million of that city
tax levy. Our requests for resources both on preventive and the
curative and social services side is going to go up enormously in
the next year. The city has given us what we asked for, but it
presents an enormous strain on the city. As perhaps a more
direct answer to your question, I think the single greatest
constraint currently -~- well, there are two great constraints to
the rapid spread of voluntary confidential testing, not only in
New York City but elsewhere.

One is the confidentiality clarification and
protections that I mentioned. The other is the availability of
trained and competent counselors to make the counseling and
testing program work. This is particularly acute in the
voluntary sector. I think we don't have enough; we need to do
more. We are doing more. But I think in the public sectors in
many of the high prevalence areas, we've gotten off and running
with counseling and testing. But the voluntaries lag far
behind. While I think that the counseling process has to be a
routine part of all clinical encounters, I believe it's true
that in the hospitals and in the clinics you're going to have to
have the availability of trained counselors both to oversee and
to train others and to be a technical backup for many other
people doing counseling. That just does not exist in this
country. If it doesn't exist in New York City, you can imagine
what the state availability of trained counselors is elsewhere in
the country. That's a second critical constraint to the rapid
expansion of testing.

MR. CREEDON: Dr. Walsh.

DR. WALSH: Steve, I wanted first to comment on the
vaginal ulcer business in Africa. You are aware that Ken Warren
and his colleagues at the Rockefeller Foundation are now mounting
a major study to verify that as the possible cause. I don't know
whether the Commission is aware of that, but they are doing that
to see if that really has a role.

DR. JOSEPH: Let me just break in on that for a moment
to say that, you know, these things are all tied together. We
have an epidemic in New York City and elsewhere in the country
of a rising incidence of syphilis, which is a genital ulcer
disease. We have tremendous problems with chancroid in New York
City and elsewhere in the country, which is a genital ulcer

‘ 198

 

 
 

 

 

disease. We've all forgotten about herpes since AIDS, but there
is plenty of genital ulcer --

DR. WALSH: That's why I was going to ask why he's
doing it in Africa. Why doesn't he do it in New York?

DR. JOSEPH: We will compare our results --

DR. WALSH: I don't know if the study has started. You
might ask Ken. You know him, I'm sure. The question I wanted to
get clear is in the confidentiality area, you made a statement
that this must also protect the physician's ethical and
professional duty to warn a patient's known contacts.

DR. JOSEPH: That is correct.

DR. WALSH: In other words, I think I understand
clearly that you feel that the physician, even without permission
from the patient, should be permitted to do some contact tracing.
Is that correct?

DR. JOSEPH: I will explain that and then explain to
you what we do do.

DR. WALSH: Yes, because I know that's a controversial
area.

DR. JOSEPH: I believe that the physician or the public
health authority when acting as a physician, for example, in our
STD clinics, does have an ethical and professional duty to warn.
Now, the first step in that process is --

DR. PRIMM: What did you say, Doctor?

DR. JOSEPH: To warn the sex and drug partners at
significant risk of a person who's infected with HIV. Now, the
first step in that process has got to be to urge the index
individual, the individual who's before that physician, to
fulfill their responsibility, which is the real responsibility,
to perform that duty and to warn their sex and drug partners.
What we do in New York is if that individual says they want help
in the process, we say to them, fine. Bring your partner, and
we'll sit down with you and help you in that process.

If the individual says that they recognize that
responsibility but they cannot do it themselves for whatever
reason ~-- they're embarrassed, they're afraid or whatever -- we
will perform that function for the individual. We will go out
and do what I call contact notification. I make that semantic
difference between contact notification and contract tracing to
underline the very important point that this is a voluntary
consensual process. You cannot drag the names of sex or drug

199

 

 
 

CO
*
‘
- : 2
'

contacts out of people in a coercive way,, but. if the person,
recognizing their responsibility to have ‘their contacts informed,
wants the public health authority to do that, I think we have an
obligation to do it, as I think the individual physician --

DR. WALSH: Steve, is your concept -- because this has
bearing on a position or recommendation we may have to make on
existing legislation on the Hill, is your concept, however, that
if a physician happens to know the contact, say the individual is
married and the patient still does not give permission, is it
your concept that the public health requirement is such that that
physician has an obligation to make the contact anyway?

DR. JOSEPH: I believe that if the physician cannot be
assured that that information is transmitted to someone in
substantial risk, that physician has an obligation to inform that
person.

DR. WALSH: Will that require legislation in New York?

DR. JOSEPH: No, it will not require legislation. What
we need is actually the other side of it in New York in a bill
that we hope to get up there in the next week or so. What we
need is a double-barrelled bill that protects confidentiality on
the one hand and on the other hand provides protection to the
physician and others in those circumstances where there is an
appropriate need to disclose or warn.

DR. WALSH: Would you and the mayor, for example, be
willing to advocate that at the national level? Because some of
the existing legislation on the Hill fines the physician if he
does it without permission, even if it's a wife, and also does
worse than that if he does it a second time. To me, it didn't
seem like a realistic thing for legislation, but would you all be
willing to, with the experience you've had in New York, to
suggest modification?

DR. JOSEPH: We will do that and will advocate on the
legislation, and I will be happy to furnish to the Commission as
soon as our bill is agreed upon and ironed out within the city
administration, which should be in the next week or so, to
furnish you a copy of the bill so that you can see in detail what
we're talking about.

 

MR. CREEDON: Dr. Primn.

DR. WALSH: Steve, one other question. I know with
your longstanding history and experience in the breast milk
controversy years ago, has New York done any studies on
transmission by breast milk?

DR. JOSEPH: No --

200

 

 

 
 

 

DR. WALSH: Because I'm not satisfied with what we're
seeing internationally.

DR. JOSEPH: No. To the best of my knowledge, there
are no U.S.-based studies on the issue. At the Surgeon General's
conference on AIDS in families in Philadelphia last spring --
that monograph is out -- there was a recommendation in terms of
women at risk or women who think they are at risk being tested
voluntarily again, and a seropositive woman not breastfeeding.
Again, there's a real difficulty in applying that recommendation
to the third world setting where breastfeeding is really a life
and death issue in many circumstances.

MR. CREEDON: Dr. Primm.

DR. PRIMM: Dr. Joseph, you spoke about explicit
culturally relevant educational material, and I need not tell you
about the controversy that surrounds the gay men's health crisis
and the explicit material that they had. So I'd like you to talk
to us a bit about what you feel is explicit. That's question
one.

The second area of concern to me, and you and I have
discussed this previously, is the increased and liberalized
methadone maintenance treatment in New York and how do you see
that? What constitutes a more liberalized methadone maintenance
treatment program in New York and the ability of cities other
than New York, for example, to set up such treatment programs
with the controversy that you know surrounds methadone
maintenance as a modality of treatment for intravenous drug use?

Then my last question would be explain what you really
mean by your availability of clean needle exchange program and,
if you would, include other narcotic implements that are used by
intravenous drug users and talk about them in relationship to the
transmission of the virus.

DR. JOSEPH: On your first question about explicit
material, AIDS is about sex and about drugs, and there's nothing
you can say about sex or drugs that will not be conflictual or
offensive to someone. So you have to start from that basis.

The second point is particularly in the world we live
in, particularly in the urban areas where people are bombarded
with media, particularly electronic media, information and
exhortations to this or that, our sensibilities and sensitivities
are dulled. The messages do not get through. We have felt that
in view of those two things that all our modalities, whether it's
our videotape mass media commercials -- I'd be happy to furnish
you with those -- or our street level materials or our packaging
and inserts into the million condoms we distributed last year in

201

 

 
 

 

the Department of Health, from the Department of Health, have got
to be directly focused on the issues with language that is
understandable to the people that will use that information.
There's no easy answer to that, and I'm thinking of two
incidents. Let me describe two incidents that will give you a
sense. Early on, about a year and a half ago almost, we had a
brochure ready to go out that talked about protection of women
from heterosexually transmitted HIV infection. We had a long
debate over whether in the brochure that we were going to produce
with the New York City logo on it we should use the word "semen"
or "cum". There were some who would object to the use of either
word, but how could we possibly convey what we were trying to
convey without that kind of explicit wording? There was a debate
over whether street vernacular or common usage or medical
terminology could be effective and understandable and
uncontroversial enough to allow us to do it at the same time.

The second anecdote, when we produced our condom
packaging for distribution of condoms, the issue came up --
again, this is something that Koop, I think, has spoken very well
about -- about people not knowing how to use condoms. Most
people, including many people who use condoms, don't know how to
use condoms in the sense that would protect against disease
transmission. So we decided that we needed to have some
instructions about condom use in the package. We then took note
of the fact that many of the people to whom this material would
be aimed do not have a great command of reading skills. So the
issue was how can you package instructions in a restricted space.
We decided to use pictorial display.

We then got into the issue, well how graphic should our
pictures be? Should we have a photograph of someone rolling a
condom down over a penis? Should we have a line drawing of that?
Should we have something that is not quite so graphic or less
graphic or whatever? Those issues become terribly important.
Again, it goes back to something that Sheldon talked about, and
there's no easy answer to them. You can't make general rules
that apply except, I believe, the general rule that in each
product the audience that it is aimed at, the product has to be
understandable and credible to that audience. So in New York
City, we have to have our messages in Spanish as well as English.
We have to have our messages at a reading level that's
understandable to the largest number of people in New York City,
and we have to use words and phrases and pictorial descriptions
of things that people understand and can relate to.

On your second issue, in the issue of reduction of IV
transmission, as with everything else with AIDS there is no
single channel, single bullet solution to the problem. In
everything we do about AIDS, whether in the prevention or the
clinical side, you have got to make use of the widest range of
imperfect solutions that you can apply. On the drug use side, we

202

 

 
 

 

have got to do everything beginning with exhorting people not to
shoot up, to telling people if they must shoot up use a clean set
of works, to telling people if they are going to use a set of
works that may not be clean here is how you clean it, and that
has to be graphic and descriptive and in the vernacular, in ny
view, in order to see if the data will prove out that the
availability of clean works will reduce transmission.

There's no data on that from the United States.
There's a lot of opinion and prejudice both ways, but there's no
data from the United States as to whether the availability of
sterile needles and syringes without prescription would have a
positive effect on transmission of many diseases. New York State
is one of only 12 states in the United States that prohibit
possession of equipment, injection equipment, without
prescription. I'11 come back to that in a moment. You need to
then move over to all the educational issues around drug abuse,
and then you need to move into as wide a mode of treatment as we
can possibly move. Now, that has to include drugfree. It also
has to include methadone maintenance because there is no way that
we could develop in the time that we have left the number of
drugfree slots -- that's a terrible word; it's the one we use in
the business -- the number of drugfree spaces for patients that
would be required.

We have, as you know very well, somewhere between
200,000 and 250,000 people in New York City who use IV drugs.
Those range all the way from the shooting gallery person to the
weekend chipper. We have 35,000 of those 200,000 people in
current treatment. If you just do the numbers to get ahead of
this virus, we have got to get somewhere on the order of
magnitude of 25,000 to 50,000 people into drug treatment in the
next two years. The only way you can do that, in my view, is
through a rapid expansion of methadone maintenance treatment
programs. That has to be done, on the one hand, by developing
new programs rapidly, and I hope we'll have something to say
about that in the next week or so, and it also has to be done by
liberalizing some of the restrictions and requirements around a
staff to client ratio, around many of those things that constrain
the numbers of people in treatment.

Does that entail risks? Yes, it does. Is this a
matter of balancing imperfect alternatives as it is in everything
in the epidemic? Of course it does. Is methadone an
unvarnished good? Of course it is not. But is it one of our
only effective weapons for beginning to rapidly slow
transmission among IV drug users and their sex partners? I
believe it is, and so we have to shave away many things that in
other times, in normal times, we would be reluctant to shave
away.

203

 

 
 

DR. PRIMM: My concern about that recommendation is
that it comes along with a warning of prudence and caution,
particularly when we're talking about deliberalization of some of
the regulations and methadone maintenance treatment programs.

I see some of that happening now in New York with
liberalization, and when you talk about testing for the virus and
counseling, it's just an adjunct to counseling, well certainly
methadone is just an adjunct to counseling. Methadone is not the
be-all and end-all. It's an adjunct to counseling, very frankly.

When we begin to liberalize the numbers on the other
side, we defeat the program and make it less effective and
particularly with certain populations. You know where I'm
talking about, in Harlem and in Brooklyn where I run drug
treatment programs. We need a lower patient/counselor ratio; in
other words, something around 25 to 1 rather than 50 to 1 that is
recommended now.

I think if we get into that ballgame, we're going to
see a deterioration of methadone maintenance treatment programs.
For certain individuals who are motivated, et cetera, who have
jobs and whatnot, yes. They can have a high patient/counselor
ratio, but others, they cannot. That's the point I need to make.
I'm very in favor of your recommendation, but with it other kinds
of recommendations.

The one, Steve, on needle exchange, I had hoped you
would talk about other narcotic implements other than just the
needle and the syringe, because I feel very strongly that the
cooker and, indeed, the little piece of cotton that is often used
over and over again is just as bad as the syringe itself. You
and I know that this is common in New York, and it's a whole
ritualistic kind of process. Addicts do things differently
geographically, but generally pretty much so they use the cooker
and they use a little piece of cotton, et cetera, and they share
it. That, too, also means sharing the virus.

MR. CREEDON: We are beginning to run into some timing
problems and specifically Dr. Landesman has to leave by 11:30.
Dr. Landesman, would you come up to the table? I know Mr. DeVos
wants to make a comment and we will get back to you, if we may.

DR. PRIMM: Mr. Chairman, I just think both the
testimonies are so very important to the Commission. I am
willing to forego my lunch. I think it is so important.

MR. CREEDON: The problem is that Dr. Landesman has to
leave by 11:30. He has an appointment on the Hill at 11:55. I
would like to give the Commission an opportunity to question him
further before he leaves, and then we will come back to Dr.
Joseph and Dr. Wiley.

204

 

 
 

 

MR. DeVOS: My comment to Dr. Joseph is one of respect
for the work you have done in your report. I admire you and I
can understand I think your frustration if the rest of the nation
and the Senate doesn't come along with you as fast as you would
like to have them. I think, however, to include in this report
those attacks is a little less dignified than I like to see. I
just want to go on the record of saying hey, we aren't just
automatically accepting that because we accept all the rest of
your report which is so well done. I don't want to get into a
debate on that. I just wanted to make the comment.

MR. CREEDON: Dr. Crenshaw, do you have a question for
Dr. Landesman?

DR. CRENSHAW: Yes. I am curious to know, on your
condom packaging, if you dealt with any qualifications on
efficacy or --

DR. JOSEPH: No; we did not.

DR. CRENSHAW: Nothing like the Surgeon General's
warning?

DR. JOSEPH: No, we did not. In a way, one of the
comments I would have made to Dr. Primm is in AIDS, we cannot
afford to let the best be the enemy of the good. We can't afford
because we lack 100 percent in the up side of methadone
regulations or in the inadequacies of total protection with
condom use, we cannot let that paralyze us from using what we
have the best way we can use it.

DR. CRENSHAW: I was thinking more of cautions that I
think aren't optional, such as their being recommended with
spermicides, oil based spermicides can dissolve latex. The
comments, do not use vaseline or saliva for lubricant. Advice of
this sort which I think is critical to the effort to use them, to
delay infection.

DR. JOSEPH: All our condoms are latex.

MR. CREEDON: I would like to get questions for Dr.
Landesman if we may. One of the interesting things about this
Commission is that I keep learning new words. I had never heard
of a weekend chipper before. It is fascinating how different
terms come up.

DR. JOSEPH: How about a Wall Street noontime user?

MR. CREEDON: I'm getting educated. I have a few
questions, Dr. Landesman, which really I would like to address to
the group as a whole. Dr. Wiley, please come up as well. One

205

 

 
, : ‘
, ,

of the comments made yesterday and was made again this morning is
this question of the need for greater coordination between the
state and the local and the Federal Government, and how important
that is in terms of getting the job done on a national basis.

Dr. Joseph, I think you have commented somewhat on that and Mr.
DeVos has commented on your comment.

DR. JOSEPH: And I won't comment, I promise.

MR. CREEDON: I would appreciate, Dr. Landesman,
getting your view about that subject and Dr. Wiley and Dr. Joseph
as well. We are talking in part here about certain high risk
groups, the homosexual community and the IV drug user comnunity.
I would like to understand whether it is your opinion that those
groups are growing or diminishing or stable. I guess to some
extent, that involves a question, especially with the IV drug
users, how many new people are coming into that group and how do
we stop that from happening, if we can.

The third question really related in part to expressing
my surprise, I guess, at the 400,000 number that you mentioned,
Dr. Joseph, as the incidence. One of our witnesses yesterday was
Dr. Redfield from Reed Hospital and the Defense Department has
done quite a few studies, both with people applying for positions
in the military and also with respect to people who are in the
military. The incidence of the virus among that group suggests a
relatively lower level of prevalence. Let me just explain what I
mean. Some people have said, that group is not a representative
group and therefore, the incidence, Dr. Curran said, has got to
be at least two or three times higher than the 300,000 to 400,000
that study would suggest. At least that is my impression.

I guess in looking at that group and trying to decide
whether that group is a representative group, I think Dr. Primm
would certainly agree it is not a representative group because
first of all, there is a requirement of a high school education;
secondly, it is probably a group that is different than the IV
drug user group, where the prevalence is much higher. On the
other hand, it may not be a representative group on the other
side in that many people who would not apply to the military may
not have as high a prevalence rate. I would ask, Dr. Landesman,
before you go, if you would comment on those three questions, if
you can.

DR. LANDESMAN: I think the first one was the question
of the organization and integration of different services. Two
points. One is that I think the very nature of the bureaucratic
structure, not to use that as a pejorative term, just the way we
organize our services in different departments at the different
city, state and Federal level, to some extent, they are similar
in all ways; Departments of Health, Social Services, Family
Services, et cetera.

206

 

 

 

 
 

 

To some extent, conspirers against effective, organized
care of patients with HIV disease or preventive measures, simply
because this disease, as I noted earlier, cuts across those
bureaucratic lines, especially around the issue of AIDS and
families, because the natural structure of the bureaucratic
process works against us in trying to organize these things, it
then becomes imperative at each level to develop a highly
integrated leadership which can in fact cut across the normal
bureaucratic lines, such that coordinated action can be taken.
That comes again to the issue of leadership and willing to go
beyond what is the normal process of how you get things done,
developing mechanisms for housing for patients is very difficult.
Developing mechanisms for services for women, you have to go to
one department. For services for drug users, you have to go to
another department.

Somebody at the top of each one of these levels has to
say, listen, we are going to throw all the old rules away, we are
going to start again with one integrated program and we are going
to draw from the others. That really comes down to an issue of
leadership at the city, state, and Federal level, and somewhere
along the way, we are going to have to do that if we are going to
be able to effectively deal with the problem.

The point is the natural bureaucratic structures that
we have impede the development of effective integrated programs
and therefore, at all levels of government, we are going to have
to develop mechanisms to break down those natural bureaucratic
structures. Your second question was? :

MR. CREEDON: The size of the groups.

DR. LANDESMAN: The military data as I remember it, the
overall prevalence there was .15 percent overall, age adjusted,
-14 percent. I think that number and the numbers Dr. Joseph
quoted about New York City are entirely consistent with one
another. In New York City, I believe the overall rate from the
military data was 1.5 percent, in New York City as a whole, of
all those who volunteered for the military were found to be
positive. By and large, people who volunteer for the military --
let me back track. The military service selects against people
who are likely to be infected, such that persons, because being
gay is illegal in the military, as is drug use, and especially
gay men are reluctant to be identified as being positive while in
the military, obviously persons who are gay often times select
against going into the military for that reason, besides the fact
that the military says they should not be there.

The numbers that we get in the military are of
necessity a group that would represent a much smaller proportion
of those persons who are infected. The numbers that were quoted

207

 

 

 
 

 

from the City are not inconsistent with the military data, if we
have 1.5 to 2 percent of all those who enter the military and
that includes people who reportedly have no risk factor at all,
that's a very significant number. You must also remember that
the military data includes farm boys from Iowa who have never
been 50 miles beyond their home. Obviously, we have a very low
prevalence group volunteering for the military as well as a high
prevalence group from the inner city areas.

I don't think you can transpose the military data into
a national figure. I think one of the things that sometimes we
stumble on is the concept of one national figure of how many
people are infected. As mentioned earlier, the AIDS epidemic is
really a variety of sub-epidemics or multiple pockets of disease
in different areas of the country. No one sample in any one
place is truly representative of what is going on in another
place. It is only applicable generally to the population that is
similar in demographic characteristics to the population that is
tested.

I think we get into a lot of confusion by saying that a
study done on a certain population in one area of the country is
directly applicable to a different population in another area of
the country. You have to look at comparable types of populations
in comparable areas before you can get a sense as to what the
numbers are. You can't generalize the local epidemiological
studies throughout the entire population of the country.
Otherwise, we will be sitting here with conflicting data from now
until the year 2000.

THE COURT: What about whether the groups are growing
or not growing?

DR. LANDESMAN: As far as I know, homosexuality is
homosexuality and will probably not change now or in the
foreseeable future. The issue of whether homosexual persons will
behave in a manner that will slow the rate of transmission in
that group of persons is a separate question. All the evidence
we have indicates that may be the case. In terms of a national
prevention strategy around that issue, I think again as I said
earlier, focusing on the young or emerging homosexual, the one
entering that lifestyle in terms of risk reduction and prevention
of infection, is where I would place the dollars around the issue
of gay men. The population of 25 to 44 in gay men either are
infected or know how not to get infected but aren't going to
listen. In some ways, the question is resolved there one way or
another.

It is the young people who are coming up now that one
must really focus on so that you decrease the total pool of
infected persons. That's the goal, to decrease the total pool of
infected persons in that population. In terms of heroin users or

208

 

 
 

 

IV drug users, I'm not a drug expert and don't purport to be one.
My sense is that the total number of persons who are using
heroin, the rate of new people coming into that pool, has
decreased somewhat from the time in the 1960's and early 1970's
when the heroin epidemic really took off. The secondary issue
that is now coming up, and I think Dr. Joseph can speak to it
more, is the issue of cocaine use, which is often times
intermixed with heroin, which is the new thing that has to be
faced and which may be a significant add on problem to the
epidemic in the future.

MR. CREEDON: Are there any other questions of Dr.
Landesman?

DR. LEE: I have been doing some calculations on your
numbers of heterosexually infected people. You say 30,000 to
115,000 persons are infected with the virus, heterosexuals. To
try to get a handle on this for middle America here, suppose we
picked a figure of 70,000. Are 80 percent of those minorities?

DR. LANDESMAN: Yes, I would probably say 75 to 80. I
stress that is what we call in the trade a guesstimate. I would
think your number wouldn't be off by any more than ten percent.

DR. LEE: We are left with 14,000. Of the 14,000, how
many get into the drug area? 80 percent?

DR. LANDESMAN: Those who aren't minorities?

DR. LEE: Of the 14,000 heterosexually infected whites
left, are 80 percent of them drug related?

DR. LANDESMAN: I don't think we have a lot of data.
There is one study I'm aware of that has been done in Staten
Island in New York City of a white middle class population on
Staten Island, where they looked at cases of AIDS, cases of HIV
infection and their sexual partners. In that study, these women
who were all white and not minority populations, the average
income in the families was about $27,000; the average education
was high school plus one year. I don't have the exact figures in
my head, but I think about 30 percent of them were infected.
These were sex partners of middle class, working middle class
drug addicts. I have focused on the issue in the minority
communities. We should also realize that drug addiction exists
outside of those communities as well.

DR. LEE: What I am getting at here is as you look at

what percentage of prostitutes, what percentage of drug related,
you come down in the white, and a figure I have heard quoted

209

 

 
 

 

before, in the white, non-drug abusing female population in the
United States of America, the incidence may be as low as 1 in
100,000. Are you people at the table today completely
discounting Langmuir's stuff? Is the plateau a fiction?

DR. JOSEPH: Yes. Alex Langmuir, is a man who we have
great affection for as well as respect, anybody in the business
does. I have particular respect for him for a number of
reasons. I think he is off base on this. There is absolutely no
single indicator that I know of that would give any indication
that we are at or coming to a plateau that will tail off in the
next few years.

I think Alex, and absolutely no disrespect to him, he
is a giant in the field, Alex is looking at models that do not
apply to this epidemic. This is the first epidemic that we have
any knowledge of that depends exclusively on voluntary human
behavior to start it, to move it, and to stop it. The models
that he is projecting, I think, do not apply. In the last year,
1987, the numbers of cases in New York City increased by 30
percent. The numbers of deaths in New York City increased by 38
percent. That is a snapshot of five years ago. I watch the
hospital numbers each month in New York City, each month we move
up about 25 to 50 hospital beds a day. There is absolutely
nothing in any of the data that I see that would lead one -- all
of us want the good news. There is not any good news. We better
be aware of that.

MR. CREEDON: Dr. Landesman?

DR. LANDESMAN: I think it is important that you not
look at the numbers in the aggregate when you are deciding as to
whether the problem is getting better or worse. In some areas,
it may be getting better while in other areas, it may be getting
worse. In the gay community, as I have said, the issue is now
going to be the emerging adolescent and young gay person.
Presumably at this point in time, the rate of infection in that
population, the younger age gay men, is considerably lower than
it is in the ones 25 to 44. It may well be that the overall
reservoir of infection in that population will go down. That
should be good news but it shouldn't lead us to think that
therefore the epidemic is in a sense getting better because there
is another population which is being fed by the drug using,
current or former drug users, in terms of heterosexual spread,
which currently and in the foreseeable future will be getting
worse. There are different elements of the problem. It is
somewhat like silly putty. You push in one place and you pull
in another. Each one will demand different preventive measures.

I think our goal is to try and keep HIV disease out of
the emerging homosexual population on the one hand, while at the
same time recognizing that we do have an issue of this

210

 

 

 
 

 

heterosexual spread and try to work on that problem at the same

time. I would caution against the concept, the overall general

terminology of it is either getting worse or getting better. It
doesn't lead us to really concrete action.

MR. CREEDON: Thank you very much, Dr. Landesman. A
number of the Commissioners have additional questions which we
would like to be able to submit to you, if we may. I hope we
have prepared you for your testimony on the Hill now.

DR. LANDESMAN: Thank you...

MR. CREEDON: Can we come back to Dr. Primm's questions
and resume there, please.

DR. PRIMM: I just have one thing to say and that is we
see a lot of intravenous drug use among physicians. Maybe not as
much now as we did in the past. I'm sure they probably use clean
needles. Is there any anecdotal experience or any experience
anywhere that could give us some idea about that population that
could be extrapolated to this population we are talking about?

DR. JOSEPH: I know of none. Let me combine my answer
to your question with some of the questions just posed. There is
some data that in New York City the heroin population is aging.
The person who knows more about this than anybody in the world is
Don Des Jarlais of the State Substance Abuse Service. His
estimates are the average age of heroin addicts in New York is
now up in the 30's. We don't really know how many young people
are coming onstream. We know we have about 100,000 uninfected
drug users in the City, IV drug users in the City. Don's
figures, which again, there is no reason to doubt, show we are
currently converting those uninfected to infected at about 8
percent per year. That's the kind of figure that gives urgency
to doing absolutely everything we can to stem that channel.

I would caution you, however, from thinking that the
only problem of substance abuse that relates to HIV transmission
is needle transmission. In New York City, with the explosive
epidemic of crack, hundreds of thousands of people, largely young
people, in this viciously addictive, devastatingly destructive
wild fire type of substance abuse, it is very common for young
women in the crack houses to sell themselves for their next vial
of crack when they run out of five and ten dollar bills. The
people they are having sex with in the crack houses are people
who on the other days are shooting heroin or are infected
otherwise.

I just can't say enough times that the AIDS, the HIV infection
problem will not be stemmed until we deal with the primary
problem of substance abuse and that goes far beyond just the
issue, as important as it is, of people sticking needles into

211

 

 
 

 

their arm with the blood of other people in the syringes. I
think we have to do everything along the way. That is why I
think the needle exchange makes sense.

MR. CREEDON: If the Commissioners are agreeable, I
would like to now move onto Dr. Wiley, and then we can come back
and ask questions. I hope you will stay, Dr. Joseph. Dr.
Wiley?

PRESENTATION BY DR. JAMES WILEY

DR. WILEY: In the interest of time, I'm not going to
read what I distributed to you but, rathér, paraphrase it and
stop to allow you to ask some questions. My testimony really
deals with the question of in what sense are heterosexuals a risk
group, and to answer that I think that we can consider an
introduction of the virus into heterosexuais as having, perhaps,
three long-term -- one of three long-term consequences.

The first might be that the epidemic would remain what
I would call a passive one; that is to say, the source of
infection into heterosexuals consists of people who are infected
in other ways, through sharing needles, through transfusions of
contaminated blood products and from homosexual exposures.
That's the one case, and we seem to be in a phase where you could
argue that, at least at present, the epidemic is primarily a
passive one. A second outcome would be a passive epidemic plus a
certain number of cases of heterosexual transmission from people
who were themselves infected heterosexually. I think that's been
referred to as secondary or tertiary transmission here today.

The third and, really, more frightening prospect is the
idea of a self-sustaining heterosexual epidemic; that is, one
that does not require continued contributions from the primary
risk groups. So what I really want to talk about is what is the
evidence for or against this outcome in the long run? It's
clearly not one we're seeing immediately -- and can we trust
that evidence to give us a reliable guide to how to think about
this? It seems to me that the one kind of evidence is the
present distribution of cases. If you look at AIDS cases and
you remove those that are of foreign born and who presumably
acquire the infection elsewhere, then you find about 2 percent of
the U.S.-born cases are presently classified as instances of
heterosexual transmission. The bulk of these are classified as
people who acquired the infection from somebody in a primary risk
group.

There are troubling numbers of undetermined source that
are about the same in magnitude as those classified as
heterosexual, and we sure would like to know a bit more about
them. But in any case, the rather small percentage of total
cases sometimes encourages the thought that a self-sustaining

212

 

 
 

 

 

heterosexual epidemic is not likely either as a description of
current affairs or in the future. Another observation that
seems to encourage that same inference is the comparison between,
for example, the epidemic in gay and bisexual men in the United
States and the epidemic among African heterosexuals in certain
areas and what we think to be true about American heterosexuals.

In the former two cases, we believe now that a good bit
of the reason for those epidemics were very high rates of sexual
contact; that is, that those epidemics were, in fact, driven by
sexual contact. I know that that was certainly a very early
inference from looking at the epidemic among gay men. The early
inference from the African observations was more frightening
until it was realized that in most cases you could attribute the
African epidemic to a combination of very high rates of sexual
contact and certain co-factors, such as the ones that have been
mentioned here already, perhaps other sexually transmitted
diseases that help AIDS get established in a population. Through
those comparisons, sometimes people infer that heterosexuals
really aren't at much of a risk because, after all, American
heterosexuals aren't like American gay men or they aren't like
African heterosexuals.

A third piece of evidence really comes from some
studies that I've been involved with in Berkeley on transmission
efficiency of HIV from man to woman and, to some extent, from
woman to man. Those studies so far have suggested a rather low
rate per contact; that is, per sexual contact, in this case per
unprotected vaginal/penile intercourse between an infected man
and an initially uninfected woman, the risk seems to be fairly
low. That, too, the notion has been that the notion that that is
maybe 1 in 1,000 has encouraged people to think that, well,
that's not high enough to get a self-sustaining heterosexual
epidemic.

Then finally, some of the seroprevalence estimates that
have been talked about today among army recruits, marriage
license applicants, STD clinic attendees and so on, suggest that
at least in the nation as a whole seroprevalence outside the main
risk groups doesn't seem to be too high, and that's again another
indication that we are currently in more or less of a passive
epidemic and that we don't have a self-sustaining one going.

Now, what are the weaknesses of that argument or, another way of
saying this, could we be wrong? Can we really rule out the
possibility of a self-sustaining heterosexual epidemic? I think
at the present time, partly because of the complexity of the
disease itself -- it has, after all, over the last five years
surprised us occasionally and usually on the down side -- because
of that and because certain information is very hard to get, and
maybe we're a little bit late in getting it, I think the answer
is no, we can't rule out a self- sustaining heterosexual
epidemic.

213

 

 
 

 

mitt Ni
Teg ,

° a 1 ’ ay err ee

| To that extent, I think the three of us really are

saying the same thing. In the long run, it certainly could
happen. Now, let me go over what I think the weaknesses are,
because it does suggest a strategy for reducing some of the
uncertainty later on. First of all, the present distribution of
cases is not a very good index of what the number of infections
is. You will, by the end of these hearings, hear this three or
four or five or a dozen times and more expertly stated than I
can, especially when the modelers come this afternoon. In any
case, what one can say is that there's no simple rule of thumb
that allows you to convert cases into infections in any risk
group. What we know is that that number tends to change, and as
the epidemic rolls along in that group, if the group manages to
stop seroconverting, the ratio of infections to cases goes down.

MR. CREEDON: Cases of people with the virus?

DR. WILEY: Infections are people with the virus.
Cases are people with AIDS. So it's very hard to project. What
I'm saying is it's hard to project from AIDS cases to AIDS
infections, and our best data by far are on AIDS cases rather
than AIDS infection. What we do know is that in the early stages
of epidemic growth, the ratio of infections to cases can be in
the hundreds, 300, 400. Certainly it was among gay men in 1981,
‘82. It can be very high. Now, given the long latency period
for this disease, the bottom line is that a small number of cases
can hide a large number of infections, and that's the first
caveat that I wanted to make on the available evidence for
heterosexual spread. It could be concealed.

Secondly, the comparisons between the sexual behavior
of heterosexuals and those of gay men and African heterosexuals
do suggest that we won't have a really rapidly escalating
prevalence of infection in those groups. They don't suggest by
any means that the level of sexual contact among heterosexuals
that have, let's say, a partner or more, one or two partners at
least, is low enough to be below the threshold for epidemic level
transmission in the United States. We just don't know that. So
while it's true that you can take a little comfort from the
comparisons between the African situation and the gay situation
and American heterosexuals in terms of the possibilities for
explosive growth, there is nothing in that comparison so far that
suggests you can't have slow, steady growth to perhaps 2, 3, 4
percent in some areas without the contribution of the primary
risk groups; that is to say, without a real self-sustaining
growth, not one that requires direct contact with those groups
that we now regard as high risk.

Okay, a third point has to do with my own findings. I
find myself criticizing my own findings here. The 1 in 1,000
figure is an average. It has been widely interpreted as

214

 

 
 

 

indicating that it is alright; that is to say that it is so low
risk that a few casual encounters don't matter much. Well, the
problem with that argument is that the average conceals
variability in the ability to acquire the virus as well as to
transmit it. Now, I have to disagree with my colleagues here
about orders of magnitude and transmission efficiency. I think
there are orders of magnitude. I think some people are much more
efficiently spreading this disease than others; that there are
exactly -- there are people that we could call superspreaders,
and we found a few of them in our work. Now, these people like
the much-celebrated case of the Belgian man who came back from
Africa to reside in Belgium and infected a large number of his
female partners who were in a relatively small number of
contacts suggests that as well. The factors that generate
greater potential susceptibility or infectiousness are just how
beginning to get studied well enough to make these magnitude
estimates, but I would say from the data that I've seen from
Africa and that I've analyzed from Africa in comparison with what
you observe in the United States that you can get as much as a
ten-fold difference in increase in infectiousness or perhaps
susceptibility under any of the three circumstances.

One, if the index case, the person infected, is in late
stage disease; there's a great deal of interest now in testing
hypotheses that people become more infectious over time, that
they don't stay at one level of infectiousness; or at least they
might have an early period of infectiousness before the antibody
production and then late in the disease when antibody production
goes down for as yet unaccounted reasons, they become much more
infective. Another case would be what has been already cited.
Genital ulcers seem to both facilitate transmission,
susceptibility to infection and possibly also infectiousness.
Anal intercourse is another factor that could possibly produce a
ten-fold increase. So I think actually there is about an order
of magnitude difference between unprotected heterosexual contact
and unprotected receptive anal intercourse, for example. So I
think there is that difference, and it certainly did contribute
to the epidemic among gay men.

On the other hand, the question is not really what the
average infectivity is, but it is what are the numbers of highly
infectious people and are they likely to change over time as the
current cohort of infected people from all risk groups ages in
its disease. So that what we're not dealing with here is a
stationary state that we can count on. Will that happen, and
what are the sexual practices at least among the heterosexually
infected people who happen to be very highly infectious?

MR. CREEDON: What evidence is there, specific

evidence, about the virulence getting stronger as the length of
the disease --

215

 

 
 

DR. WILEY: Well, there are two sources. One's direct,
and the other's indirect. The direct evidence, I'm not as
familiar with it as the indirect. The direct would consist of
following culturing virus out of the peripheral blood and semen
of infected men over time and then watching how much virus you

get.
MR. CREEDON: Has that been done?

DR. WILEY: That has been done. There are certain
estimates of it, but as I understand, the laboratory procedures
for culturing semen, for example, are not altogether reliable.
But there is a pattern -- in the few studies that have been done,
of early shedding and late shedding just prior or after having
been diagnosed. Now, the indirect evidence comes from looking at
the rate of male/female transmission in AIDS where the male is an
AIDS case versus where the male is an asymptomatic but infected
person. Those are my studies, and they indicate that, indeed,
AIDS cases are more efficient transmitters than asymptomatic
people. I'm really citing a reanalysis of some data that Dr.
Fischel published, and it's one of the few studies which look at
the sexual behavior of people after they've been diagnosed with
AIDS and the seroconversions in their partners. It is
interesting to note that in her study, after diagnosis the
average number of sexual contacts per week was somewhere between
two and three, and not by far the majority of them were
unprotected. So people do continue to have sex certainly with
permanent partners after an AIDS diagnosis. We found some
similar findings among gay men in San Francisco, too.

DR. PRIMM: Dr. Wiley, have you done hematological
studies on those people who continue to have intercourse and have
shown antigen at the time?

DR. WILEY: No. I don't. Somebody might, but I don't.
So that's a third caveat, that the low apparent rate of
transmission from male to female and the even lower apparent rate
from female to male may hide very important variations in
infectiousness and susceptibility that we need to know about not
only in terms of the adequacy of our warnings to people but also
because the population dynamics of infection depends critically
on how that average is composed. Then finally, the
seroprevalence estimates that I talked about before, they really
are seroprevalence estimates based on quite unrepresentative
populations. For that reason, a number of us have worried that
they may be misleading us, so that we are not taking perhaps as
much comfort from the low rates as we would like. The conclusion
I draw from this is that we really don't know the risks we face
with respect to the possibility of a self-sustaining heterosexual
epidemic. Our best current knowledge doesn't allow us to reject
that possibility, and we would certainly like to be able to
reject it. That's why many of us have advocated more direct

216

 

 

 
 

 

measurements of seroprevalence in the population, some of which
have been talked about, the CDC sentinel studies, various
seroprevalence studies that are being proposed or started around
the country. I'm involved in two of them, basically heterosexual
populations in San Francisco.

We are finding, in fact, at least in San Francisco if
not in Peoria, that you can do that kind of study; that, indeed,
the feasibility question is not entirely settled, but it looks
good from our point of view. The early results look like we can
get sufficient cooperation; that we're not getting so much
self-selection, as you might imagine; that people who could be
seropositive are not avoiding our study like that plague; and
that we are able to get them to give blood in the home with those
who are trained to do interviewing. Now, that may not hold
outside of San Francisco, as my CDC colleagues here often remind
me, but it is possible to do such studies, and they should be
done in those areas. where we already have epidemic AIDS, and
they should be done among heterosexuals.

But I think that at long last we're starting to see it
get done or at least get planned. Those of us who wanted to do
this three years ago were laughed right out of the study
sessions, so to speak. Nobody wanted to talk to us about that.
They thought we simply wouldn't find any infection out there
among heterosexuals, so why do the study. I think it would have
been wise, looking backward anyway, for us to have set up
sentinel studies in those days even if we weren't going to find
anything, because once you've done one of these studies, you
realize it takes you a couple of years to get going. You can't
just throw money at it. You have to throw people at it, and they
need to be trained. :

So it takes a while to gear up to do these things, and
the things being planned now probably won't be implemented for
another year, which delays the time that we can pin down how much
infection has occurred another year. Now, you mentioned
something about what sort of data would you like to see. Well,
certainly with respect to case reporting, we would like to see
what can be done with the undetermined source cases that CDC is
reporting, and it may be that there isn't anything that can be
done; that is, they just simply don't have the data. But if any
more work will yield good results, it ought to be done.

One specific thing I think needs to be looked at very
carefully, and that's whether it's possible to actually count the
number of what I think Dr. Landesman called tertiary cases; that
is, one key index for the outcome that I've been discussing, a
self-sustaining epidemic, will be an increasing number of people
who acquired HIV infection heterosexually from people who were
themselves heterosexually infected. Now, there will be, I think,
some difficulties in classifying cases in this way, but it's of

217

 

 
 

a
\
,

really critical importance in case reporting to know whether that
number is increasing. The other thing is to look for such
individuals in the case tracing of heterosexual cases and in the
seroprevalence studies as far as they can be done in connection
with interviews to find out how many people who are HIV
seropositive in these surveys, in fact, have no other risk factor
except that they've had heterosexual contact with people who are
not likely to be drug users or male bisexuals.

MR. CREEDON: Are the studies that are now being done
trying to develop that information or not?

DR. WILEY: I think they are. I mean, certainly the
ones that we're fielding in San Francisco are directly designed
to do that.

MR. CREEDON: Questions of Dr. Wiley?

DR. LEE: Dr. Wiley, I tend to agree with you myself
and not with Dr. Landesman. I learned as a doctor there's an
enormous difference between a vagina and a rectum. For the sake
of the Commission members who are not doctors, I might point out
that the French take their medicine in the rectum because you
get such high blood levels so fast. You get better absorption in
the rectum than you practically do by swallowing it by mouth. So
anything that's injected into the rectum immediately gets into
the bloodstream. What's injected into the vagina is -- the whole
mechanism there is geared towards rejection. So what do you
think about this business of 1 to 100,000, first of all, 1 to
100,000 white nondrug abusing females being HIV positive, number
one?

Number two, I wonder from Dr. Joseph and from San
Francisco why there hasn't been more or why I don't see more
about discouraging prostitution when up to 75 percent of these
streetwalkers are positive?

 

DR. WILEY: I'm not sure to what extent the first thing
you mentioned, the question of anal intercourse versus vaginal
intercourse, is going to decide the question of how many people,
how many white American women are infected. I think those are
two different issues. I do agree, and at least my data
suggests, that the anal intercourse is a more efficient mode of
transmission, but I do want to caution against saying that,
because that is so, that heterosexuals don't have a problem. To
me, it just doesn't follow.

DR. LEE: But there is a magnitude of difference, a
big magnitude of difference.

DR. WILEY: But the other thing that made a difference
in gay men was also a very high rate of sexual contact in the

218

 

 
 

 

years before anybody knew about AIDS, before there was any real
warning. So it's some combination of transmission efficiency and
fairly high rates of contact that would produce that.

MR. CREEDON: But you did say the order of magnitude is
very different.

DR. WILEY: It is 1 in 100 versus 1 in 1,000. Now,
that is one order of magnitude difference mathematically.
Whether that difference will protect the heterosexual population
from eventual self-sustaining epidemic is a whole other question.
As I said, since infectiousness might be changing, maybe we're
looking at numbers later on that are more like the numbers that
gay men were facing in the '80s as far as transmission efficiency
is concerned.

DR. LEE: I would like Steve to address this, too.

DR. CRENSHAW: Before we leave this point, Burt, could
I make a comment on vaginas that I think is very interesting.

DR. LEE: Certainly. I didn't know where that voice
was coming from.

(Laughter. }

DR. CRENSHAW: I think there's a really big
misperception about vaginas in general and their absorbency.
There are studies on aspirin; there are studies on other drugs
indicating that vaginas absorb them into our general bloodstream
and our system as efficiently and even more efficiently than the
rectum, number one.

Number two, the artificial insemination cases that are
very well documented in Australia where four women, separate
women, inseminated with the same sperm in a clinical setting, one
exposure, no thrusting, no blood drawing, all became infected
with the AIDS virus. So even though when you calculate
probabilities as you have, you can say 1 in 1,000, you can do
that for pregnancy, too, if you take into consideration
menopausal women and infertile women and come up with a similar
number. But when we get right down to it in and in view of the
macrophages in the vagina and the macrophages in the rectun, I
think we're going to find out that this is, as Dr. Landesman
said, ultimately splitting hairs. Even you're saying the order
of magnitude, if this is true, may not have a meaningful bearing
on modes of transmission and efficiency.

DR. WILEY: Certainly it has something to do with it,
but we're not dealing in different worlds here. The other thing
that we know from some of the recent studies of prostitutes in
Nairobi are that oral contraception seems to increase the

219

 
 

susceptibility to infection by at least an order of magnitude,
maybe as much as 20 or 30 times. So with certain co-factors, you
can produce a vagina that has the same absorption capacity as an
anus does.

DR. CRENSHAW: The problem with that study is that they
are not comparing it to what effect that oral contraceptives
might have on a rectum, because you can't compare the apples and
the oranges of the male, so it may be -- and this is the point
I'm trying to make -- it is not determined. It is not
established, and it may be old news that there is a discrepancy
between anal intercourse and vaginal intercourse and efficiency.
That's an open question that you need to look at.

DR. LEE: Could you also address the bisexuality issue.

DR. JOSEPH: Yes. I was going to that. I think there
are two other factors besides your anatomical illustration that
are important here. One, we know virtually nothing about
bisexuality in the sense of how many bisexual men, who are
infected, have sex with women in terms of the potential of that
transmission. I think the further we go in the thing -- you may
have some cross-country data on this -- that the more we are
finding out that another important channel is the infection of
the heterosexual woman by a bisexual man, who does not identify
himself certainly overtly and maybe even consciously as a gay or
bisexual man.

And the second thing, which may be even more important
than that, is that we know very little about the frequency of
anal intercourse among heterosexuals, and we should not make the
assumption that receptive anal intercourse is a risk factor only
for men infected by other men, but it clearly is a risk factor as
well for a number of women infected by infected men. I would
guess that on this hairsplitting business, I would wonder if Dr.
Wiley wouldn't agree with me that we certainly are going to have
a self-sustaining epidemic. The only issue is, is it going to be
a large and increasing epidemic in both growth and absolute
terms, or will we end up in this fifth stage of -- fifth
overlapping stage of the AIDS story with small pockets that don't
have enormous epidemiologic importance?

DR. WILEY: Well, I think the way I'm using the word
"epidemic" is to describe uncontrolled growth; that is, we
certainly will have cases, tertiary cases. The question is, will
each tertiary case produce one further fourth order case, at
least one? And that's what we're talking about here. It's not
so much that we won't see -- I think there's a Persian poem that
talks about the soup of the soup of the soup of the soup - - it's
not that we won't see these chains of transmission produce a
fourth and fifth and sixth order level. The question is, is each
level going to be weaker than the preceding one? Is each number,

220

 

 
 

 

set of numbers, going to be smaller than the primary? Is the
secondary going to be smaller than the primary, the tertiary
smaller than the secondary, and so on? Or do we have a case,
once the virus finds its way around the population, it's going to
run into some super- spreaders, some sexual networks that are
particularly conducive to generating HIV transmission, and all of
a sudden explode on us in a -- it's not going to explode at 50
percent prevalence in a few years, but I would say a 5 percent
prevalence in certain areas qualifies as a, you know, detonation

at least.

And so that's our real question about the question of
self-sustaining, not will we get these long chains of infection,
but will we get them in such a way that each case produces more
than one additional one? And that's when we start getting into

exponential growth.
DR. LEE: You didn't mention the prostitute problem.

DR. WILEY: Oh, the question of -- what was your
question?

DR. LEE: Well, I never see anything about the
prostitute problem, yet in 60 Minutes they mentioned that 75
percent of the African prostitutes are positive. Joyce Wallace
in New York says the streetwalkers are a third, and the call
girls are 1 percent. But this is a very, very dangerous -- and
God knows what the male prostitutes are -- a very dangerous
population.

DR. WILEY: I think Dr. Joseph can probably speak to
that more effectively than I can. But I know in the West, there
hasn't been so much seropositivity among prostitutes as we
thought there would be. And I don't think that's true in the
East. So I think I'll defer to hin.

DR. JOSEPH: There have been a number of studies now in
this country, as well as the studies done in Africa, looking at
seroprevalence in prostitutes, and we know two things.

First of all, unfortunately when people talk about
prostitutes, they automatically think they are talking about
female prostitutes with male customers, and I want to come back
to that. But with female prostitutes, we know that infection
rates in this country are high. They range from about 15 percent
to about 60 percent in various studies.

But the second and most important thing is that
virtually -- well, the vast majority of infected prostitutes in
all those studies are either themselves IV drug users, or they
have a steady sex partner who is an IV drug user. The second
thing is that those studies that have looked for ongoing

221

 

 
contact, transmission from female prostitutes to male customers,
have found to date very ‘low levels of transmission. There is
virtually no data that I know of that looks at male prostitutes
who have male customers or commercial sex in which the vendor is
a male and the customer is a female. And certainly you would
have to hypothesize that in an urban area such as New York, where
there are large, significant numbers of male prostitutes or
transvestite prostitutes, that the risk of transmission in those
situations is very, very high. But I think all the studies done
so far on female prostitutes with male customers have shown very,
very low risks of transmission.

DR. LEE: Doesn't that back up Dr. Wiley's hypothesis?
DR. JOSEPH: That what?

DR. LEE: Well, that vaginal transmission is more
difficult.

DR. JOSEPH: Well, I personally believe that there are
physiologic differences that make it much more efficient for a
male to transmit to a female than vice-versa, but that there are
many things that can alter that balance, and I think that's what
he's saying. It seems quite clear that genital ulcers of either
sex are one. Other infections may be another; stage of the
disease may be another, et cetera. And I would draw cold
comfort from that degree of difference of efficiency,
particularly when we have the African and Haitian experience that
can show us that it can happen otherwise.

I think we'll always have a differential between
efficiency -- between the two directions of transmission. But I
think basically what Sheldon said is correct, that differential
will not stop onward transmission, and the question really is, at
what level will we have that onward transmission.

MR. CREEDON: I'd like to just indicate for the
Commissioners a proposal as to how we might proceed. We're
thinking we will have lunch brought right down here, and we can
have lunch while we're continuing to be educated. We will take a
five-minute break when our discussions with Dr. Joseph and Dr.
Wiley are finished. And then that should give us enough time for
the next panel before moving on to the last panel at 1:15. So
why don't we continue now for a few minutes and ask whatever
remaining questions we have of Dr. Joseph and Dr. Wiley? Penny,
do you have a question?

MS. PULLEN: No.
MR. CREEDON: Rich?
MR. DevoOS: No.

222

 

 

 
 

 

MR. CREEDON: Ms. Gebbie?

MS. GEBBIE: I have a question for Dr. Wiley. If I've
understood what you said so far, you've been fairly clear that
despite some research that says the heterosexual concern may not
be as large as it's made out to be, you don't think that is
convincing evidence that we should just turn our back and walk
away from it. I see, then, two options for this panel in our
continued deliberations. One is to conclude that we don't know
enough about it. We should say it's an interesting problem.

Let us not make any policy recommendations in that area other
than to say, let's research it along the lines that Dr. Wiley and
others have outlined, but sort of walk away from it for now until
the studies are in. The other option or an other option is to
say, clearly studies are needed, but the best public health
judgment is to take the idea that it's real and growing and make
policy as if that's true, and then adjust downward later if, in
fact, the studies shows that was in error. Which of those would
you recommend?

DR. WILEY: Well, I think the second, mainly because
dealing with the existing forms of heterosexual transmission gets
you involved in certain kinds of educational work anyhow. And
some of that really applies to the general population of
heterosexuals as well. And so it seems to me just prudent to go
ahead and extend certainly educational efforts into the wider
community, and it may be that we will look back on it in a decade
and say, well, look, it really wasn't all that necessary.

But I think the dynamics of HIV transmission in this
population are sufficiently subtle that it may be awhile before
we can really pin down what the risks are, and I see no downside
risk in continuing to alert the general population that this is a
possibility, so long as the alert does not contain stigmatizing
messages of the we/they sort. So I think the second alternative
is what I would be in favor of.

MR. CREEDON: Dr. Walsh?
DR. WALSH: I'd like to get out of the genitalia.
(Laughter. }

And get into something more practical at this point,
and that is, Steve, you made some comments which I think are
germane and interesting in your testimony, that we need some
advice on. The appearance that there is substantial funding for
AIDS or adequate funding for AIDS is out there; it appears that
way. Nevertheless, you cited just one example of where money was
being taken from the section for sexually transmitted diseases
and so on and shifted over. Every Administration, as you know,

223

 

 
 

is subject to the advice and guidance of career bureaucrats who
tell them how to allocate funds wisely, and we get back into
what the Chairman has so often mentioned about the fragmentation
of that advice that is being received by laypeople in high
places. I wonder if you and Dr. Wiley both would like to
comment on a couple of things in this area for our benefit.
Number one, we hear a great deal about counseling and education,
which none of us disagree with, and you have made the point, Dr.
Wiley, that it takes people, not money. That's only half true,
because it takes money to train people. So you need money for
training counselors, and I don't care what group you're dealing
with.

Secondly, I am concerned with the budget reductions
that we're facing anyway, that maximum utilization must be made
of what funding is made available for AIDS, and minimum damage
must be done. For example, on the drug rehab program, if money
is being taken away from that and theoretically put into AIDS,
and then has to be channeled into drugs, you know, we're only
kidding the public. And I wonder if you felt that you could, if
not here, advise at some time, if you were in a position to
influence where the available funding would go, be it Medicaid
funding or anything else, how you would attack the priorities to
the greatest benefit of the public, because I think this
Commission is not going to be considered expert in rectal or anal
intercourse or vaginal intercourse. We're going to be considered
expert on many more of these more general things, which are
vitally important to win this war, and you're in the front line
in New York. You are having problems with apportioning money in
your own state and getting it in your own state. We can benefit
greatly from that experience, and I wondered if you'd like to
comment on that.

DR. JOSEPH: Let me tell you about some experience,
Bill, that I think bears upon that. I serve as the Coordinator
of the Interagency Group in New York City that has primary
activities with AIDS, and that includes not only the Health
Department, but Health in Hospitals, the Board of Education, the
Corrections Department, et cetera, et cetera, about eight
agencies. Now for me, as the Health Commissioner, HIV infection
is the number one item on my plate. Everything else either
relates to it or comes second to it. But to my colleague, the
Chancellor of the Board of Education, though he sees the
importance of the AIDS issue, AIDS is not and never will be and
never can be the number one issue on his agenda. And for my
colleague who runs the Human Resources Administration and Social
Services and Welfare Department, with all the problems that he
has, although he sees the important -- so the kind of, the kind
of coordinated Nirvana that Sheldon describes just isn't going to
happen. Nobody is going to get a blank check. Not everybody in
the government is going to say, "This is the number one priority
among all priorities."

224

 

 
 

 

I think what would do more than anything else in the
Federal Government to deal with that problem -- how I wrestle
with those problems in New York is another issue -- but I think
it would do more than anything else at the federal level to deal
with those analogous problems, is to have an Executive
Coordinator of AIDS activities. You can't put it in one place.
It can't be in Health and Human Services; it can't be in NIDA; it
can't be here or there, and you can't take all the AIDS out of
everything and put it in one new Department of AIDS, because AIDS
is so embedded in everybody's problems and everybody's programs.

So you need some coordinating authority. But a
coordinating authority that is in name only or has a sort of
policy judgmental advisory is not going to get the muster cut. I
think you have to have someone with a small body of staff that
has some direct executive leverage, and somebody has to decide
how tight that is or how heavy that is, but have some direct
executive leverage on the various agencies, specifically those
with regard to drugs and specifically those with regard to
health, but there are many other agencies such as the Department
of Education, et cetera. It has to be an executive appointment.

DR. WALSH: I'm not getting at whether we have to
convince somebody that this is the number one priority. From the
standpoint of this Commission, AIDS is the number one priority.
What I'm interested in is, can anyone advise us if there is $800
million or a billion dollars --

DR. JOSEPH: Where the money is to go?

DR. WALSH: What are the list of priorities in the war
against AIDS that we should be addressing? We know the drug
issue is a great one. We know education is a great one.

DR. JOSEPH: I gave you the three in my list.
DR. WALSH: How can we do it?

DR. JOSEPH: On the prevention side, I think the three
stand right out at you, and they all have to be done. It's the
education; it's the voluntary counseling and testing; and it's
the specific activities with regards to drugs. On the clinical
and services side, there's just going to be an awful lot of money
rolled into that, no matter what anyone wishes, and I think the
concentration in terms of new funds needs to be on the prevention
side. I'm really not the person to talk about where the research
dollars ought to go in biomedicine, but in prevention, it's
education, testing in an appropriate context, and a serious
primary effort to deal with the substance abuse problem.

225

 

 
 

 

_ DR. WALSH: With the reduction of the federal share of
Medicaid and the increase in Medicaid funding going to care of
AIDS patients, we are going to and are already running into
special interest groups who benefit from Medicaid, like the
elderly and the others. What is the answer to those groups?

DR. JOSEPH: AIDS only magnifies weaknesses in our
health and social services system that we already have. On any
given day in New York, we have about 60 patients who are
medically ready to leave hospital, who have AIDS or ARC, but
can't because they have nowhere to go. But we have 1200
patients, old people with strokes, broken hips, et cetera, who
are ready to leave hospital but can't because they have nowhere
to go. And AIDS will be the thing, in my view, that eventually
puts this country into a system of universal health insurance
that we probably should have been in several decades ago. I
don't think there's any choice about it. That's just going to
happen. The demands for funding those clinical services and
brokering them out across all the needs are going to be
irresistible.

MR. CREEDON: I would like to turn it over to our
Chairman, but I don't want anybody to feel slighted at this
point. Cory, you had a question?

DR. SERVAAS: Well, I have a question for Dr. Wiley, if
it's all right, if there's time. Dr. Wiley, I'm interested in
the prevention and the spread of AIDS, and I'm curious about the
JAMA report where it was reported that more women, by Dr. Bruce
Volker from a foundation in California -- and I don't know if you
know him -- but he reported more women were vulnerable for anal
sex than all the homosexual community together who practice anal
sex. Do we have any studies, and do studies show how much
difference there is in this, whether we're guessing or whether we
know, and do these women have the same increased incidence of
rectal hemorrhoids and all the things that the homosexuals have?
Do they have more hepatitis and syphilis and all those things
than women who don't practice anal sex?

DR. WILEY: Well, I think there is some, oh, anecdotal
evidence that anal intercourse as a heterosexual practice
increased over the last 30 years and became more common, but
never quite as common.as it has been in recent years among gay
men. I think that you don't observe, as far as I know, the
so-called gay bowel syndrome in women who practice anal
intercourse. There may be some additional problems that people
experience because of it, but I think that in gay men what seemed
to be an increasing frequency of anal intercourse over the late
‘60s and '70s became associated with a whole syndrome of
conditions, because there were epidemics preceding AIDS, six or
seven of them, many of which were facilitated by anal
intercourse. I'm thinking of shigella and shigellosis and

226

 

 
 

 

giardiasis and amoebic infections and hepatitis B and hepatitis A
and so on. So, you know, I think those consequences of anal
intercourse among heterosexuals haven't yet been seen, so far as
I know, but I am not an expert in the treatment of sexually
transmitted diseases, so I don't know what people are seeing in
the clinics.

DR. SERVAAS: In your opinion, should we be alerting
women to avoid --

DR. WILEY: Oh, to avoid anal intercourse? I think
perhaps not any more than we alert them to avoid vaginal
intercourse under certain conditions, and that if they're going
to have intercourse at all with people they don't know, I think
-- again, I suggest the rule of prudence. Use a condom and
practice safe sex.

So I don't think you could single out anal intercourse
as a particular risk among heterosexuals and create a special
program of saying, "Don't do that." I think it's really part of
the whole package of safe sex among heterosexuals. It's not an
enormously common practice, but it does happen.

MR. CREEDON: Chairman Watkins?

ADMIRAL WATKINS: Two specific questions of both of
you. I'd like to have each one of you address the answer to it.
We're supposed to be here today on incidence and prevalence
database, what we might do better in future surveys that are now
contemplated. You talked about the family of surveys. We really
have to focus on that issue. That is the purpose of this one.
We're going to ask perhaps even both you back again, if you'd be
willing to come back. None of us own big stock in Pan American
or the Eastern Shuttle, and so we're going to ask you to come
back, because we're going to be unfortunately having to take on
these issues sequentially. We're trying to take on incidence and
prevalence. Next week is going to be focused on drug abuse.

Care in mid- January. Discrimination is going to come up.
Education is a major hearing underway. We're trying to prepare
for those hearings. So I don't want to get too far afield. I
want to get back to the incidence and prevalence issue right now.
We talked this morning a little bit -- in fact, I think it's been
very educational on these tertiary case numbers to fix the
heterosexual risk pool. What does that mean in terms of
epidemiological database that you would like to see in the CDC
reports, for example?

Is there a display that is better than we have now to
predict what we might want to fill up in an area that's voided
now? Is there a breakout that you would see, as the surveys that
are coming up, just approved by the Domestic Policy Council and
the White House? Are there new ways to approach it? Should the

227

 

 
 

data in the survey itself contain other information? After all,
if we're going to go out and get this information, let's get the
information that you're concerned about. So can you recommend to
us perhaps a new display, Dr. Joseph, in the CDC data that you
might provide us that would be a catalyst to move in a new
direction? Let me expand that even further. Is the gray zone of
ARC, where we've just expanded to AIDS versus ARC, is that really
a temporary aberration on the statistical database, and is there
even merit to displaying it a different way, so that HIV
infection becomes our greatest concern? If we're talking about
education and that sort of thing, we seem right now to feel
comfortable that if you're not talking about AIDS, you forget
about the HIV infection. I mean, generally that's the feeling
you get, and there's a great deal of confusion when you're
talking about AIDS as opposed to asymptomatic infection.

So is there another approach now, or are we going to be
locked into the bureaucratic techniques that we now have, and
we're just going to regurgitate data that is not going to be that
satisfying, either in projecting the perceptions in the American
public of where the real focus should be, or in getting the
right data to try to pin down some of your concerns, the people
who work in the field and research in this area, that might
improve our understanding as this thing begins to unfold? And
isn't now the time, before we survey, to make sure we've wrung
that out and are satisfied? Would you take that one question on?

DR. JOSEPH: I can give one comment and five series of
data-collection alternatives. The comment is, your point is very
well taken about being misled by looking at AIDS or even AIDS and
ARC. The incredible thing about the reversal that we've
undergone in New York, looking at deaths in IV drug users who
were never categorized as AIDS or ARC is that it has shown us a
very different epidemiology than we have assumed for several
years was the epidemiology. It has shown us that the burden of
significant disease and death among IV drug users is greater than
that among gay men. So your point is very well taken.

The five points around data collection and
interpretation. I would once again say -- we've all said it --
that basically I think the projections that you are working with
currently, the CDC range of estimates, I think are an adequate
range of estimates to work from, certainly at this point in the
epidemic. I wouldn't beat that to death looking to either
refine those numbers or looking for the outliers, either higher
or lower. In terms of work that needs desperately to be done
epidemiologically, the sentinel survey approach that you've
heard from all of us is very important. It's important
geographically; it's important in terms of demographic groups. I
think we need more surveys in particular population groups. I
don't mean that in any way as a code word for racial and ethnic
groups. I mean teenagers. I mean women in the reproductive age

228

 

 
 

 

range. I mean older people. I think we need to pick certain
population groups and use them as sentinels as well as geographic
sentinels. Again, that all must be done on a voluntary basis.

The next piece of evidence we need is that we need to
be trying to find a way to classify and project what Sheldon
Landesman refers to as tertiary cases. We need to have the CDC
project those people who are cases at this point, or those people
who are infected, as we get our survey data in whose only risk
was sexual exposure to a person who was infected, but who was not
in one of what used to be called the primary risk groups. And
finally, what we badly need -- and it goes back to my first point
-- we need studies that will teach us more about the natural
history of infection as it relates to disease. We are still
dealing -- it hasn't come up this morning -- we are still dealing
with this very muddy concept of perhaps within five years 30
percent of people with full-blown AIDS and another 30 percent
with ARC and who knows at seven years. We need to refine our
understanding of what the relationship is between infection and
disease as we go out in time. And there really is not much
that is continually projected in that regard. CDC could do more
there.

MR. CREEDON: We had data from Dr. Curran on that
yesterday, but I mean, where he left us was, it was unclear where
it was going.

DR. JOSEPH: But even though it's‘unclear, we need to
keep projecting it.

ADMIRAL WATKINS: So as not to be redundant, Dr. Wiley,
would you just add any comments on top of that.

DR. WILEY: I basically agree with what Dr. Joseph
said. Just two comments. There is no simple way to improve the
routine data presentation which you referred to. The data that
comes to CDC comes from a variety of sources, and it doesn't
include, for example, the counts on the number of ARC patients
that have been discovered in various medical contexts all over
the country. And there's no way that that's going to happen.

It would be really important for us to know those things, but I
don't think it's going to happen very soon, because we don't have
reporting of ARCs.

ADMIRAL WATKINS: But in New York, they don't use ARC,
I understand. That's not a helpful and useful classification to
them in the actual clinical work that they're doing. It's people
ill that have the virus. Now is there a new technique that
should be used in the CDC reporting procedure that would make us
understand that a little better, or is ARC clear enough to
everybody? Doesn't it send a signal that somehow we have
separated AIDS from HIV infection?

229

 

 
 

 

it 1

, pte doy t ee: ce 1.

DR. WILEY: They don't get reports of people who don't
meet their existing criteria for case definition. They don't
even get them.

ADMIRAL WATKINS: Let's go back to case definition
then. Is there a new definition that we should apply that would
give us a better understanding among the American people of what
this is all about?

DR. JOSEPH: May I jump back in for just a second?
DR. WILEY: Yes, go ahead.

DR. JOSEPH: What we need to know are the numbers of
people who are infected but asymptomatic, the numbers of people
wno have any sort of symptoms from their infection, the number of
people who have CDC defined AIDS and the relationship over time
between those three groups. That is what needs to be projected.

DR. WILEY: Yes, but you can't construct that from the
sort of thing that CDC gets and although they could say to health
departments, "Give us that," the health departments couldn't give
it to them either because the cases come in by diagnosis.
Usually, there is some sort of operation or biopsy proven this or
that that leads somebody to believe that this is an AIDS case.

ADMIRAL WATKINS: So I take it you are making a
recommendation to the Commission that we do nothing to make
recommendations on changes to the sero prevalence data base
presentations?

DR. WILEY: No, not that, that it is not going to be
easy to recommend anything that is useful and feasible at the
level of case reporting except perhaps for reporting cases of
tertiary cases among heterosexuals. I think that would be
something that could be done.

MR. CREEDON: Why wouldn't it be feasible to ask the
departments of health that report to CDC any case where medical
treatment is required and is.related to the fact that the person
has the virus?

DR. WILEY: If you knew the person had the virus.

MR. CREEDON: Well, if you don't know he has the virus,
isn't every case that is reported, isn't it established that the
person has the virus?

DR. WILEY: Every AIDS diagnosis. This tends to be
people pretty far along in disease. They are very sick and they
had to come to doctors. People with ARCs can go for two or

230

 

 
 

 

three or four years without ever seeing a doctor depending upon
their condition.

MR. CREEDON: Then, of course, those cannot be
reported.

DR. WILEY: That's right. So you miss the great bulk
of HIV infections that have produced symptoms and there isn't any
way to recover them in the normal reporting mechanisn.

DR. CRENSHAW: Yes, but there could be.

MR. CREEDON: Could we require reporting of people who
have the virus and who have appeared for medical treatment
related to the virus? You don't have to call it ARC or anything

else.

DR. CRENSHAW: You don't diagnose ARC unless you have
an infection.

DR. JOSEPH: The concept of ARC as was mentioned a
moment ago is the person who has HIV infection and tuberculosis
and no other symptoms, is that a case of ARC or not. Those are
the people, that IV drug user who died of tuberculosis or
endocarditis never had an HIV test and when you go back and look
at his chart, that is how we turned up these 2,500 people, you
find out that this was an adult male who had Thrush so you make
the reasonable presumption then that he was HIV infected.

Most of those people we never see. We can't pull them
out. We have to estimate them by Sentinel surveys and other
sampling techniques and try to get a better fix on this
relationship between asymptomatic infected, all other symptoms
and illnesses with HIV infection and the CDC-defined cases.

ADMIRAL WATKINS: But your "ill with AIDS" really ina
way has been an attempt to come out of the ARC dilemma. It
sounds like to me that your "ill with AIDS" is still a valuable
piece of information irrespective of the fact whether
tuberculosis is solely AIDS driven.

DR. JOSEPH: Terribly.

ADMIRAL WATKINS: It seems to me that kind of data
would be very valuable in the whole staging of the from
infection to fully symptomatic AIDS.

DR. WILEY: Many ARC patients are treated as ordinary
patients with particular complaints and their sero status is not
known at the time of treatment. They may know it and not wish it
to be in their medical record.

231

 

 
 

ADMIRAL WATKINS: I understand that but then it is not
reported. But when you do know it, then it seems to me that it
in itself becomes a very valuable tool in understanding the
staging of the disease. Anyway, I think you have answered my
first question.

DR. LEE: Mr. Chairman, are you asking would they
recommend that all HIV positivity be reported to the CDC which is
not now done?

ADMIRAL WATKINS: I am asking them for any
recommendation they would make for us, for the Commission, to
make a representation to Health and Human Services and Centers
for Disease Control and others perhaps, to the President, through
the President, on changes that you would like to see inserted
into the means of obtaining this information and that would
better fill the voids of understanding that we now have. We need
to start leading to that now rather than wait until two or three
years of studies and find out that we just wasted three years
when we could have gotten data that we desperately needed so why
not get it now even though we are not sure what to do with it.

DR. JOSEPH: We will write up those four that I
mentioned and send them to you in short order.

ADMIRAL WATKINS: We would like to have that.
Secondly, I think, to close out this because we really have to
move on is, you are going to be, both of you, involved to a
certain extent, you, Dr. Joseph, perhaps more, in conducting the
30 standard metro statistical area surveys in the family of
surveys now underway and being prepared.

Is the data that we are going out to get sufficiently
surrounded by those surveys; are you satisfied that we are not
losing an opportunity here in a number of areas, that is, just
the survey information getting; should it be enhanced? Should we
make recommendations to CDC to enhance that information, to grab
hold of it and if you are going to go out and get this, let's
get some additional data along the lines we just talked about
perhaps to put it into our data base and most importantly, can
you recommend the elimination of certain obstacles that will be
in the way of achieving the full potential value of those surveys
in the nation.

If, in fact, we are going to be rejected by some of
the people in the high risk behavior areas as the statistical
survey gets conducted, then is there a way to prepare the path
from the President to the state governors and to others in the
health profession business to maximize the value of those
surveys? If we just let the people who are in the business do
their normal thing, they will do the best job possible but have
we done all we can as national leaders to ensure receptivity, to

232

 

 
 

 

ensure anonymity is given maximum credibility so that the people
will come forward. Can you give us your concept of a strategy
that would enhance the value of those surveys? This would be
very valuable because that is something we can do right away and
make recommendations if we could get a cohesive strategy to say,
"This is the way we think, Mr. President, you should precede
these surveys nationally and that they should flow then through
the governors, the highest level of leadership to the mayors
right down into the population in order to maximize it" and get
the television programs and so forth building the credibility of
the system with the American people to enhance their value of the

survey.

DR. JOSEPH: We will send you some specific technical
comments on that point.

ADMIRAL WATKINS: Do you think that is important?

DR. JOSEPH: I do think it is an important thing.
Remember, of course, that it is we who will do the surveys. The
cpc will send us a Texas Ranger or two but it is going to be the
local health departments who will be doing the surveys.

ADMIRAL WATKINS: But won't you be frustrated by the
turn downs and your inability to get the very best data that you
would like to get? Wouldn't it be better to optimize that before
we start the survey?

DR. JOSEPH: I think that point is well taken and we
will send back some ideas both about the fostering and advocacy
that is necessary to make the surveys run as well as possible and
any particular technical points we have.

ADMIRAL WATKINS: And proof of anonymity so that
credibility is really going to be there. Can we enhance that
credibility in the high risk group or is it an impossible
situation?

DR. JOSEPH: If I may, Mr. Chairman, and I said I
wouldn't come back at Mr. DeVos' remark and those who know me in
the panel know that I mean no personal disrespect but I call them
as I see them, if I may, in New York our problems with anonymity
are not around the way we conduct our program in New York; our
problems with anonymity occur when Congressional or
Administration officials launch trial balloons which seem to
threaten the protection of anonymity and confidentiality. For
us at the local level, that has an enormous impact. That is why
I went to such great pains to comment on Mr. Helms' amendment.
They will do more to threaten the ability to carry out those
anonymous surveys in New York than anything that we will do in
New York in the way we carry them out. That is a very important
issue.

233

 

 
 

 

DR. WILEY: I agree. I think’all of us who have been
in the survey business and have dealt with problems of refusal to
participate on other studies shudder to think of the problems
that we would have on doing AIDS studies and so geared up for it
in different ways and one of those different ways was to mobilize
community support for surveys that required people to give blood
and talk about their sex lives and to do a lot more of that than
we had ever done in past surveys.

ADMIRAL WATKINS: May I ask Dr. Wiley who is in the
survey business to coordinate with Dr. Joseph on your input on
this particular issue to us. I think it would be very helpful
since you are here, you are the witnesses today, both of you are
experts in this area and have been at the grass roots. We need
to have that and I think that is something that we would want to
do right away. I don't know exactly where they stand in the
execution of that program but I don't believe it is too late yet
if we move expeditiously.

Thank you very much for coming today. It has been
extremely helpful and we would like to again keep our dialogue
open. The questions that many of our panelists were unable to
ask this morning, we will send you and would ask that you try to
get back within a couple of weeks of receipt of those to give us
your best insights for the record and for our recommendations to
the President in February on this particular issue of incidence
and prevalence.

MR. CREEDON: Thank you very much, Dr. Joseph and Dr.
Wiley. We are going to take a break for about ten minutes.
Lunch will be served up here during that time and we will
reconvene in ten minutes at which time I would hope Mr. Edwards,
Dr. Delgado and Mr. Levi would all be up here and ready to
testify.

(Whereupon, a luncheon recess was taken at 12:35
o'clock p.m., to reconvene at 12:55 o'clock p.n. ]

234

 

 
 

 

 

AFTERNOON SESSION

[12:55 p.m. ]

MR. CREEDON: Our next panel is intended to give us the
perspective of some community based organizations and our first
witness is Mr. Don Edwards of the National Minority AIDS
Council. Mr. Edwards.

PANEL 4 - COMMUNITY BASED ORGANIZATIONS PERSPECTIVE
PRESENTATION BY MR. DON EDWARDS

MR. EDWARDS: Thank you, sir. Good afternoon, Mr.
Chairman and Commission members, my name is Don Edwards. I am
the Executive Director of the National Minority AIDS Council.
NMAC is an organization dedicated to working with and through
national minority leadership organizations, minority businesses,
local minority community-based AIDS projects and national and
local health agencies in assisting to prevent the spread of HIV
and helping minority communities to better care for minority
persons already infected.

I am very pleased to be here to offer testimony as
part of these hearings which have been looking at data collection
and related issues. Persons perceived as advocates like myself
are often excluded when the focus is on the more technical
aspects of scientific phenomena.

But I am very happy to be here because it is very
important to practice advocacy in this area as well. I come as a
novice epidemiologist who was the first person in the U.S. to
look at the CDC's Public Access Data Set to describe the
epidemiology of AIDS among Black persons born in the U.S. Thus,
I want to assure you that with my distinguished co-panelists, I
am quite interested in the numbers as regards the HIV epidemic.

I spoke with a number of individuals in the Black
community in preparation for this presentation. I would like for
you to know that in the last 60 days, there has been a quantum
leap forward on the part of leading minority clinical researchers
and health practitioners. Just last week, for instance, the
National Institutes on Drug Abuse followed a technical review
meeting hosted by your fellow Commissioner Dr. Beny Primm. He
and his colleagues at Addiction Research and Treatment
Corporation in Brooklyn brought together minority scientists,
health providers, and community activists from across the nation
to discuss issues relevant to the subjects on which you are
taking a look at today. In two prior meetings across the
country, one sponsored by the Detroit chapter of the National
Medical Association and again by the Provident Medical Society,
the Brooklyn NMA affiliate, critical attention was again devoted
to these issues.

235

 

 
 

 

I would like to just raise some of the questions that I
got back as feedback from the folks I talked to around the
country. First of all the question that seems to come up most
often is, is it as bad as they say it is for the Black community
and this seems to be a question essentially of is the news really
something we can lend credibility to. As you know, the reality
is that it is. As of November 23, 1987, 47,022 cases of AIDS
have been reported to the CDC and of that number, 11,746 were
Black. This number represents 25 percent of adults and
adolescents over the age of 13. The rest are 373 and these cases
are in children under the age of 13.

I guess what I really want to say about this
information, you can look through my testimony, is that the
response that I have gotten around the AIDS surveillance program
from the Centers for Disease Control is one of essential support
and commendation for the efforts that they have taken to ensure
that the nation has a regular update of this data.

Yet there are some concerns and I would like to share
some of them with you. First of all, a number of persons raised
the issue that it may be needed to institute systematic
standardized surveillance and reporting of cases in the top 25
high prevalent cities. These cities coincidentally also have
significant percentages of Black populations.

The point is not to describe one national epidemic and
leave it at that because it seems for some folks who are on the
street level that there may be, in fact, a number of regional or
local epidemics occurring which, in fact, look very different
from one other. Additionally, considerable dissatisfaction seems
to be existing regarding the lack of standardization and the
quality of surveillance coming out of states and municipalities.
For instance, 40 percent of the cases in Alabama are Black, but
hardly anyone in Alabama or out of Alabama knows this.

Black activists are also nearly unanimous in their
desire to have national and state geographic data aggregated by
race, data on gender aggregated by race and especially AIDS cases
by risk factor combinations aggregated by race. There has also
been a concern that it might be useful in separating for
purposes of following cases and developing programs to see cases
in the homosexual/bisexual category separated since it has been
reported that there seems to be a higher percentage of Black
males who are practicing male-to- male sex who do not identify
themselves as homosexuals in the Black community.

This aggregation of data might be useful to have on a
monthly or bimonthly basis but it is crucially needed to confirm
or deny anecdotal community information. The question is if this
responsibility falls to the states, the CDC should provide the

236

 

 
 

 

standard for the collection and reporting of this information
thus ensuring its standardization and generalizability. There
are legitimate issues regarding breach of confidentiality in
cutting data this fine but I believe safeguards can be in place.

The main point that I bring is that surveillance needs
to reflect a greater sensitivity to race and ethnicity as a
highly significant variable. My second point, did AIDS become a
health problem for Blacks only recently? If yes, how did it
happen so quickly and if no, why didn't we know about it sooner?
I have to tell you that as more information is disseminated many
people at the community level feel it is going to become clear
that the scientific awareness of the disproportionate impact of
the HIV epidemic on Blacks and other minorities was evident long
before the publication of the report entitled, "Acquired
Immunodeficiency Syndrome Among Blacks and Hispanics - United

States." published by the CDC on October 24, 1986.

There is concern among Black AIDS activists that as
more Black persons learn about the early evidence of the
disproportionate impact dating back to at least 1983, that there
may develop a credibility gap as to whether any other data is
being held back. The suspicion and anger potentially engendered
may detract scarce resources from education and direct services
at a point where progress is just beginning to be made. The
issue around this question is not simply mere curiosity. At the
heart is again the question of credibility of the current
message about the impact of AIDS in the Black community and of
the messenger.

The fact is that there was available information had it
been interpreted in a more informed manner as early as 1981 which
might have led to Blacks understanding that there was a higher
impact than initially understood. Had this information been
disseminated, we might have had the kind of report that came out
in 1986 much earlier. Another issue, why does it seem that all
the bad news on AIDS is about Black people? Are all the studies
just another way to blame Blacks for AIDS? I think you know as
well as I do that numbers paint pictures. For Blacks, the
picture has been consistently suspect and in the case of the
Haitian community, in particular, particularly ugly.

In the minds of Black persons, the Haitians were
essentially blamed for bringing AIDS to the United States.
Initial reports on AIDS in Black Africa seemed more interested in
pointing out how the governments of these countries were standing
in the way of scientific progress, of help for their citizens,
and of an explanation of where AIDS started.

Lastly, in the U.S. data has been looked upon and used
to fuel perceptions that it is the Black drug addict who is
spreading AIDS across the country and that because of them, AIDS

237

 

 
 

is going to infect heterosexuals and kill hundreds of
unsuspecting women and innocent babies. I think one of the
things that we should keep in consideration is that the Black
person on the street looks at perceptions much more from the
perception they have of their place in the society and we must be
careful that any research that is conducted, any data that is
gathered, does not poison an atmosphere by relying on essentially
stereotypical, insensitive and uninformed processing of
experience and of data.

I hope that by now the fundamental importance of
sensitivity to culture and tradition is an accepted and integral
aspect of our research and will no longer contribute to the
erection of tedious and divisive barriers. Researchers also need
to keep in mind some points. First of all, the U.S. Black
community is not a monolith. In the U.S., Black identity
includes early and recent immigrants from Caribbean nations who
may speak a number of different languages, native languages as
well as those of their former colonizers.

The Black community consists of separate ethnic
communities joined into one multi-ethnic community by African
ancestry, common cultural values, cross-cultural marriage, and a
common experience of racism and racial pride. There is also a
psychology at work in the Black community that we need to
understand that one of the key aspects of that psychology is to
function in the role of self- preservation. Many people are
recognizing what appears to be a kind of denial already
aggravated by the stigmatization associated around AIDS in the
Black community that has produced a "See no Evil-Hear no
Evil-Say no Evil" process that resists education related to AIDS.

Disrespect of this psychological process only serves to
reaffirm its need to exist. It would be particularly ironic that
in that case of AIDS if this psychology did not lead to self-
preservation but to death. Another question, is the impact on
Blacks really so disproportionate? There is considerable
disagreement on this question. From a scientific standpoint, a
number of Black researchers have pointed out that using general
population figures for denominator data drives the proportion up.

Epidemiologically, it would make more sense to use
either the percentage of Blacks in the urban center, or in the
metropolitan SMSA as the denominator. On the East Coast in
particular, given the high percentage of Blacks in these cities,
the ratio would probably begin to approach one to one. Others
hold that the two to one ratio is correct considering the
projected spread of the epidemic into the heartland of the U.S.,
and the higher degree of undercount that is suspected for Black
cases. Is there a genetic link which explains the high
incidence of AIDS infection among Blacks? You should understand
this line of investigation waves a red flag in particular for the

238

 

 

 
 

 

Black community. Rightly or wrongly, anytime genetics is
mentioned, it raises the specter of ever- recurring racist
attempts to explain the lives of Black people with an update on
archaic pseudo-scientific social biological theory.

There is legitimate research to be followed on the high
incidence of HTLV-1, for instance, in some parts of Africa which
are contiguous with the presence of HIV, as well as in the
presence of HTLV-1 in Caribbean nations and that should be given
appropriate attention. But the Black community is particularly
sensitive to anything that remotely suggests something innate to
being a Black person increases one's risk for infection.

With Whites, it seems only gay men get AIDS. With
Blacks, men, women and children get AIDS. Why? Though this
epidemic is six years old, you should remember that it was only
13 months ago that the CDC published their initial report on AIDS
Among Blacks and Hispanics. Thus, it is still difficult for many
in the Black community to accept that data and information even
though it delivers a clear and unmistakable message and that
message is that AIDS is not a gay white man's disease.

But for the most part, where it appears that so much
has been determined about the sexual practices, the degree of
behavioral change, the percentage of population, et cetera, for
gay men, Blacks have been mostly left with a lot of interesting
but unanswered questions. Where are the longitudinal studies
exclusively for Blacks or at least oversampled for Blacks? Where
are the case control studies linking cases and controls to
ascertain independent or multiple variables? For the category
homosexual/bisexual male, where do most Black men fall?

For homosexual/IV drug user/ which category contributes
the greatest to risk? If there is one area in which there is
almost complete unanimity, it is that the degree of scientific
investigation absolutely contradicts the declaration that a
crisis really exists. In addition to the potential "genetic"
factor that some have raised, there are two other more
legitimate avenues for exploration and research. One is that the
racial distribution of AIDS cases does in fact reflect the racial
distribution of populations that are affected in the high
prevalence areas. A second factor suggested that has gotten very
little attention is that economic and cultural factors in this
country also may account for the observed differences in
incidence.

Both of these lines of investigation lie dormant.
Perhaps in one of them lies key information which can help to
save lives in the Black community as well as bolster the efforts
that this country is engaged in with tremendous credibility. The
question, however, I want to end on and that has been raised in
the community is, "Do people in this country really want to know

239

 

 

 
 

|

about the conditions of Black health?" "After they know, will
they spend money that will make a change." "If they start with
AIDS, will they really use what they know to make a difference?"

Commissioners, this is a mere sampling of the questions
which I was asked to share with you by key individuals in the
Black community. I would like to particularly acknowledge Dr.
Wayne Greaves, Chief of Infectious Diseases at Howard University
Hospital, who was to my knowledge the first Black physician and
epidemiologist to raise some of these hard questions around the
country.

I want to call your attention in particular to an
article I included at the back of my testimony entitled, "AIDS
Update: No Longer Gay White Man's Disease" dated November 25,
1983 which to my knowledge was one of the earliest news articles
mentioning the impact of AIDS in racial and ethnic communities
and I also want to draw your attention to a subsequent article
talking about Dr. Greaves' own research in 1985.

Additionally, I have submitted documents that give you
some information about the epidemiology and some aspects of AIDS
as related to the Asian American community and the Native
American community which I hope you will look at and include in
your deliberations. Thank you very much.

MR. CREEDON: Thank you, Mr. Edwards.

DR. CRENSHAW: One of the things I found as a
continuing source of confusion and frustration in the epidemic
is experts in America looking at things that are going on in
Africa, such as years ago when it appeared to be heterosexual in
Africa, but we hadn't accepted it in America yet as a
heterosexually transmitted disease. The constant answer I heard
was "Africans are different." Somehow not saying but implying
they are not quite like other human beings.

I think we can learn so much from what is going on in
Africa and yet there seems to be a real barrier in America to
accepting that and extrapolating that and generalizing it to
other populations, even though there may be various other factors
that play a part. Can you comment on that and share any of your
thoughts?

MR. EDWARDS: I think that the black community, and
this is who I feel most comfortable speaking about, in terms of
what educators have run into, has been very suspect of how the
whole African issue was raised initially, and that has really
focused on what a number of people perceived as essentially a
very superficial kind of determination that again the African
people, the black people of the African continent were at fault.
They were at fault because potentially they may have eaten the

240

 

 

 
 

 

 

green monkey -- I don't even want to elucidate all the kinds of
ridiculous notions that came up, as to how this animal virus,
what we suspect to be an animal virus got into the human host.

I think there is a tremendous amount to be learned from
the experience in Africa, but I have to tell you, AIDS in most
people in the black community's perception has simply brought to
the attention of the world pre-existing health conditions in
Africa that escaped the attention of the world prior to AIDS.

I don't think we can superimpose on that perception
that had it not been for AIDS, the health conditions of Africa
would essentially still be ignored. I think there is a
credibility gap in terms of what is driving the interest. Is the
interest simply a solution and once you find a solution, you are
going to extract your information, your resources, pull out and
leave Africa to fend for itself, and I can tell you from an
anecdotal basis, this is something I think the black community is
very concerned about and waiting to see how it is going to play.

I think from a scientific standpoint, there are a lot
of studies that need to be done in cooperation with African
countries. If this can be done in such a way that is sensitive
to the national agendas that support the development of the
health sectors of that country, I think America as well and in
particular the black community of this country will see that the
approach to what is being done on the African continent bespeaks
the sensitivity that could represent the way things could be
done, for instance, in this country. Black people pay a lot of
attention to what happens to black people around the world. I
think recognition and sensitivity to this can go a long way in
this country and in Africa as well.

MR. CREEDON: Dr. Primm?

DR. PRIMM: Mr. Edwards, you talk about whether the
impact on blacks really is as bad as they say. Is it really so
disproportionate. You go onto talk about a one to one ratio, two
to one ratio. Could you explain that for me? I don't see how
numbers can really lie. Are you doubting the reports of CDC?

Are you doubting the reports of the New York City Medical
Examiner's Office? Are you doubting the reports of the National
Funeral Directors Association, which is a black funeral directors
association? What are you trying to get at there, if you would
explain that.

MR. EDWARDS: Certainly, sir. The way this issue was
raised and perhaps I have done it a disservice, is not so much
that doubt is raised as to there is clearly a significant
problem facing the black community in this country. The question
is not so much whether it is two to one or one to one. The
question is what is the usefulness of how you look at the

241

 
 

 

problem. If the problem is described as a total population
problem, then you need to have that linked to an input of
services and resources that are going to approach it as a total
population problem. Some people have said this is a very useful
way of looking at it and it may in fact drive funding. It may in
fact drive, if you look at the classic formula around the
prevalence, the prevalence determines the allocation of resources
and services. Yet, many people in the community have raised that
in previous and still outstanding health problems facing the
black community, that this formula has not been followed.

The question of whether it is two to one or one to one
is raised as a question for usefulness. If it is one to one, it
may be more useful in relieving some of the stress and some of
the stigmatization that black people perceive around the question
of AIDS which makes them more open to AIDS information. I think
in many ways people really don't care how many, whether it is 25
percent, 30 percent, we recognize that as another health problem
in the black community, we are over represented. The question is
what is it going to get us?

DR. PRIMM: I understand that. What I would like to
deal with as much as we can is that we do know blacks are
disproportionately represented in those who have been diagnosed
thus far to have full blown AIDS. When I am looking at things
like this, I am wondering whether it is that we don't believe
that. I think that is a fact. I think we have to base all of
our decisions on those facts. The other thing that I would like
to talk about is that you spoke about AIDS being a window of
opportunity to do something about the health problems of blacks
in Africa. Indeed, that is so. In this country, I think it is
just as apparent as it is in Africa. If you look at Harlen,
Washington, D.C., the services, they are poor quality health care
services, not only are they poor quality and inaccessible, they
are dehumanizing. I can go on and on. I think it is a window of
opportunity for this Commission to look very closely at the rates
and incidence of tuberculosis in certain communities, certainly
to look at subacute bacterial endocarditis and
glomerulonephritis, et cetera, all of those different diseases
that plague intravenous drug users and inner city dwellers, and
come up with some recommendations around those issues as well.
When you talk about Africa, I think we have to talk about home
also. I think that is really important.

MR. EDWARDS: I could not agree with you more, Dr.
Primm. I think that the perception that was created around the
country unfortunately with the dissemination of the Secretary of
Health and Human Services!' report on black and minority health,
helped raise some sensitivity and awareness of the fact that
there are 56,000 excess deaths in the black community alone,
simply related to the big six diseases. AIDS was not even part
of that consideration in that report. What we see in this

242

 

 
 

 

country is both a crisis in the black community related to the
health of that community raised by AIDS, but in fact, Dr. Primm
has identified that within that crisis, there is tremendous
opportunity. If we look at teenage pregnancy, substance abuse,
sexually transmitted diseases, AIDS not only gives us an
opportunity to make a fundamental injection of resources and to
change people’ s consciousness and awareness about those issues,
but it may in fact allow us to break what has been a kind of
policy deadlock regarding the link between looking at these
health problems in black and other minority communities as it
relates to different financing mechanisms and other health care
financing issues in this country.

The black community has been burdened for some time by
a number of epidemics. We really should take advantage of the
kinds of technologies in education and behavioral change that we
are developing around AIDS and apply them as well to other
behavioral problems that lead to health problems in the black
community.

DR. PRIMM: I think the National Minority AIDS Council
has a very unique opportunity. I would like to ask my Chairman
to ask the Council, wouldn't it be good if you provided us with a
whole laundry list of things that we could recommend that would
help straighten out some of these very problems that you cite.
You have a board now that is constituted. You have an office
here in Washington. I think it is a very unique opportunity to
contact say the National Medical Association, some of the other
black organizations, the National Urban Coalition, the National
Urban League, which are very interested in this problem, and get
their recommendations and give us a laundry list of things that
those organizations and you feel ought to be done. That perhaps
could be coordinated out of your office.

MR. CREEDON: I would like to move onto Dr. Delgado, if
we may, and ask Mr. Edwards if he would remain so we can address
questions to the group as a whole. Dr. Delgado?

PRESENTATION BY DR. JANE DELGADO
NATIONAL COALITION OF HISPANIC HEALTH AND
HUMAN SERVICES ORGANIZATIONS

DR. DELGADO: Good afternoon. I enjoy being the
luncheon speaker. I am the President and Chief Executive Officer
of COSSMHO, which is the National Coalition of Hispanic Health
and Human Services Organizations. We have been in existence for
about 14 years. I have been President and Chief Executive
Officer for two and a half years.

The organization has 507 members. Our members are
community based organizations, people in positions of power in
health institutions and community health centers, faculty and

243

 

 
 

many researchers. We have about a $2 million budget; 80 percent
of that goes out to the local level. One of our major
activities is research. Before I start talking more about
Hispanics, I just want to get us all on the same understanding
about Hispanics. First of all, most Hispanics, nearly 75
percent, are second generation. That is an important fact to
remember. Second, over 75 percent of Hispanics speak English.
Most Hispanics feel more comfortable in Spanish. They speak
English but feel more comfortable in Spanish. That is why when
we talk about things, we say translation is important, but you
have to understand what you are translating, why and who
developed those materials.

The next thing is people who speak Spanish speak a
common Spanish with regionalism and colloquialisms which makes
some differences when we are talking about things like body
parts, where slang may be very important and may differ by state
or region. The other thing is that unlike the non-Hispanic white
community, we are not as diverse, we are mainly
Mexican/Americans, Cubans and Puerto Ricans, Central and Latin
Americans, while non-Hispanic whites can come from all over
Europe. The paper that I provided to you is very important. It
was called "AIDS, the Impact on Hispanics in Selected States."

We did this paper because one of the major problems we had are
the data which are available on Hispanics. Besides my role with
COSSMHO, I'm on the National Advisory Committee on Health and
Vital Statistics and was also one of the people who helped pull
together the Black and Minority Health Task Force Report. You
will never see a Hispanic talking about the number of excess
deaths because we don't have that information. In fact, the best
data we have on anything is on AIDS. What we have nationally, we
have found real problems with. It just clumped everything
together.

We decided to go back to selected states and get
information as to the categories of transmission. What we were
hearing from CDC was AIDS and Hispanics is IV drug abuse. We
heard it and we were parroting it, too. When we went back to our
data, we found some very interesting things. One of the things
we found was if you look at IV drug abuse for the States of New
Jersey, New York and Puerto Rico, it is very important, but it is
not so important in Florida, Illinois, and Texas. In fact, in
those states, homosexual/bisexual male transmission is much more
important. That means that our community based organizations
needed to know that. They needed to know who they should work
with, what organizations they needed to work with and reach out
to in order to get information out to the community.

It took us a long time to get this paper together.
California is missing. The reason California is missing is
because they would not provide us with information on Hispanics.
We tried very hard to get information. We just got news from

244

 

 
 

 

Colorado that they will be providing us a:run. If you
understand what is going on at the community level, they need to
know as much information as to whom in their community are at
risk and under what categories of transmission.

ADMIRAL WATKINS: Excuse me, Dr. Delgado. Are you
saying California has the information and wouldn't provide it or
they don't get the data at all?

DR. DELGADO: They must get some data because they give
it to CDC. However, they would not provide it to us. It took
months to get it from the other states. I think part of it is
because when you get information, information creates power to
ask for resources, which is what Don was talking about. If you
don't provide the information and you don't know what the
incidence or the transmission categories are, you don't have the
tools you need to get the resources.

If you look at this, you get some real interesting
differences by state. We also have information on Puerto Rico.
You will find this paper useful. We are finding it useful. We
have disseminated this to all our members and every place that we
can. We think that one of the most important things that we can
be doing now on AIDS at the local level are things in the area of
public education, treatment and services. When I got to COSSMHO
in 1985, I started to approach CDC about doing something in AIDS,
and they knew about the problem in AIDS and Hispanics. They may
not have published a paper until 1986, but they knew in advance.
The numbers of blacks and Hispanics have been consistent, 14 and
25 percent, from the very beginning of the AIDS epidemic.

One of the concerns we have is that people will make
forays into the Hispanic community and say they are doing
something in terms of AIDS education, treatment and services,
without working with the existing institutions that are there.
This is a major concern for us because we want people to work
together. We want institutions to come in and work with us but
we also want the institutions who are there to be part of the
AIDS network, which is for example, one of our major activities
in AIDS, that we have a newsletter that goes out to all our
memberships on some AIDS update.

The idea that we worked in AIDS was very unpopular in
the beginning. They didn't know why in earth I had picked this
as a topic to work on. The community was not at a level of
information. One of our first steps was to get a booth at the
National Hispanic Media Association, and give out pamphlets to
all the media on AIDS so they would learn. We got a lot of
negative feedback from that. We also got much more positive
feedback, that people didn't know this was a problem.

245

 

 
 

 

As we do more in the AIDS area, we want to make sure
that we learn from the models of health promotion, disease
prevention, outside in the community which have been useful for
Hispanics. I think this is very important. Some of the models
for non-Hispanic whites will not be as successful with Hispanics.
Our communities don't have the resources. They don't have the
networks. They also may not have the access to health care that
we would like.

I am sorry to say I didn't bring another paper that we
brought which was called "AIDS and the Financing of Care in the
Hispanic Community." This was an important one to us. I heard
you talking about Medicaid before. I was chuckling. One of the
problems we have with Medicaid is close to two-thirds of the
monies from Medicaid goes toward long term care, even as it was a
program intended for poor mothers and children. Keeping that in
mind, as our population has changed and gotten older, Medicaid
has been something which has been more to deal with elderly
problems.

When you look at Hispanic participation in something
like Medicaid, you see it is relatively low. For example,
Hispanics are second only to blacks to the extent of poverty in
the population, yet they are five times more likely not to be
covered by Medicaid. The reasons? If you live in Texas, it is
very hard to qualify for Medicaid. Many Hispanics live in Texas.
It is difficult. We have to look at the financing of health care
for Hispanics once they have AIDS. I have a lot of things to
share with you and I am trying to hit everything. I'm sorry I am
jumping around so much.

The other thing which is important is the idea of
getting our member organizations to work with some of the gay
service organizations. This sounds like it should be an easy
thing since we are all fighting the same disease. In fact, in my
community, what we are fighting is also a lot of denial about
sexual activities. Working through that, you can't do everything
at the same time so we always talk about high risk behavior. We
don't talk about high risk groups. We just say that we as
Hispanics have to know about high risk behavior, what we should
and should not do.

I think it is important to keep in mind also as you
think of your recommendations, three very important things. I
hear a lot of talk about coordination, coordination within HHS,
coordination within HHS and other agencies. Coordination at the
Federal level often means people sit around a table, everybody
says what they are doing, and that's coordination.

To me, that's not coordination. Coordination means we
all sit around together and we parcel out the pieces of what is
going to be done and we are informed. It is more like a

246

 

 
 

 

continuum of care concept. That's not the kind of coordination
we have now in AIDS, at the national level, and consequently it
filters down to the local level.

The second concern is the idea of gatekeepers. Who
will be the gatekeeper for AIDS education, AIDS treatment, AIDS
services? If they are the same people who have always been
gatekeepers, Hispanics have not benefitted from health care in
America to the extent they should as contributors to society.
That's another issue.

The third, which is very important, I always hear,
well, Hispanics are Catholic and they don't want to do anything
with AIDS. If you ever go to El Paso, Texas, cross the border to
Juarez. Juarez has AIDS education going on at a level that El
Paso will not get to for a long time. The reason is in Mexico,
family planning and talking about family planning and the
paraphernalia associated with family planning, has been an
ongoing discussion. For Hispanics living in America, we are also
a product of the national policies which sometimes make this very
difficult. I think those are the broad things that are
important to keep in mind when you are trying to think of
programs that impact on Hispanics, please remember Hispanics are
14 percent of those persons with AIDS. We have data. We aren't
getting any resources. We would really like your support. Thank
you.

MR. CREEDON: Thank you very much, Dr. Delgado. You
know, the main thrust of our meetings yesterday and today is this
question of incidence and prevalence of AIDS. I wonder if you
have focused on whether the CDC reports adequately deal with the
AIDS in the Hispanic community? If not, have you tried to get
cpDc to adequately deal with it?

DR. DELGADO: One of the reasons it took us a long time
to get this report together was because we went to CDC and asked
them to give us data on AIDS in Hispanics by state, and they said
they could not do that. So because they could not do that, we
had to go to different state agencies.

MR. CREEDON: Are they trying to do it now?

DR. DELGADO: Not that I know of. I mean, they said,
well, it's confidentiality issues, that's why we can't give it to
you. I was like, confidentiality? I did not understand it.
There was obviously a gap there, and I don't think the gap was on
my end.

MR. CREEDON: Do you think the gap is being closed
now, or is it still a problem?

 

 
 

 

DR. DELGADO: It's still a problem.. I think part of it
is, you know, for Hispanics, we have never had any good data.
This is the first time. The fact that we have good data and
nothing is happening is quite disconcerting, to say the least.

I think one of the problems that Don mentioned which
for us is very important also is to have that breakout between
gay and bisexual to understand what's going on, because we also
have a lot of denial in our population. For example, in El Paso
-- and I speak a lot about El Paso because we were very active in
getting the Hispanic community to take ownership of the AIDS
problem. The local gay community talks about -- the local
Hispanic gay group does not talk about being gay but about men
who have sex with men. To me, that's a different type of
communication than a community that takes ownership of their
sexuality.

So how we deal with each of our communities is very
different, and, for certain, a Hispanic who's in one part of
Texas will be very different than a Hispanic in San Francisco or
one in New York or Hartford, so we would really go for tailoring
the message. CDC has not been very good at providing us with
specific information.

MR. CREEDON: I think it would be helpful if you would
write to us and tell us in light of the way CDC now reports how
you would like to see their reports changed, so that we would
have a very specific recommendation as to how it could be more
helpful to you.

DR. DELGADO: I will do that. I will also add that on
this national advisory committee which is the advisory to the
Department of Health and Human Services, that is one of the
things that was also discussed, and I will look back on my notes
and see what we said in the national meeting, too.

MR. CREEDON: Thank you very much. Any other
questions?

DR. LEE: Straighten me out on the figures here. Are
the black and the Hispanic rates higher than the white rates,
three times the white rates as in your paper.

DR. DELGADO: Well, it depends what subgroup you're
looking at. For Hispanics, there are 7.9 percent Hispanics in
the population, but 14 percent of the AIDS patients are Hispanic.

DR. LEE: You say in your paper that it's three times

the rate in Hispanics and blacks as it is in white. Is that
accurate?

248

 

 
 

 

DR. DELGADO: That's cumulative, yes, and that's
looking at cumulative incidence.

DR. LEE: And that's the way it is?

DR. DELGADO: Right.

DR. LEE: Now, what do you say -- how do you add up in
the states that don't have a lot of drug addiction like Texas,
Florida? That's not what we found in Florida. According to you,
you don't have much drug addiction in those two.

DR. DELGADO: Right, but if you look at --

DR. LEE: Why is there so much more homosexuality than
in the black and Hispanic? Do you have an explanation for that?

DR. DELGADO: Well, let me go back. First of all, to
be considered gay or bisexual, you only have to have one contact
with a male. Second of all, i think people's sexuality is not
something -~- and sexual choice is a very difficult question to
answer, and I'm not in a position to talk about that, and I think

it's very --

DR. LEE: I guess I didn't mean to say why. What I
meant to say is, is there an increased incidence, then, in those
states of bisexuality and homosexuality minorities versus the
whites?

DR. DELGADO: It would be hard to say. It would be
hard to say. We're talking about transmission categories, and I
think what you would find in those states, also, they do have
less use of IV drug use in those states than you do in a state
like New York or New Jersey. So there are other variables which
compound this thing. It's just a two-way analysis.

DR. LEE: Dr. Edwards, do you have any comments on
that? My analysis of the numbers is probably wrong somewhere,
because I don't see an explanation for it.

MR. EDWARDS: I'm not sure I'm even understanding your
question, Dr. Lee.

DR. LEE: Well, if there are three times as many people
with AIDS proportionately in black and Hispanic populations as
there are in the whites and in some of these areas you have very
little -- some of these states are reporting very little drug
addiction, to me that means an increased incidence
proportionately of homosexuality or bisexuality in the minority
communities. I'm not -- medically, I'm not aware that's the
case. I'm not sure I understand --_

249

 

 
 

MR. EDWARDS: I think I understand what you are asking
-- what do you get from looking at cumulative incidence data? I
think what we're getting is that when you look at the country,
the cases per million in the black and Latino communities are
three times the cases per million in the white community. What
happens is that kind of population-based data really washes out a
lot of the kind of focus data I think you're asking questions
about. What would be more useful in looking at the kind of
incidence per transmission category that I think you're raising
on a state level would be the kind of questions that Jane has
raised that we aren't getting that data from the states. We
really need to look at the percentage of Latinos or blacks or
whatever in the state and then use that as a denominator data for
the kind of -- for what we're looking at in terms of IV drug
users and gay people and other transmission categories. That is
going to vary by state.

I think one of the questions that Jane has also
raised, however, is that prior to AIDS, this kind of data was not
being kept. One of the things we found in New York City where
you probably have more intravenous drug users than anywhere else,
that we're still dealing with estimates of population.

We have no real data, and if we don't have it in New
York, you can imagine what is missing for the rest of the
country. That is the kind of issue that we're really looking at,
that we can't find -- the initial paper that was printed, I think
Ann Hardy from the CDC did it two or three years ago, we have no
good denominator data in looking at the number of black and
Latino men coming out of New York City to project an incidence
rate model for New York City. This is a problem we're facing all
over the country.

MR. CREEDON: We're running about a half hour late, and
if the Commissioners are agreeable, I would like to ask Dr. Levi
or Mr. Levi to talk now, and then we can come back and ask
questions of the entire panel.

PRESENTATION BY MR. JEFF LEVI

MR. LEVI: I want to thank the Commission and its staff
for this opportunity to testify this morning. As you know, the
National Gay & Lesbian Task Force is the nation's oldest and
largest gay and lesbian advocacy organization. Since the
beginning of the AIDS crisis, we have sought a committed,
compassionate public health response to the epidemic from all
levels of government and all levels of society. Before
addressing the subject at hand, I wish to briefly comment on your
interim report recently submitted to the President.

While you reached no conclusion, the report praises the
work of the gay and lesbian community in providing services and

250

 

 
 

 

"4
1

education at a community level is an appropriate and important
recognition by a body appointed by the President. Indeed, the
work of the gay community is deserving of direct presidential
recognition, something that has been sadly lacking. But because
you have not reached your conclusions yet, I offer this word of
caution. Simply because the gay community has done such a
tremendous job in addressing unmet needs in the AIDS crisis, it
does not in any way mean that the job is done in the gay
community or that the task can be continued without government
support, federal, state or local.

The federal government has been particularly negligent
in supporting prevention programs and services for those affected
by AIDS. It has been afraid, I believe, to work with those most
engaged in this effort because those most involved are gay. The
lesbian and gay community's efforts are not a substitute for the
government taking responsibility for a public health problem, a
public health problem that affects all Americans and should
concern all Americans. The gay community cannot bear on its own
indefinitely that which is the entire society's burden.

Over the last two days, you have heard a good deal
about numbers: the number of people who are or might be infected
with HIV; and the number of people who have and who might go on
to develop AIDS or ARC. There has been a good deal of
speculation in the press about whether the Public Health Service
has over- or underestimated the number of people infected. I do
not claim to be an epidemiologist. I will not attempt to enter
this dispute. I would urge you, however, to view your role as
less one of playing arbiter among the epidemiologists and more
one of depoliticizing this debate and focusing on its real
implications.

Why does it matter if there are 600,000 people infected
or 1.5 million? Why does it matter if HIV infection is moving
quickly or slowly into the heterosexual population? We do not
have to have to precise projections to know that AIDS is a
serious public health problem that requires attention in a number
of areas. I would like to make several general points with
regard to seroprevalence data.

First, you must be careful about the use of numbers.
Inflated numbers can be used as a scare tactic. Interpreting
numbers as unexpectedly low can be used to denigrate the
seriousness of this epidemic, justifying less public attention
and resulting in less public funding to cope with this crisis.
Quite frankly, this is what many of us fear is the genesis of the
current debate over seroprevalence.

Second, we must avoid painting constituencies or risk
groups with a broad brush. The level of infection in any group
will vary according to the overall level of infection in the

251

 

 
 

 

geographical area. Behavior changes in the gay community, for
example, vary greatly, depending in part on the level of
prevention activity and the level of infection in the comnunity.
In other words, variance can be related to the degree to which we
have educated persons about their risk and the degree to which,
given the incidence of disease, people perceive themselves to be
at risk.

Third, we must remember that we are dealing with very
soft data. The seroprevalence studies that can give us hard data
have only begun. All of the estimates put forward by the various
epidemiologists are educated guesses at best. It will take some
time until we have a good baseline data regarding infection
levels, and it will take even longer to determine trends in
infection rates. It bears restating that the gay community has
always supported the seroprevalence studies under way. Indeed,
gay volunteers have formed the data base for much of what we are
talking about today.

Fourth, seroprevalence data tells us who was infected
in the past, not who is engaging in behavior that places them at
risk today. Our emphasis, therefore, should not be on risk
groups but on risky behavior. Thus, while a majority of people
infected may indeed be gay or bisexual men and IV drug users, we
do not know, we do not have hard data on where the infections are
occurring today, but we do know who is engaging in behavior that
could place them at greater risk for HIV infection. This
country is experiencing an epidemic of sexually transmitted
diseases among heterosexuals. More than 14 million Americans
contract a sexually transmitted disease every year. Those rates
are skyrocketing nationally among heterosexuals while declining
among homosexuals, though this may vary by community.

If these people are contracting STDs, then they are
engaging in sexual activity that places them at risk to AIDS.
This is disturbing in and of itself. It is made more so by the
fact that there is increasing evidence that STDs may be a co-
factor in development of HIV infection.

One final concern about the emphasis on seroprevalence
data. There is a danger in focusing too much on the disease
model, of identifying the number of people infected and seeing
that as the universe of people affected by this problem. Such an
approach only addresses one side of the equation, that which
needs to be done after infection has occurred; namely, treating,
caring for and hopefully curing those who are infected.

It ignores the other side of the equation. How can we
prevent people millions of people whom we know to be engaging in
risky behavior irrespective of their so-called risk group, how
can we prevent them from becoming infected in the first place?
Our obligation in this crisis is two-fold. We must care for and

252

 

 
 

 

plan for the needs of those who are infected; and we must work
with greater urgency towards effective treatments and a cure.

But our obligation is also to make sure that the majority of
Americans who are uninfected, including the overwhelming majority
of gay men in this country, that they have the tools to remain
uninfected. The seroprevalence studies under way will give us
interesting data; but as your interim report indicated, we are
coping with a crisis that requires immediate attention by the
policymakers in our government.

I would, therefore, urge you to base your work on the
ranges that have been offered to you by the Public Health
Service. No matter which numbers we use, the next administration
will face hundreds of thousands of cases of AIDS. This
Commission must address the needs of these people and how the
existing health system will need to adapt to this new burden. We
must guarantee quality care to all persons with AIDS or HIV
infection. To do that, we must successfully answer a number of
questions. Will we be able to assure that access to quality care
will not be dependent on our ability to pay? Financing care of
those with AIDS poses some of the most complex challenges to our
health care system. How do we help the health care delivery
system from private hospitals to public hospitals prepare for and
more evenly distribute the added burden of AIDS so the quality of
care offered those with other diseases is not diminished?

I think some of the most compelling stories have come
from the public hospitals, who are overwhelmed by AIDS and yet
are also the primary care facility for millions and millions of
poor Americans. How do we solve or reduce the discrepancy in the
quality of care offered the person with AIDS in San Francisco as
opposed to Tulsa or the discrepancy between the gay white man in
New York and the black IV drug user in the same city? This
question is especially compelling as the demographics and
geography of AIDS changes.

Everyone seems to be putting in their plug for a study
that CDC could do. One of the studies that I wish the CDC would
do and have actually asked them to do and have been told it would
be too difficult is to do a comparison of life expectancy between
diagnosis and death by geography, by all the various risk groups,
as well as by race or ethnicity.

I think that will tell us a lot about some of the
delivery of care issues we have. I travel a lot, and one of the
most compelling things that I hear is particularly in the lower
incidence communities where there are physicians who are
desperately trying to care for people with AIDS as best they can,
saying I just don't have the same resources that a person with
AIDS would have or a physician treating a person with AIDS would
have in another city like San Francisco or New York; that I don't
have access to the latest treatments, I don't have access to the

253

 

 
 

latest equipment or diagnostics; I don't have access sometimes to
the same kinds of drugs that I could offer my patients.

That is compounded in lower incidence cities with
health care facilities that are not terribly eager to care for
people with AIDS. Finally, are we providing -- or almost
finally, are we providing enough funds to research better ways to
manage the disease in addition to finding a vaccine and a cure
for the underlying infection? Finally, are we using the
projections to assure that the public sector is assuming its
responsibilities rather than continuing to rely on an already
overburdened voluntary sector? These are frightening issues to
confront at the current level of infection, but we must take
steps to assure that the number of people infected stays
constant rather than increases.

We must return the focus of our educational efforts to
prevention, prevention that helps people change their behavior,
that helps the 14 million Americans contracting sexually
transmitted diseases each year change their behavior, that helps
both heterosexuals and homosexuals and that helps IV drug users
as well. So I end where I began, with an appreciation for the
praise of the work of the gay community contained in your interim
report and with the hope that you will use the lessons of the gay
community's pain and experience to encourage the public sector to
take responsibility for this crisis and to support those methods
that have been proven successful in our community, most
particularly in the area of prevention, prevention that relies on
education and counseling, not testing and punishment, prevention
that is frank, explicit, affirming and nonjudgmental. That kind
of prevention saves lives.

I might add that it also has to occur in the context
of nondiscrimination. People will not come forward to be tested
or counseled or participate in education programs if they fear
that that participation might cost them their jobs, their
insurance, custody of their children or their homes. Finally and
most importantly, we must not allow our discussion of numbers to
make impersonal what is a very personal crisis for millions of
Americans. This disease affects people, not risk groups of
others different from you or me. It's not some abstract dots on
a curve but human beings, members of our families. The urgency
of your task, given the scope of this epidemic, cannot be
overemphasized. The gay and lesbian community urges you and the
nation needs you to act forcefully and courageously to bring
about a speedier, compassionate response to AIDS by the federal
government. Thank you.

MR. CREEDON: Thank you, Mr. Levi. Certainly that 14
million number that you mentioned is a very sobering number and
one we should definitely pay attention to. You said you had asked
the CDC to amend its reports to include some data?

254

 

 
 

 

MR. LEVI: It was not to amend their reports but to
take the data they collect, and it is all on computer down there,
and to do a comparison. It doesn't even have to be a national
comparison.

MR. CREEDON: Not a regular report?
MR. LEVI: It could be a snapshot.
MR. CREEDON: Do you have correspondence?

MR. LEVI: I must confess it was telephone
conversations so I don't have correspondence. I would be happy
to write to the Commission and explain what I had in mind.

MR. CREEDON: I think that would be helpful. One of
the issues that certainly concerns us is the exact issue you
mentioned, and that is what is the timeframe from the time
someone contracts the virus until various stages. While we had
some testimony on that yesterday from CDC, I don't think we have
a really good fix on it and maybe they can't give us a very good
fix on it. .

MR. LEVI: The only study I've seen is the New York
study where they were able to show that people with different
opportunistic infections lived longer, people who are IV drug
users live the shortest period of time. I think that is one
piece of the puzzle. What I am looking at in some respects is
related to access to care, and that is a secondary issue. In
fact, I'm not sure the New York study answers all those questions
either. Is the IV drug user living a shorter period of time
because the disease is more variant in his case or is it because
that person isn't getting access to the right kind of care.

MR. CREEDON: That is exactly what we are interested
in, in knowing what are the care needs going to be, given a
certain incidence of the virus and so forth, and appreciating
your comment that we are not just dealing with numbers, we are
dealing with human beings.

DR. DELGADO: When you have the care needs, I think cDC
just wants to count who is out there while the HRSA wants to have
more information on the care needs. HRSA is looking at care
needs and CDC is counting. You don't have the cross
fertilization you need.

MR. CREEDON: Maybe we can encourage some of that cross
fertilization. We would appreciate recommendations as to how,
what CDC is doing can be made more effective. Any further
questions of Mr. Levi or of the panel?

255

 

 
 

MS. GEBBIE: ‘I. think all of you have mentioned
additional studies. We were just talking about one that Jeff
Levi was asking for. I for one would appreciate in written form
either the outline or the subject matter of any additional
studies that each of you think would be most pertinent in
understanding this epidemic, either generally or in regard to the
specific groups you are most familiar with and who you represent.

What I also heard were points about attitude, concerns
that groups were being inappropriately blamed or inappropriately
targeted and the data might be manipulated to make certain points
or to ignore certain points. Can any one of you or each one of
you identify easily the thing or things that this Commission
could do to start attacking that attitudinal problem, that might
relieve some of the attitudinal barriers so that we could get on
with the more factual discussions? Is that something you could
put in a capsule for us? .,

DR. DELGADO: I think the major thing is focus on high
risk behavior. If you would do that, talk about high risk
behavior in terms of AIDS prevention, that would go a long way.
That would be my comments in a capsule.

MS. GEBBIE: Instead of groups?
DR. DELGADO: Yes.

MR. LEVI: If I could pick one thing that is disturbing
perhaps is the publication of CDC's surveillance statistics.
There is a hierarchy of groups. It is homosexual and bisexual.
It is intravenous drug user and it is heterosexual and other.

I'm not sure we know necessarily which is the predominant mode of
transmission and it works in such a way that if you are
homosexual and an IV drug user, you are put in the homosexual
category. While CDC does provide breakdowns later on in their
reports of how people overlap, I'm not sure we know which is
necessarily the more efficient route and what it tends to do is
create some confusion as to exactly what is going on out there.

DR. CRENSHAW: Dr. Delgado, in relation to the
obstacles you have been confronting, getting information out of
cbc, I would like you not to feel like a minority. As a
Commission member, I have run into many similar experiences. It
has been six months and with relatively simple requests put in
both words and in writing, I have gotten not a single response.
I brought this up to Polly and Polly indicated that the
Commission staff is having the same problem.

I am going to call on the Admiral, although we have put
this in writing and it has been a month since those questions
were submitted and there is no response, that with the urgency of
this issue and the AIDS epidemic, we need somehow to draw on your

256

 

 
 

 

expertise in removing obstacles, to eliminate the problems in
getting quick access to information, not just for Commission
members, but for special interest groups that desperately need
that information, as long as it is not classified and not
associated with names or identifiers.

ADMIRAL WATKINS: I will bring it out of the woodwork
and we will take a look at it.

MR. LEVI: You never know what is classified. The size
of the fuel tank in the President's limousine is classified.

DR. CRENSHAW: That wasn't a letter to you, Admiral.
That was a letter Polly submitted to CDC on behalf of the

Commission.

DR. PRIMM: I wanted to ask Mr. Edwards about the
funding of the National Minority AIDS Council. Do you have
adequate funds to function? It comes to my attention that your
office functions on very meager funds, that you have made
requests that have not been honored. You are the only national
organization dealing with this that represents all minorities.

MR. EDWARDS: Dr. Primm, I think you have been one of
the people who have been advocates for NMAC. No, we don't have
that kind of support and funding.

DR. PRIMM: What is your support at the moment?

MR. EDWARDS: Our support comes from community based
organizations. In fiscal year 1988, the expectation of working
with the information campaigns in the Centers for Disease Control
and we have in fact submitted requests for support in the private
sector as well as other sectors of HHS.

We think that the recognition of the kind of work we
are doing along side Jane and the National Urban Coalition,
Dorothy Height, organizations across the country, it is being
recognized as key and important. We don't claim the center
stage. We claim simply to try to network and bring more people
together. I would be more than happy in response to your
initial request to submit that kind of laundry list to the
Commission. We would appreciate any support and recommendations
of sources of support. I think that right now people across this
country, black, Latino, Asian, Native Americans, are very much
interested. You have their attention. AIDS has gotten their
attention. We are all looking to see if we are going to be
allowed to in fact use what we have gained and put it into some
kind of action. Your support would really be very helpful. In
answer to your question about what level of support we have right
now, we are volunteer.

257

 

 
 

 

ADMIRAL WATKINS: Is there a need to find a menu of
co-factors that link with the epidemiological data? You were
talking about that, Mr. Edwards. I think Dr. Delgado referred to
it. Isn't there a better way? As long as we are getting the
information, isn't there a better way to link it to the other
kinds of analogues that might give us a better handle on the
broader issues, environmental impact and so forth, on these
matters?

It seems to me that can be defined and it need not
necessarily come from the Federal level, in the same degree it
might be at the local level, which should vary by region, by city
and so forth. It seems to me that here is an area of
public/private relationships between community based
organizations like yourselves, and local health officials, such
as Dr. Joseph from New York, that could work out a regime when
presenting local epidemiological data, to relate to some of those
co-factors which are very significant, where you can begin to
make an impact on the current ratios. It could also impact one
poverty ratios, health delivery ratios, and a lot of other
things.

. , Is it not possible, in any area you want to pick, as an
example of what might be done at the same time, to begin to do
the research on what is really driving those kinds of ratios that
are out of line with the demographic data.

DR. DELGADO: When we did the Black and Minority Health
Task Force report, and I was the Secretary's person on that
Committee and served on the Steering Committee, one of the things
which was important is we had all this data and we knew what was
happening and what the problems were but nothing happened after
that. Very little happened. An Office of Minority Health was
established to try to get some funding to local community based
organizations. There is a good handle on that. Although we can
wait to get more information, we have enough information now to
get going on things. I think that is the problem we have. Given
what we know, we haven't moved as effectively. I think that is
what Jeff was talking about also. There are facts we know. We
do have some numbers. We have to move on. My own organization
has since 1985 only $200,000 to do any AIDS work. Most of our
work has been on a volunteer basis also. The funding isn't
there. Yesterday I was at the Ford Foundation talking about AIDS
and what they should be doing. In fact, I was hoping one of the
big private, public areas of cooperation would be with the
foundations. As you know, with the stock market situation,
foundations have their monies in stocks. That will affect those
foundations which want to have an impact on new areas, if they
haven't done anything on AIDS. I think there are a lot of things
going on and we are looking to the Federal Government to pick up
on that.

258

 

 
 

 

MR. LEVI: Just as a follow-up, I think it is also
important to note that this is the first time the Ford
Foundation is considering getting into the AIDS issue. The
traditional health foundation money that is out there has only
begun to look to enter the AIDS area. Here we are six and seven
years into the epidemic. I think that has a lot to do with the
stigmatization that is associated with this disease and one
hopes this Commission can help overcome.

DR. DELGADO: There is a report by Michael Seltzer
which details everything the foundations are doing for AIDS,
which might be an important piece for the Commissioners to look

at.

ADMIRAL WATKINS: I was impressed by the 1985 report of
HHS on black and other minority health projections. It was an
inspiration to me in the whole field of American youth and
health. Those reports gather dust. If you don't present data on
a routine basis that does some linking, it seems to me that you
waste opportunities for the visual continuing identification of
those things that impact, particularly where the linkages are so
._@irect. I am just saying isn't there an opportunity here to keep
_this more dynamic and exposed than the one time reports that were
excellent but --

DR. DELGADO: That report was very special for two
reasons. First, as a result, an office was set up. There are
nine staff people there. There is legislation to get that office
funded to do AIDS work. Additionally, I think they have 13
states that have started black and minority health task forces to
actually implement those findings at the state level. The very
design of that report was actually to implement, but it is very
hard. Unfortunately, most of the key people in the Department
who worked on that report are gone now.

ADMIRAL WATKINS: This is one of my problems with it.
It comes and goes. We had legislation in 1979 to establish an
Office of Health Education in the Department of Education. It
was never funded. Now it is coming up again. What I am saying
is just because it is there, it doesn't mean it doesn't need a
jab and some impetus.

DR. DELGADO: It needs a jab.

ADMIRAL WATKINS: If you could make a recommendation
along with the others, along those lines, it may be we will want
to review where that is. It seems to me that may be a embryo of
an opportunity here with the additional AIDS impetus to move that
more aggressively, which is a broader issue, but very germane to
what we are talking about.

259

 

 
 

DR. DELGADO: Senator Specter has a bill to that effect
that all of us have been working on.

DR. PRIMM: Mr. Chairman, I want to add onto that.
That office is not appreciated at all as it should be. First of
all, the amount of monies allocated to take care of the health
problems of minorities in this country was something like $3
million for blacks, hispanics.

ADMIRAL WATKINS: This is why I am raising this.
DR. PRIMM: It is an opportunity.

ADMIRAL WATKINS: There is an opportunity here now to
move that much more aggressively. An earlier witness, Dr.
Landesman, from New York, made a recommendation that there has to
be greater coordination between community based organizations and
the formal health care organizations. I don't want to take too
much more time now because we are running behind. We would also
like to hear maybe from the three of you, and maybe you can do
this through the Leadership Coalition on AIDS or some other
context, where you would give us what you would recommend there,
specifically. Give us a template, a concept, of what you might
see in a variety of regional areas, that might cut across a
spectrum of needs out there, that would set up such a cooperative
effort without heavy bureaucracy, that might achieve some things.
This would help us at the national level as we grope with what
legacy we would like to see the President leave in terms of the
institutional process changes that might be necessary to keep
this in the forefront of thinking. One last question. Do local
health officials reach out to your community based organization
to help with the database, with the epidemiology? Do you see a
willingness and excitement on their part to get with you and to
coordinate?

MR. EDWARDS: I would like to talk about that, Admiral.
I think that is a mixed bag. We see a lot of work that needs to
be done in that area. I think there frankly is a kind of
institutional jealousy that we see operating, when you look at
community based organizations, advocating for specific kinds of
funding and having the track record to show that they are doing
things that public health departments have not been able to
accomplish, either because of lack of funding in their own
metropolitan areas or because of lack of interest. We have seen
in some metropolitan areas, more of that kind of cooperation, but
across the country, in the southern part of the country and the
heartland of America, we still have considerable gulfs to bridge
between community based organizations.

ADMIRAL WATKINS: Would you give us some
recommendations where you see the best in the country, for
example, is there something going on in the State of Colorado,

260

 

 
 

 

anywhere, where you feel it might be a model in a variety of
areas in the country where you think there is exemplary work
between community based and local health officials, and even at
the state level. We have seen some of that in some counties,
where local leadership is so strong that it has pulled it
together. The question is do you have any examples of where it

is run right?

MR. EDWARDS: I would say there are a few, in places
like San Francisco, which I think is still providing considerable
leadership, even in its outreach and support for minority
community based activities. Washington, D.C. is an outstanding
example. I think we would really have to stretch, however. We
see the beginnings of that kind of support in Chicago, Atlanta.

Y am not aware of a lot of other cities, at least speaking for
the communities I am in touch with.

DR. DELGADO: For the Hispanic communities, I would say
it would be very hard, as I don't want to repeat, but it is
important, Hispanics have not been traditionally part of the
usual health care system so they have not benefitted from that.

I believe about one year ago, the black and hispanic community in
Los Angeles sued the county health department for lack of
services with respect to AIDS. There are some problems but we
will look for some good models.

MR. LEVI: I think there is another problem and I'm not
quite sure how one addresses it except I think there has to be
almost a Federal mandate for this to happen. The communities
where the linkage between community based groups, whether they
are black, Hispanic or gay, where they are needed most, they are
sometimes least likely to occur because of the political climate.
There are very conservative states where a health department may
consider it political suicide to be giving money to a comnunity
based organization that is in some way gay identified, yet it is
the only organization in that jurisdiction that is doing the
prevention work and providing the services that people with AIDS
or people with risk for AIDS need.

One of the ways we have been able to work around that
has been the fact that there has been Federal money given to the
states to pass onto community based organizations. One of the
concerns that I have is that the Administration is now
considering blocking that money. In other words, the money that
goes to a state for community based prevention or surveillance or
a vast array of services that are provided that right now are
delineated and divided, will just go as a block to the state for
the state to make a decision as it wishes. My concern with that
approach is that in states like California and New York where the
gay community is fairly organized, where the AIDS community is
fairly organized, that money will probably be divided as it ought
to be divided. My fear is that in more conservative communities,

261

 

 
 

 

there will be a tremendous hesitancy to put the money where we
will get the most bang for the buck.

MR. CREEDON: I think we should close it out. We are
running about an hour late. Thank you very much. We look
forward to receiving the additional information that you agreed
to supply. Our next and final panel is a group of four experts
on statistical and mathematical modeling. What I would like to
suggest, if the Commissioners are willing, is that we ask each of
the panelists to make his presentation in serial order and
perhaps commenting on the comments of the prior panelists and
then we will reserve our questions until all four of the
panelists have made their presentations. Maybe all four could
come up. Mr. Michael Cowell, who is from UNUM Life Insurance
Company. Dr. Victor DeGruttola from Harvard University,
Department of Biostatistics. Mr. Anthony Pascal, RAND
Corporation, and Mr. Anthony Robertson, from Research Testing and
Development Corporation. Our first witness will be Mr. Michael
Cowell from UNUM Life Insurance Company. He is not here. We
will move to our second panelist, if we may, Dr. Victor
DeGruttola from Harvard University. Dr. DeGruttola?

PANEL 5 - STATISTICIANS/MATHEMATICAL MODELING
PRESENTATION BY DR. VICTOR DeGRUTTOLA
HARVARD UNIVERSITY, DEPARTMENT OF BIOSTATISTICS.

DR. DeGRUTTOLA: I'm going to speak a little bit about
the uses and the limitations of mathematical modeling in trying
to estimate prevalence of HIV infection in the United States. As
I'm sure everyone realizes, there are a number of different ways
of estimating prevalence. One is to try to estimate the number
of people who are at risk from different kinds of behaviors and
then take samples from those groups, estimate prevalence within
those samples, and then apply it to the total size of the
population.

The problem, of course, is that we don't know what the
sizes of those populations at risk are, and we don't know how to
sample from them. There are other approaches that make use of
mathematical models to try to estimate what the prevalence of
infection is, knowing something about the incidence of AIDS in
the United States from surveillance data and something about the
latency between the infection and the onset of AIDS.

A very simple example of this that probably most people
are aware of is when people try to estimate the ratio of the
prevalence of infection to incidence of AIDS for certain
well-studied groups. Probably most people have heard the ratio
of 30 to 1 or 50 to 1 cases of HIV infection to cases of AIDS
that have been used as a sort of back-of-envelope estimate of the
number of cases of infection in the United States.

262

 

 
 

 

What I'm going to do is just show how that idea can be
made a little bit more sophisticated using some simple
mathematics to try to get a better sense of what the ratio of the
prevalence of infection to the incidence of AIDS is. Of course,
if you knew that ratio, if you knew it was 10 to 1, for example,
simply by looking at the number of cases of AIDS, you would have
a good estimate of the prevalence of infection. In this
modeling exercise, there are basically three different
distributions that we have to worry about. The first one is the
distribution of times of HIV infection or simply the curve of
infection.

The second one is the distribution of times between
infection and onset of disease, and we know that that time can
either be very short or within a year for a very small proportion
of people, or it can be quite long. There's a whole range.
Finally, the third distribution we need to worry about is the
distribution of times of AIDS cases, and, in fact, for people who
like mathematical symbols, you can express the distribution of
incidence of AIDS as what's called a convolution of the other two
distributions, but I won't get into the technical details of
that. I'11 simply mention that it is possible mathematically to
work backwards from the incidence of AIDS to the prevalence of
infection, if you know something about the distribution of the
latency periods.

The latency period can be estimated with reasonable
accuracy from one well-studied population in San Francisco. It's
a population of men who were enrolled to study hepatitis vaccine,
and because bloods were stored starting in the late 1970s, there
is a fairly long period of follow-up of about seven years. And
from those data, people have estimated that the risk of AIDS
after seven years is about 30 percent. But you can estimate the
whole distribution up to seven years. Beyond that, you have to
extend it using some mathematical extrapolation technique.

But although the distribution of the latency period,
the time between infection and AIDS, is relatively well
characterized, the distribution of times of HIV infection is not.
The basic issue might be, for example, was the rate of infection
of HIV fairly constant since the start of the epidemic, or did it
increase at a much more rapid rate, say closer to an exponential
rate, like a chain reaction. And we really can't say. But it
would be very useful to know that, because if we had a better
sense of how the infection grew, we could have a much better
sense of what the prevalence of infection was right now.

On this graph, what I show are three different curves
for the times of HIV infection. All of these curves match, when
combined with information on the latency distribution, match the
observed data on incidence of AIDS equally well. In other words,
each of these curves is equally possible if you take a latency

263

 

 
 

 

distribution which was typical of the group in San Francisco and
you look at incidence of AIDS. But you can see that what they
would tell you about the current estimate of infection is
radically different. If you believed that the rate of HIV
infection increased very rapidly over the past seven years,
something close to an exponential increase early on, like a
chain- reaction type of increase, you would get a prevalence that
was very high, over two million.

If on the other hand you believed that the rate of HIV
infection grew at a fairly constant rate -~ in other words, there
was a fairly constant rate of people getting infected each year
-- you would expect to see a much lower prevalence of infection.
Basically if you look at these three different curves, one of
them has a linear rate, a constant rate of growth of infection.
One is close to the exponential early on; it shoots up rapidly.
And the third one is sort of in between. You can see that you
can estimate, depending on which curve you believed, you could
estimate anything from about 400,000 to over two million people
currently infected.

So if you also were interested in the ratio of the
prevalence of HIV infection to the incidence of AIDS -- in other
words, you wanted to have some ratio to multiply times the
incidence of AIDS to estimate prevalence of infection -- that
ratio, depending upon which of these three curves you believed,
could be anything from 12 to 1 to about 70 to 1. What that
means is that you can't expect simply to have someone come up
with a ratio that is useful for all groups and for all different
points in time, but if you think you know something about the
dynamics by which the infection was spread, you might be able to
do fairly well.

Simply by looking at incidence of AIDS and the latency
period, we can't reject any of these three curves. But we do
have other information that might tend to indicate that this
curve of very rapid increase, particularly in the last couple
years, is not valid. And the reason is, if you look at cohorts
of men who are at risk through homosexual behavior, the rates of
infection, instead of this dramatic increase from '84 to 1987,
appear to have declined. Of course, the problem with
interpreting those data is they only apply to cohorts that have
been enrolled in studies and presumably are well-motivated to
participate in the study, and they may not be typical of all men
who are at risk for homosexual activity. The other major
problem, of course, is that we know very little about young
people, about adolescents. We don't know anything about
incidence of AIDS from that group, because in general it takes a
long time before they develop AIDS, and so we wouldn't expect to
see it until they were in their mid to late twenties, or not to
see very much of it. And the other thing is that adolescents are
not enrolled in cohort studies.

264

 

 
 

 

There is some information from Army recruit data
suggesting that adolescents in the late teen years are getting
infected at rates that are increasing over time, not increasing
as rapidly as this exponential increase, but nonetheless at a
fairly high rate. So the bottom line is what can -- how can we
use additional information to try and get a better estimate of
prevalence besides just saying that somewhere -- probably
somewhere between a half million and two million. One thing you
can do is look at a well-studied population like the gay
community in San Francisco. The number of cases of AIDS in San
Francisco is about 4100 -- it's a little over four thousand -- I
should say cases reported to the CDC [sic].

We can get some evidence -- some estimate of what the
prevalence of infection is in the city of San Francisco for two
reasons. One is demographic studies have been done to estimate
what is the size of the population of men who have homosexual
behavior, and that estimate was about 60,000. Of course, it's
going to be a rough estimate, but you know it's definitely below
100,000 and certainly more than 20 or 30,000. So roughly 60,000.

There's also a population-based study, survey, of
infection in San Francisco that indicated that about half of men
who reported having homosexual contact were infected. So that
you have a fairly good estimate of the prevalence of infection in
San Francisco as 30,000, and the number of cases of AIDS is about
4000. That leads you to a ratio of prevalence of infection to
incidence of AIDS at 7.5, which is lower than anything we saw on
the curve before. The reason for that is that the epidemic is
much more, in a sense, mature in San Francisco. It started
earlier. It appears to have started there earlier than anywhere
in the United States, but New York City, and because the
infection started earlier on, the ratio is lower. The ratio of
prevalence of infection to incidence of AIDS decreases over time,
because the risk of AIDS increases.

Now even though San Francisco is not going to be
typical of all cities in the United States -- in fact, it's quite
different -- still it tends to indicate that unless the epidemic
process were radically different in other cities of the United
States, you would not expect to see a ratio that was ten times
greater in other cities. Ratios of 70 to 1, the kinds of ratios
we would have to see in order to get a prevalence of HIV
infection over two million, rates of 70 to 1 just don't seem to
be very plausible. So what can we say? Well, the prevalence of
HIV infection from homosexual contact in the United States is, I
think, likely to be less than a ratio of 20 prevalence to
incidence times the number of cases of AIDS from homosexual
contact are probably less than a half million. Now although the
other slides had some technical justification, this is more
subjective. It's impossible to estimate precisely that ratio,

265

 

 
 

 

unless you knew a lot more about the dynamics of the epidemic
than we do at this time and a lot more about how different cities
vary. But since we do know that it's unlikely that the rate of
growth of infection was exponential in the past three years,
exponentially increasing - - if anything, it seems to be
decreasing -- and since we know that ratio in San Francisco is
probably about 7.5, 20 seems to me to be a reasonable upper bound
for what the ratio could be, and that leads to an estimate of HIV
infection from homosexual contact of less than a half million.

I should point out, another caveat is that since we
know very little about what is happening with young people, the
men who are enrolled in these studies that show decreasing rates
of infection are usually older, they're not adolescents, we can't
say anything with any precision, and that's why I emphasize this
is a guess; it's not a formal estimate.

Now what about the remainder of the epidemic?
Homosexual contact appears to account for about two-thirds of the
cases of AIDS in the United States roughly. If we thought that
the ratio of people infected to cases of AIDS was the same for
people infected through homosexual contact as for other kinds of
contact, then we would be done. We would say, well, just add a
half to this, and it would be about three-quarters of a million.

The problem is that those other epidemics may be very
different. The remainder of the epidemic partly refers to people
infected through transfusion or through contaminated blood
factor, and those rates have dropped off sharply since 1985. On
the other hand, it's not clear what's going on with IV drug
users, if the rates are still continuing to increase rapidly. We
also don't know much about the heterosexual epidemic. The spread
which appears mostly now to be from IV drug users to
heterosexuals, mostly women, has not been well characterized, and
we really -- it would be very difficult for us to estimate what a
ratio of prevalence of infection to incidence of AIDS is for
women infected through heterosexual contact.

But still they probably constitute even now a
relatively small proportion, let's say less than 10 percent of
all the people infected. So that I would guess once again that
the ratio of prevalence of infection to incidence of AIDS for the
remainder of the epidemic is probably less than 40. That would
lead me to guess that prevalence of infection due to causes other
than homosexual contact is also probably less than half a
million. So my guess would be that the total prevalence is less
than a million in the United States. It could be considerably
less than that, depending on what behaviors have been for IV drug
users and their heterosexual partners in the past three years, or
it could be more than that if there has been a lot of spread that
we have not had an opportunity to see yet among young IV drug
users and their heterosexual contacts. And that's one reason

266

 

 
 

 

 

that I think it would be very appropriate to try to start doing
studies to estimate what are the rates of infection for
adolescents and young adults in urban areas.

Finally, if we want to get some idea of where the
epidemic -- first of all, what is the prevalence right now of
infection due to heterosexual contact, one of the largest
unknowns, and what can we expect from the future, I did a very
simple modeling exercise, mostly for the purpose of finding out
what would we need to know before we could say anything
definitive about the heterosexual epidemic. There are too many
unknowns right now. And for this exercise I just chose numbers
pretty much out of my head, chose numbers that there were about
10 million heterosexually active individuals in major cities who
had at least some possibility of contacts, either direct or
indirect, with IV drug users and about 250,000 IV drug users in
major cities. So I had 10 million heterosexuals and 250,000 IV
drug users mixing pretty much randomly in urban areas.

This is much too simple to describe a population as
complex as the United States. The reason I did this, I emphasize
again, is just to get a’sense of what we would need to know
before we could say anything definitive about the heterosexual
epidemic. Features that I included in the model were
differences in efficiency of transmission male-to-female and
female-to-male, allowed for the fact that the latency between
infection and AIDS, you have to consider a whole distribution,
not a fixed time, that there's a high proportion of heterosexuals
infected by people with other risk factors, and also that there
may be delays in infectivity. People may not become infectious
right away. And what I showed even from this very simple model
was that it would be impossible to say whether or not a self-
sustaining heterosexual epidemic could occur in the United
States as of now. You simply can't give a definitive answer, yes
and no, even if you did pin down the sizes of the populations,
which of course we don't have, and the way that they mixed.

A typical kind of outcome of these models -- and I
should say that there's a huge range of possibilities that are
all consistent with the available data, so I just selected this
as simply being representative -- it allows -- what these models
allow you to do is, within this exercise, to plot what the
prevalence of infection would be for IV drug users and for
individuals infected through heterosexual contact. The solid
line in this top curve is for IV drug users, the dotted line for
heterosexuals, and the top curve here refers to men. What this
particular modeling exercised showed -- and I think it would be
typical of any model that you would fit right now -- is that for
men in the United States right now, there is a much higher
prevalence resulting from IV drug use than from heterosexual
contact. Even if you think that heterosexual contact was
efficient enough, so that the epidemic could be propagated, the

267

 
 

 

proportion of men infected through heterosexual contact might be
quite trivial. But for women, that's not true. The bottom curve
here refers to women, the solid line once again IV drug users,
the dotted line heterosexual women who don't use IV drugs. At
this point in time, according to this model, you would get
roughly equal numbers of women infected through IV drug use and
heterosexual contact, even though your incidence of AIDS is
greater for women infected through IV drug use, the heterosexual
epidemic having occurred more recently. But as time goes on,
there is the potential for considerably more spread.

And I would emphasize once again that this is just a
modeling exercise designed to give some sense of what are
plausible ranges for what could happen in the future and to give
us a stronger sense of what we need to know before we can say
anything more definitive. And one thing that grew out of that
exercise was to show that we need much more precise information
about infectivity, particularly a delay between infection and
infectivity.

We know that there have been many people infected
through IV drug use, even going back to the 1980s, and we've seen
relatively few cases of AIDS attributed to heterosexual contact.
There are about 500 among American-born heterosexuals who are
believed to be infected through heterosexual contact, maybe a bit
higher right now. But the big question is, is the reason for
that that heterosexual contact is relatively inefficient, or is
it because there's a long delay between the time at which an IV
drug users gets infected and the time in which he or she becomes
infectious through sexual contact? And before we know much more
about that, there's no possibility of saying anything definitive
about the potential for heterosexual spread.

Had we a better sense of heterosexual spread right. now,
I think we could pin down our prevalence figures much better,
because we would be able to say that we didn't believe that
heterosexual spread was going to be contributing a great deal to
the epidemic, either now or in the future, we could be much more
precise about the rest of the epidemic. With that as uncertain
as it is right now, it makes it much more difficult to estimate
prevalence. Thank you.

MR. CREEDON: Thank you very much, Dr. DeGruttola. I
understand that Mr. Cowell is here now.

{[ Pause. ]

Welcome, Mr. Cowell. You are free to proceed whenever
you are ready.

268

 

 
 

 

 

PRESENTATION BY MR. MICHAEL J. COWELL

MR. COWELL: I apologize for getting here late. I was
wandering around this building for about 15 minutes, looking for
an entrance, and then they sent me over to the other side of C
Street, and I wandered around over there for about ten minutes.

MR. CREEDON: The actuaries from Metropolitan had the
same problem.

MR. COWELL: The problem is we follow directions.

(Laughter. ]

MR. COWELL: When you're spending your whole year
figuring out numbers in this epidemic, it's a wonder you even can
see straight, let alone find Washington, D.C. Mr. Chairman, as
you know I'm Michael Cowell. I'm Vice President and Corporate
Actuary of UNUM Life Insurance Company in Portland, Maine. I'm
on the AIDS Task Forces of both the Society of Actuaries and the
American Council of Life Insurance, and I'm the co-author of the
report that you have before you or will have once they get
distributed, AIDS, HIV Mortality and Life Insurance. This is a
report that I co-authored with Walter Hoskins at the request of
the Society of Actuaries and the insurance industry to get an
estimate of the impact of the epidemic for the insurance industry
in general. In order to do this, of course, we had to study the
impact of the spread in the general population, and we followed
some of the same methodologies that I'm sure you've been hearing
about over the last couple of days. I won't go into all of the
basics about the nature of the epidemic because I'm sure you've
heard those several times over.

From an insurance company perspective ~- and this may
be not directly related to what you are concerned about this
afternoon but it points to the reason that we had to study this
in the detail that we did -- studying the mortality of any new
underwriting class when you are dealing with impairments, and we
viewed HIV infection as another kind of impairment, the normal
process, as you know, is to underwrite enough cases to follow
your claims experience and to see how it compares to your
standard mortality or morbidity. However, it is unlikely that
any life or health insurance underwriter would accept insurance
applicants with the HIV virus level on full clinical AIDS, so the
insurance industry realizes that it's not likely to have or even
to want to have the opportunity to study this disease in its
Classical mode.

Life insurance companies are seeing death claims on
business that's been underwritten for many years, long before any
of us even knew about AIDS, on people who suspected they had the
virus and others who probably knew nothing about it. As you

269

 
 

know, we know something about the disease in the general
population; obviously, not as much as we'd like, but we certainly
know a lot about the deaths, and we know something about the
number of AIDS cases and far too little, as the previous speaker
was obviously indicating, about the population that was infected.

Mr. Hoskins and I more or less backed into this
problem. We do recognize here that the disease that we are
addressing, the title of our report notwithstanding, is really
HIV infection. AIDS is simply the penultimate stage. Of course,
once the disease has progressed to full clinical AIDS, life
expectancy is similar to that of terminal or cancer patients.

The first slide here shows the survival of a cohort of
patients from the time the disease is diagnosed as full clinical
AIDS. This is based on the over 45,000 AIDS cases reported to
CDC by mid-November. I understand that number is now close to
46,000, of which almost 26,000 have died. What I've shown on
here on the red-line is the actual CDC data by six months
survival cohort, and the green line is our mortality model, which
is based on 45 percent mortality for the first two years, 35
percent in the third year, and 25 percent annually thereafter.

From onset of diagnosis as full clinical AIDS, this
produces a complete life expectancy of about 26 months. Fewer
than 10 percent of full clinical AIDS patients survive five
years, and at the time we collected our data there were no
reports of any survivors beyond nine years. So much, then, for
the end stage of the disease. We then step back to what I call
the middle stage of our report, which is a study of the progress
of the infection from initial seroconversion through the various
stages to full clinical AIDS.

We relied on two reports, both of which I'm sure you
are very familiar with. The report about a year ago from the
Center for Internal Medicine of the University of Frankfurt in
West Germany analyzed the progression of this disease through its
more serious stages to AIDS and death, and it provided a
particularly valuable link that had been previously missing from
the literature. The Frankfurt study used the Walter Reed
staging method that was developed right here in Washington, D.C.
at the Walter Reed Army Institute showing the progression from
initial otherwise apparently healthy seropositivity through each
of the four subsequent stages to AIDS and then death.

The heart of our study that Walter Hoskins and I
developed, really, I guess what set it off from most of the
other studies that have been done in the insurance industry was
we constructed a simple Markov chain model which combined the
results of the progression from the Frankfurt study with the
mortality just discussed from the CDC data. We developed a
model that simulates the progression of the newly-infected HIV

270

 

 
 

 

cohort through successive stages to AIDS and death. This is a
visualization of that model. The initial cohort shown at sero
years in turquoise or aqua is'a group of 100 or 100 percent of
the cohort that's HIV positive, otherwise asymptomatic, and it
shows how that population would progress through seven and a half
years based largely on San Francisco City Clinic and West German
data, and then we projected that 15 years. We actually went all
the way to 25 years, but I don't place too much credence on the
numbers beyond 10 to 12 years, maybe 15 years. I'll get back in
a moment on the reliability of these data.

The implications, the life insurance which, of course,
was the principal purpose of our study, was extra mortality from
AIDS produces a life expectancy of about 11 years for a 35 year
old HIV infected cohort of males. We compared this to, or
perhaps I should more properly say contrasted this with, the 43
year life expectancy for a healthy male the same age.

We concluded that infection with the virus had the
mortality equivalent of moving somebody 40 years along the
mortality scale, and in terms that you, Mr. Chairman, would
understand from your life insurance background, you'd have to
increase standard mortality more than 5,000 percent to produce
the same reduction in life expectancy. I don't like to describe
AIDS mortality in standard experience because, as I will
subsequently demonstrate, the mortality patterns bear no
resemblance to those in which most companies write insurance
business.

Now, we didn't rely solely on the West German study.
There were two supporting studies, one by the National Cancer
Institute, again right here in D.C., developed by Dr. Goedert,
and I believe Dr. Goedert has addressed many of these
conferences. This showed a three-year incidence of AIDS in four
populations in the United States and one in Denmark of about 725
people.

The third study, of course, is the famous CDC San
Francisco City Clinic study conducted by Dr. Volberding and going
back all the way to 1978 on the 6,700 male homosexual cohorts
from that volunteer group. It was from that that we developed
estimates of the progression from initial HIV infection otherwise
asymptomatic through the various stages to full clinical AIDS. I
have shown here, New York City is in the dark red, the CDC in
orange, the Danish yellow, the D.C. study is -- I guess the D.C.
and New York City colors are sort of comparable. The New York
City one is the first line that goes up to a little over 30
percent in three years. That was the Manhattan homosexual group.

Now, the Frankfurt study, so labeled, is the model that
we developed from the Frankfurt data by chaining together
successive periods of three-year progression rates. The

271

 

 
 

 

progression patterns admittedly are not identical. They were
based on different clinical approaches, nonuniformity, a
diversity of subjects on two different continents, three
different countries. However, we felt the results were fairly
consistent. When we moved the axes over a little bit to allow
maybe for some delays in reporting, I think we're able to
demonstrate that the progression rates were remarkably similar.
The latest data, of course, from the San Francisco City Clinic is
that approximately 45 percent of the initial cohort that has been
under study nine years and has now progressed to full clinical
AIDS.

This progression from HIV infection produces estimates
of survival that were certainly reliable for insurance purposes.
They predict the patent mortality of a cohort of HIV infected
subjects with reasonable certainty. This information is helpful
for insurance companies for determining the financial
consequences of not screening out a group of HIV infecteds when
they apply for insurance.

We readily concluded that anybody that was HIV
infected, that tested HIV positive with two ELISAs and a Western
blot, which we believe is 99.99, almost .999, percent certain of
infection was simply not acceptable as individually insurable
risks. So much, then, for the middle stage, the progression.

The next quest, of course, was to determine the extent
of the disease in the general and insured populations which
causes the principal purpose of your Commission's investigation.
I included this slide. It's simply to show the difficulty of the
problem with only 26,000 deaths and about 18,000 alive with AIDS.
We are seeing, I describe this as, not even the tip of the
iceberg. It's barely the tip of the tip. An unknown number
with ARC, an even less well-defined group with lymphadenopathy
syndrome and on down the earliest stages and a big question mark.
I use the Coolfont estimate of 1 million to 1.5 million because I
believe it's reasonable, but I'm going to get into more detail in
describing how we analyze that number a little bit further. We
don't know the extent of the infection in the existing population
and even less about that in the insured population. Again, we're
trying to measure this iceberg by citing just that small tip
that's peering above the water. The San Francisco City Clinic
studies are the only published data that we have that show
trends in the specific population over most of the period since
HIV entered the population. So what Mr. Hoskins did at this point
was sort of run a model in reverse. In other words, we said,
given the CDC's number of reported AIDs cases and deaths, and
assuming the progression along the curves described earlier, what
would the number of new infections in each of 1986, '85, '84, '83
and going on back to the mid-'70s have had to have been in order
to support this number of AIDS cases and emerging deaths.

272

 

 
 

 

We were closely able to replicate the AIDS cases and
deaths if we assumed that HIV entered the male homosexual and IV
drug abusing populations about 1975. It could have been '76, '77
or '78. It really didn't matter an awful lot which year we
chose, but sometime in the mid- to late '70s, and that it
progressed along your classical logistic curve.

You probably are familiar with the latest report from
this department, the Centers for Disease Control in Atlanta, the
November 30 report. They describe three kinds of curve. They
describe the exponential, the logistic and the log logistic, and
then they have various adjustments of those. We chose the simple
logistic curve. We did suppress it slightly, and that is the
reason that you will see the green model.

Now, the effect of suppressing it slightly gives you a
higher earlier but a lower later spread of the model. It was
this -- for any of the mathematically minded, I'll be glad to
give them the parameters of this curve. By assuming that the
disease had progressed along this curve so that by early 1987 it
had infected on the average approximately 27 percent of the high
risk groups, we recognize that it was much, much higher in the
most promiscuous male homosexual and probably very much lower in
some of the other groups, but on the average about 27 percent.

That the progression from HIV infection to full
clinical AIDS followed the CDC San Francisco study and the
mortality of AIDS patients follows the rate calculated from the
cpDc data, we developed a single model but a number of runs of
that model. If you will open your report to the centerfold in the
appendix, you will see the numbers that we came up with which we
believe to be a reasonable representation. The numbers show
total HIV infection somewhere toward the end of '86 of about
670,000 and towards the end of '87 of about 920,000.

Now, you will also note in our report that we included
only the highest risk groups. We did not include hemophiliacs.
We did not include heterosexual IV drug abusing females, and we
aid not include heterosexual males. The reasons that we didn't
were not that we didn't think there were some in these groups
that were infected but, rather, the hemophiliacs we would not
have included for insurance purposes most likely anyway.

We didn't think there was any likelihood that we had
any significant cohort of IV drug abusing prostitutes in our
insured population. We simply didn't know the numbers of the
heterosexuals. But we were seeing significant numbers of deaths
in the insured population, specifically in 1986 in which there
were 8,140 reported AIDS deaths, AIDS deaths reported to the CDC
as of mid-year. That number has gone up slightly since then
because of reclassification, but essentially just over 8,000. Of
that, approximately 3,000, slightly less than 3,000 for

273

 

 
 

individual policies and slightly more than 3,000 for group
policies were covered by life insurance. Now there are also odd
data on the number that were covered by disability and health
insurance, but the numbers there are much more difficult to
classify because of the disease. Once an AIDS victim is
reported as dead, there is very little question about the
classification, whereas during the progress of disease for
disability and health insurance, the classification isn't nearly
so precise, so that numbers are more difficult to measure.

Nevertheless, it was guite clear that the insurance
industry had paid out approximately 1 percent of its life
insurance claims in 1986 to people who were reported to have died
from AIDS. That did not include any who had died from other
causes, other infections, other opportunistic infections related
to HIV. We believe in the future that those numbers will also
be available and that we will get a better estimate of the total
financial impact of HIV as opposed to just AIDS. The next chart
shows the total AIDS cases in our model in the CDC with the
projection from the Coolfont study. As you see, we are
essentially tracking the CDC numbers through 1991. This shows
the total AIDS deaths from our model and the various low, mid and
high numbers in the Coolfont study and the further projection.

We went way out on a limb to the year 2000 in order to
be able to estimate the impact on the life insurance industry for
the end of the century. Of course, again, I would have to say
that we are using epidemiological models that are sort of based
on classic epidemics, and what we're looking at here is an
epidemic that is certainly far from classical, with its long
latency period, and hence we have to be very careful in assuming
that any of our numbers are that reliable. However, because
we're all more or less coming in at the same order of magnitude,
I conclude that either we were all right or we were all very
wrong.

I tend to think that we're all fairly close to the number,
but I said in the report that the number of HIV infecteds is
about 1 million as of mid-year this year, and I find some
difficulty in getting that number much below -- the low is
three-quarters of a million or much above about a million and a
quarter. But I would not be willing to stake my actuarial
credentials on giving it any greater precision than saying it's
in the order of magnitude of 10 to the 6th power.

ADMIRAL WATKINS: Mr. Cowell, do you have any feeling
for any of your curves about its sensitivity to the latency data?

MR. COWELL: Yes.

ADMIRAL WATKINS: Because we just went from 4.5 to 8
years nominally. The question is suppose we go to 12? Then how

274

 

 
 

 

significant is the fluctuation against those kinds of latency
steps? Have you got any curves that show you the kind of
flexibility you have?

MR. COWELL: Yes. I don't have them with me here, sir,
but we have modeled -- we have run our models at different
sensitivity at different levels. You will find, for example, if
you have -- do you have a copy of the report in front of you?

ADMIRAL WATKINS: Yes, I do.

MR. COWELL: If you will turn, for example, to page --
the first part of the report. The report is in two parts. If
you will turn to the first part of the report to chart 4, which
is on page 14, you will see the progression. Do you have chart
4, sir?

ADMIRAL WATKINS: Yes.

MR. COWELL: You will see the progression under the
five assumptions I describe plus also mathematical formulation
referred to as Weibull, which is, for the epidemiologist, a very
simply log logistic progression which tends to have a somewhat
slower progression earlier and a higher progression later.

We believe that this reasonably reproduces what I call
the envelope or the family of curves of progression. What we did
which may make sense only to somebody in an insurance
environment, and I'm sure Mr. Creedon will understand this, we
translated this into the dollar impact on the life insurance
industry. That is, if the faster progression shown here by
Frankfurt were to be used, it would mean that in insuring an
individual for $1,000, the present value of extra AIDS claims
would be about $545. If we used the slowest curve, listed here
as Weibull, the present value of additional claims discounted at
6 percent would be $479. In other words, the financial impact
projected out into futurity or through the end of the normal
mortality of any of these cohorts was relatively insensitive to
fairly significant shifts in the progression.

Now, our progression assumes a mean progression from
infection to AIDS of nine years, with a standard deviation of
3.5. That seems reasonably consistent with the San Francisco
City Clinic data of 45 percent progressing to full clinical AIDS
after nine years. It suggests that 50 percent after 10 years
might not be an unreasonable estimate.

ADMIRAL WATKINS: Does that also take into account the
fact that the latency period is extended even further, that you
have a pool of infected individuals that have more potential to
infect others during that same period of time?

275

 

 
 

 

MR. COWELL: That is correct, sir, and that is
something that we explicitly recognized. We did not explicitly
model for insurance purposes in this report. Recognize that this
was simply the first report. The work of the Society of
Actuaries and the ACLI task forces that I mentioned earlier are
ongoing. By late March/early April or late April of 1988, we
will have carried our modeling one step -- well, several steps
further to include better estimates of the progression within
subgroups of the population, and we hope by then to be able to
get a better measure on the impact on the industry not only on
life insurance but some of the other lines of business, taking
into account this phenomenon that you're describing. That is,
with the slower latency period, longer latency period, you will
have a larger pool to infect.

Now, the only qualification I'll make to the point that
I only implicitly recognize this is maybe if I show you the next
slide, the following slide which shows our estimates of the
numbers of infected, this is what I come up with. By the way,
you see this is just U.S. males because we totally ignored the
number of females because it is, we believe, to be so small,
probably fewer than 100,000, and we concentrated on age groups 20
to 59 which, of course, is where the primary insurance impact is.
Here is our million, and I intentionally had that come out at one
million even, mostly because I didn't want to have any numbers
beyond the first digit take on an aura of precision that I don't
believe they really deserve. So I guess I somewhat reluctantly
broke it down into groups of homosexuals and bisexuals a little
over three- quarters of a million, and IV drug abusers just under
a quarter of a million, both of which numbers I've heard even
after getting into the room late here this afternoon.

My estimate is that there are fewer than 100,000
females infected. The point of all this by way of saying that
even though we were -- even if we were to assume a longer
latency period and my mean of nine/standard deviation of 3.5
would suggest that two-thirds of the time -- it will maybe be as
long as 12.5 years, so it will be between 8.5 and 12.5 years --
even within that range, since the disease is presently believed
to be concentrated primarily among high risk groups of male
homosexuals and male bisexuals and male IV drug abusers, I would
not expect -- and I'm going to be interested to see what our
models show, but I would not expect that even modest variations
from that latency period would have a significant effect in the
early years.

I say this because evidence that we are picking up in
our task forces, from listening to epidemiologists and medical
specialists, that the diseases spread is still largely confined
to these groups and that the principal spread into the
heterosexual community is most likely to be the single channel of
heterosexual males visiting IV drug abusing female prostitutes,

276

 

 
 

 

that that is still a fairly narrow channel and that even
substantial variations in the latency period probably for the
next several years won't significantly affect those numbers,
probably won't make significant -- at least significant from our
standpoint in the insurance industry -- impact until probably
into the early part of the 21st century.

I did not attempt to model this beyond the year 2000.
I thought I had sort of gone out on a limb going out 13 years,
but I feel reasonably comfortable with my estimates. Sorry if
that was a long explanation, but that sort of gives you an idea
of our thinking. That pretty much concludes my prepared
remarks. The rest of the numbers are largely for insurance
consumption only, and I'll be glad to answer them, but we did
conclude from all of this that the epidemic will likely have
about a $30 billion impact on the insurance industry through the
end of the century. It will be larger or smaller, depending upon
the estimates that we have made.

Assuming that the number of HIV infecteds does peak out
at somewhere in the 2 million range by the late 1990s and doesn't
go any further and that there is no significant spread into the
heterosexual population, I concluded in our report, and I realize
this is not maybe of direct interest to this Commission, but the
impact on the insurance industry is probably of interest to Mr.
Creedon, would be serious but still probably manageable,
provided, of course, we have the option of testing people for the
virus. ,

MR. CREEDON: That's the life insurance industry.
MR. COWELL: That's the life insurance industry.

MR. CREEDON: That does not include the health
insurance.

MR. COWELL: That's correct, sir. It does not include
the impact of health insurance. Preliminary evidence is that
health insurance claims, both medical and disability income, are
probably running about seven to eight-tenths of one percent of
total claims. Of course, if that number, one percent for life
insurance and, say, seven to eight-tenths of a percent for health
insurance, are true in 1986, were true in 1986 when the overall
mortality from AIDS in the United States was 8,000, then we think
it not unlikely that those numbers will increase approximately by
an order of magnitude to maybe 7 and 8 percent for health
insurance and maybe 10 percent or slightly higher for life
insurance by the mid to late 1990s, at which time we project --
we concur in the CDC projections that the total deaths will
probably be running somewhere in the 50,000, 60,000, 70,000 range
per year.

277

 

 
 

 

So we see approximately a ten-fold increase or an
increase in the order of one magnitude in the impact on the
insurance industry before the end of the century, assuming, of
course, no major breakout into the heterosexual population.

MR. CREEDON: While you were wandering around the
building, Mr. Cowell, I mentioned to the other participants that
we would like to hear from each of them and then we would ask
questions of the whole panel. So I would like now to ask Mr.
Pascal to speak. Thank you very much.

MR. COWELL: Thank you.
PRESENTATION BY MR. ANTHONY PASCAL

MR. PASCAL: Thank you very much for inviting me here.
I'm going to have to leave at 4:00 for a plane back to the West
Coast, and so I hope I'll be around for questions, if there are
any, but I can't promise to be. I'm the author of a report that
we did at RAND for the Health Care Financing Administration on
the cost of AIDS to the Medicaid system. I suspect the reason I
was asked to come here and testify today is because of the case
load estimates I had to make for that report, although the report
concentrated strictly on the economics of the epidemic and not
the epidemiology. I am not an epidemiologist or a public health
expert. I'm an economist. Neither am I a model builder.

MR. CREEDON: We won't hold that against you.

MR. PASCAL: There is an epidemiological modeling
effort going on at RAND under David Kanouse and Scott Cardell,
and they were coming up with estimates which I can't quote. The
report will be published early next year. But they look
considerably larger than the CDC numbers. Since I had to have a
case load estimate and couldn't use the RAND numbers, I started
looking into the underlying construction of the CDC numbers. As
you all know by now, they're based on incidence curves only; that
is, it's kind of a top down approach based on what we see in
terms of AIDS incidence.

We know that those are highly sensitive to reporting
biases; to changes in reporting biases over time; to delays in
reports, which are getting worse instead of better. Although
these might be good numbers for the short run, HCFA was
interested in a five-year projection into the mid-'90s, so I
needed to find an alternative to make a bound. I took the cpc
numbers as my low side estimates but was interested in bounding
on the high side. I did something very simple, which was to take
the CDC prevalence number of 1.5 million and the IOM estimate at
that time -- now remember, this is the high side, the worse case
~~ of 50 percent conversion to full-blown AIDS over a five-year
period and came up with a 750,000 cumulative at that point.

278

 

 
 

 

That seemed quite high, and so I finally settled on an
intermediate number to do my main line estimates. It turns out
that the 400,000 figure I used of AIDS cases by the early '90s is
not so much different than you'd get if you added the 20 percent
underreporting factor to the CDC numbers and then added a 30
percent factor again for the new AIDS definition which, as you
know, includes wasting and dementia to the original definition.

I have a handout here by Kanouse and Company which
explains the reasons that they are doing this dynamic modeling
which takes into account the transmission modes, behaviors and
the natural history of the virus and alternative combinations of
assumptions about those items. We know clearly, of course, that
behaviors are changing in the various risk group communities.

We know that we're finding out much more about the natural
history of the virus, particularly the latency period that was
brought up. As Dr. DeGruttola said, we have to also take into
account the relationship between infection and the degree of
infectiousness. So all of these are built into the RAND study.
I think it's going to be out for peer review probably to some of
these people at this table within a few weeks, and then they
expect to submit it to a journal early in January. I think the
turnaround will be quite fast.

MR. CREEDON: Can that be submitted -- do you think it
could be submitted to us at the time it's submitted for peer
review?

MR. PASCAL: I will check. I couldn't answer for then,
but I will check on that. In the meantime, I wanted to mention
about the CDC numbers. I have just come from a two-day meeting
sponsored by the Council of Professional Associations on Federal
Statistics in which CDC people were grilled very extensively by
public health and epidemiological people, health economists and
the whole community about the numbers they produce. There
tended to be a high level of dissatisfaction expressed,
particularly with the disaggregations that CDC feels they are
able to provide. I'm talking particularly here about geographic
disaggregations, although as people at the earlier session
mentioned, the ethnic and racial breakdowns are also a problen.

cDC's position is that they can only be as informative
as the least responsive state from whom they collect the data
wants them to be. I think that's a very, very conservative
position on their part, and I'm wondering why and whether they
really need to stick with such a position, why we can't get
numbers from them, for example, even if we're not going to be
told how many cases there are in Greenville, South Carolina, why
we can't be told how many cases there are in SMSAs between
100,000 and 200,000 in the South Atlantic region.

279

 

 
 

 

The CDC position now is not to provide data at even
that level. Clearly, they're following their public
responsibility; that is, in terms of disclosure and
confidentiality and not putting people at risk of being
identified. But to take the standard of the most restrictive of
the states rather than trying to convince the states to be less
restrictive on the one hand or to publish what they can for other
states seems to me something that needs to be looked at quite
carefully.

There are other problems, I think, in coordinating cDC
data presentations with presentations of other federal agencies
whose data we have to combine with the CDC. They classify cities
into different regions of the country than the Census Bureau
does. For example, there are other situations of a similar kind
that make difficult the use of data by people studying the AIDS
epidemic either in terms of projecting seroprevalence or
projecting case loads or trying to estimate costs.

Finally, I might say that from what I heard in the past
two days at the COPAFS conference that there are cases, it seems
to me fairly clearly, where a state limits the access to data not
so much because they are afraid of disclosure or of identifying
persons with AIDS but because they're afraid of the effect on the
tourist industry or they're afraid of the effect on the real
estate market in particular communities. I think that needs to
be looked into very carefully as well. I'll stop there. Thank
you.

MR. CREEDON: Mr. Robertson.
PRESENTATION BY MR. ANTHONY ROBERTSON

MR. ROBERTSON: Mr. Chairman, thank you for inviting
me. I'll be very brief. I'd like to say, first of all, that I
think I agree with everyone else sitting at this table, that
essentially modeling efforts all lead to equally plausible
results, and we have a great lack of data. My main thrust is to
ask you to recommend that we collect data in a more expeditious
and efficient way. However, the data we do have can tell us
something, and I'd like to go through a few aspects of that.
This, of course, is a very strange disease. I think in any other
situation, it would be axiomatic if we were going to try to
control the spread of a disease that we need who is infected, who
isn't, and can inform them and try to hold the spread of
infection in that way. But not only do we need that kind of
information for slowing the epidemic, we also need it to know --
to be able to plan for the insurance industry, for the health
care industry and so on, and we need to know how effective our
actual interventions have been. From all these aspects of the
epidemic, we are woefully lacking in data. We really have no
normative data now that will allow us to say what effect, let's

280

 

 
 

 

say, behavioral changes, education and so on will have in the
future. Now, as many people must have emphasized to you, we are
largely dependent at the moment on the reporting of the terminal
-~ the preterminal stage of a new disease. That's very
dangerous. HIV infection is what we should be concentrating on,
and I think the history from other countries, particularly
central Africa, tells us how dangerous it is to count things by
the last stages of a disease, particularly a new disease.

Another danger which is purely theoretical at the
moment but for which we have some evidence is that this is a very
rapidly evolving virus. Now, this doesn't mean that it's going
to find new modes of transmission, for example, but it does mean
that it will vary in infectiousness and virulence. Now, these
variations may not be important, but we know from the molecular
virulty of this virus and its near relatives that they're
occurring rapidly. Again, this means that we need much better
data right now so that we can compare the virus as it is now, its
infectious and virulence and latency, with the virus that it
becomes, because this is the first stage in a new epidemic in the
human species, we think.

It's typical of new viral diseases that they evolve
very rapidly in the new host and that one really can't depict the
intermediate and long-term parts of such an epidemic. We've seen
this before, perhaps not in people but in animal diseases that
are spread from species to species. So this is something to bear
in mind, and it emphasizes the importance of keeping really good
data now. What do the current data show? I think they show
that, except for the earliest infected populations, which are the
male homosexuals in San Francisco, drug addicts in New York, that
the rate of growth, of new cases is still exponential. This
probably reflects an exponential growth in infection five to ten
years ago. Of course, in some populations that have been well
studied, the infection saturated a few years ago. This hasn't
happened in the largest part of the population. Prevalences are
very low, and we really again have no data on which to base
projections. It's dangerous to --

MR. CREEDON: The exponential growth is in what area.
specifically?

MR. ROBERTSON: At the beginning of an epidemic, —
typically one sees exponential growth and then a saturation, and
that's certainly happened. That saturation has occurred in some
populations. But in the general population, we're at a very
early stage, and we lack information about growth of this
epidemic at early stages.

MR. CREEDON: Are you saying that you have seen data
that suggests that there is now an exponential growth in other
than the homosexual and IV drug users' communities?

281

 

 
 

 

MR. ROBERTSON: No. I'm saying we don't have the data,
but I'm saying that we do see -~- well, let's take the
heterosexual risk group as an example. This is the cpDCc
heterosexual contact group. This is a very narrowly defined
group. It doesn't include people born overseas, for example,
which distort the pace of the epidemic. If we look at that
group, its growth rate, its doubling time is approximately nine
months at the moment, but it's at a very early stage in the
epidemic. If we plot out the cumulative incidence for that risk
group in terms of AIDS cases diagnosed, I think it's now 1,050.
Two years ago it was about 250. If we look at the curve and
Slide it along the time axis until it lines up with the
homosexual curve, it's about 4.5 years later. What I'm really
trying to emphasize is that we're very early in that epidemic,
and it's very dangerous to make assumptions about the prevalence
of infection from the current prevalence of AIDS cases because
the growth rate of those curves is essentially the same for all
the populations we're looking at.

The time is the biggest co-factor when we compare these
populations, so we mustn't confuse infectiousness or
transmissibility with current prevalence of AIDS cases. I'm sure
you've heard this before. We do have some limited data on new
infections and, again, it's very dangerous to jump to
conclusions from them. As I'm sure you know, the Army data can
be used to argue that the disease has essentially stopped but, in
fact, they don't show that. They show a continuing new incidence
in specific cohorts, in specific age groups. This new incidence
is certainly an underestimate of what is happening in the
population at large, but it's very useful in the sense that it
gives us some data to compare from year to year. We know from
the sentinel hospital survey that in low risk populations from
low risk parts of the country that there is a current incidence.
We don't really know what it is, but it's fractions of a percent,
up to about one percent in the midwest. I think that's quite a
high number, in fact, given the nature of these populations.
Unfortunately, the data are woefully inadequate. We only have
samples from about 4,000 blood tests now, so any number one
quotes has large uncertainties, large statistical uncertainties
associated with it.

I say this not to say that these are poor data but to
emphasize that we should be collecting sentinel data of this sort
with much greater efficiency and with a much greater sense of
urgency. This is relatively cheap to do and is very significant
in its value to us. Another kind of data which is easy to
collect, although it has certain ethical problems, but which,
again, is very cheap is the data from newborns. I think you know
that studies of newborns in Massachusetts show an incidence of
infection in their mothers which ranges from a couple of percent
down to a fraction of a percent, depending on the location of the

282

 
 

 

hospital. Again, that information is cheap and easy to collect,
and I would recommend that it be collected from as many

locations as possible, as quickly as possible. It's much easier
to do than a random survey over several years, and it's very
informative. It will show us quickly whether there really is new
incidence or not and how prevalence of infection is changing is a
better substitute for the general population.

I'd like to give an example, too, of the danger of
looking from month to month at the reported numbers. We have
very little information about reporting efficiency in different
groups. One would assume, for example, that it's fairly high
amongst hemophiliacs. We think that about 10,000 hemophiliacs
were infected, and the current CDC data show that about 450 have
now been diagnosed with AIDS; in other words, 4 to 5 percent of
that group, which would give a ratio of 20 to 25 to 1 for that
part of the epidemic for a group of people which was infected
fairly rapidly between 1981 and 1984 and isn't, we think, being
infected now. Such a saturated group, and there we have probably
a fairly accurate ratio of infection to AIDS cases.

We have that kind of ratio for other groups, but we
really don't have it for the latest risk groups, such as the
heterosexual one, although that is clearly a much smaller group
at the moment. If we look at the cases reported for homosexual
men, for heterosexuals, men and women, and so on, and look at the
growth rates in new cases amongst the various risk groups,
there's really not much we can say about transmissibility or any
other factor in those groups. They're all growing at probably
the same rate, given their different total size. If we placed
them along the time axis of the epidemic, they're all roughly
similar at the moment. So the information, really, about
efficiency of transmission isn't there. The only place it can
come from is from properly conducted prospective cohort studies,
some of which are being done now. There's a great urgency, to
enroll more heterosexuals and particularly young people in these
cohort studies.

I think it's very important. to do that correctly, too,
so that we see the early stages of the parts of the epidemic that
we've so far missed. We know the epidemic is there. We just
don't know its size, and I think it's very important to get
those data, both to know the problem we face and to know what
impact any actions we may take have on it. There are surprises
that we might get. I think we've had some in this epidemic so
far. We need very good data on transmissibility -- we haven't
got it yet -- by each root of transmission. We need data on the
virus that's required for infection. We've got that ina
limited way, but it's still too limited. A lot of money, I ;
think, should be put into epidemiological studies of this sort.
We cannot afford to wait for a cure or a vaccine.

283

 

 
 

 

We need to follow other sexually transmitted diseases
for two reasons. First of all, while they must be co-factors of
the development of AIDS, they're certainly markers for behavior
that could lead to AIDS. In this context, it's very interesting
that two of the classical sexually transmitted diseases,
chancroid and syphilis, have shown dramatic increases in those
parts of the country where the AIDS virus has been endemic
longest. Now, there could be many explanations for that. More
people might be going to STD clinics. They might be more
nervous. There might be a real increase in these diseases, and
the data tend to suggest that.

The increase might be because there's a community of
immunosuppressed people who are now more infectious, although the
actual diseases, syphilis and chancroid, are being seen in the
heterosexual groups rather than the homosexual in these new
infected. But it's very important to keep data about all STDs
and about their relationship to HIV infection and to gather those
as quickly as possible so that we don't have nasty surprises. If
we really are seeing epidemics of new STDs amongst the
heterosexual population, these could have a great impact not only
on the spread of AIDS or of HIV but on the total medical costs
that we face.

We've already seen an increase in TB in AIDS endemic
areas, and I think it is vital to keep good data for other
diseases, particularly STDs both as markers of AIDS to come and
of signs of the effect of having a large part of the population
immunosuppressed. I think I'd like to stop there. I just
recommend that we really do need to spend more time and thought
on the epidemiologics.

MR. CREEDON: I would like to ask a number of general
questions and perhaps Mr. Pascal could react, for instance, first
since you have to leave at four o'clock. We had one witness
yesterday, Mr. Langmuir, who at one point was the chief
epidemiologist for CDC. He is retired now and he is, I guess,
about 80 years old but he is quite active and he gets the CDC's
statistics regularly and has been plotting them and applying
Farr's Law to the data.

He has constructed a curve that goes something like
that and I guess one of the questions is, are each of you
familiar with his work and what do you think are the implications
of it for us? I mean, is it completely off base or is something
like that happening or going to happen? Would it make a
difference if you regarded this not as one epidemic but as a
series of sub-epidemics and then applied his curve to the
sub-epidemics? Do you have a reaction to that?

MR. PASCAL: I think I am the least relevant to comment
on that.

284

 

 
 

 

MR. ROBERTSON: It certainly is important to treat it
as a series of sub-epidemics and this is because not only because
it is but because the nature of the disease with its long
latencies, very long latencies, which may be different from group
to group, there is no evidence for this yet, except perhaps from
old and young people, but because of the latencies involved we do
have very separate patterns of disease in the different groups so
that there are separate epidemics and eventually they will follow
separate time courses. As I mentioned, as other people have
mentioned, we have seen a saturation in some aspects of the
homosexual epidemic so it is not useful to think of the epidemic
as one whole for some purposes. So, yes, I would agree.

MR. CREEDON: Yes, Dr. DeGruttola.

DR. DeGRUTTOLA: I have seen a presentation of that
model and as I understand it, what he was doing was taking a
curve with a particular shape, I think it was a log normal curve,
and fitting it to data and saying that because this model fit
relatively well to data it would predict the future of the
epidemic. I really think that, in fact, there is a huge range of
models that would fit. There is so much uncertainty in the data
and there are so many different kinds of models that fit that you
can't really believe any one. You have to attempt to fit several
different models and see if the data can distinguish among them
and in point of fact, it can't as I tried to demonstrate in my
talk.

Even the ability to discriminate between a sigmoidal
shape curve, something that started out exponential and then
flattened out and something that was flat all along, it is not
possible to do. The other thing is that model did not take
explicit account of the latency period which we now believe has a
mean, at least for the one well-studied population, of eight or
nine or ten years and for other populations it is unknown; no
reason to believe it is considerable shorter than that, possibly
longer if they have fewer co-factors. So for both of those
reasons, I really think that you can't accept that model as
giving you very precise information about the future but I think
what you do have to do is try fitting a variety of different
models, both models that are simply extending the incidence curve
which is what the CDC does as well as models that attempt to
distinguish between the curves of infection with HIV and the
distribution of latencies between infection and AIDS.

I think if you do that, most of those models are going
to be predicting much worse outcomes than the one predicted by
Dr. Langmuir's model although I must say that I think projection
beyond 1991 and 1992 at this point is extremely difficult because
of lack of precise information. I think we should attempt to do
it but I think that our uncertainty grows as time goes on.

285

 

 
 

 

MR. CREEDON: Do you want to add to that, Mr. Cowell?

MR. COWELL: Let Mr. Pascal, if he is in a hurry, go.
MR. PASCAL: No, it is not my area so I will defer.

MR. ROBERTSON: I would just like to interpolate an
example I meant to mention which is that we are totally dependent
on a belief that, for example, reporting efficiency hasn't
changed. We have really no independent estimates of that. We
don't know how good reporting is in various groups. We are also
very dependent on how we define the disease. Now clearly, one
has to have a very clear definition and the CDC is absolutely
right to have one for reporting purposes. That may not be
relevant to total cost, however.

If you look at the effect of the new CDC definition,
the rate of reporting in the last ten weeks has gone up exactly
40 percent. Now that probably is a transient but we don't
actually know what that means. We don't know if these are people
who would eventually have been reported or whether they are a new
group or whether the relaxation of criteria for diagnosis is just
broadening the epidemic. But again, if we tried to model from
that which gives us at the moment a doubling time of 12 months
rather than 16 a few months ago, well, the process is clearly
fraught with danger.

MR. PASCAL: We certainly know from this recent study
of the New York City Health death certificates for the IV drug
users that even in a place with a very sophisticated public
health service with probably among the best reporting in the
country, there can be enormous underestimates of AIDS incidence.

MR. ROBERTSON: The real question is, is that changing.
MR. PASCAL: Is it changing over time, right.

MR. CREEDON: One of the key questions it seems to me
that we have to consider as a Commission is whether the data that
cpdc is publishing is reliable data. In other words, is it ina
range that it is reasonable for the Commission to accept.
Obviously, the process that is followed here leaves a lot to be
desired and Mr. Pascal, you said that there was a lot of
criticism at the meeting in the last two days and I wonder if
you would each comment on that. In other words, is the
Commission safe or acting reasonably or prudently in relying on
the range that CDC is coming up with or do you have specific
recommendations as to other things that we should do or
consider. Why don't you start, Mr. Pascal, because you were
subject to all the criticisms over there.

286

 

 
 

 

MR. PASCAL: My problem is not so much with the
reliability of the data as the level of aggregation at which it
is published.

MR. CREEDON: What does that mean exactly? ©

MR. PASCAL: The detail that you get in terms of
geography, in terms of risk group, in terms of transmission mode
and so forth. It think that more could be done with some
pressure on the states perhaps or even with some more courage on
the part of CDC to present data at the levels of disaggregation
that are needed to do the studies.

ADMIRAL WATKINS: I think it is very important, Mr.
Pascal, you have really opened Pandora's box with me on your
commnts from that group. I need to know what the name of it is.

MR. PASCAL: I have a program here.

ADMIRAL WATKINS: Is there a report from them yet or is
there one in preparation?

MR. PASCAL: It is a group made up of professional
associations such as public health associations, economics —
associations and so forth that use federal assistance.

ADMIRAL WATKINS: Do they have a small executive staff
with an executive director?

MR. PASCAL: Yes.
ADMIRAL WATKINS: Who is that?

MR. PASCAL: Kathryn Wallman. I can give you the
address.

MR. CREEDON: Coincidentally, they were having two days
of hearings yesterday and today.

MR. PASCAL: Yes.
MR. CREEDON: Same time as we were.

ADMIRAL WATKINS: I think it is extremely important
that we be able to liaison with that executive director, get a
summary of the meeting because we hear these things from grass
roots as we move around and yet we haven't had any definitive
presentation to us of what a large group of people who have to
use that data and see it feel about it. It seems to me that it
would be extremely valuable for the Commission for us to get that
presentation or certainly to get any sort of paper that comes out
perhaps with the presentation later on.

(287

 

 
 

 

L
MR. PASCAL: There were no formal papers ‘presented.
They were panel discussions led by moderators but I think Kathryn
Wallman who was the organizer of the meeting could provide you
with a good bit of what you need and there is also an excellent |
participants list of people who were there to voice their
concerns.

ADMIRAL WATKINS: We have had other presentations today
on the need for disaggregation from minority groups, Hispanic,
Black, co-factors, many other things that may not normally fall
within the epidemiological data base, but with a new disease like
this, it becomes extremely important not to throw away
information at the same time we are getting other data.

Certainly we need to focus on the HIV infection and not just talk
about the things that would normally be talked about in a
morbidity and mortality report but rather expand our vision for
AIDS get the perceptions out there that the HIV infection rate is
something that we need to focus on a great deal more. We keep
seeing reports coming out on deaths and morbidity which somewhat
disguises the rather significant HIV infection number, even
though it is presented somewhere else in the report at a million
to a million and a half. Do you see what I am saying?

MR. PASCAL: Yes.

ADMIRAL WATKINS: So I think your input to us, and the
input. of that particular group could be very valuable to the
Commission.

MR. CREEDON: Do the other participants want to
comment on their view as to the reliability of the CDC statistics
as to the degree of infection? Is it reliable? If it is not
reliable, is it unreliable in an order of magnitude that is
significant for us? Both Mr. Cowell and Dr. DeGruttola seem to
come in at around a million or somewhere between 750,000 and a
1,250,000 which is in the same ballpark. Would you comment on
that, please?

DR. DeGRUTTOLA: I think that there are a number of
different issues here. First is reporting of incidence of AIDS
itself and as we know, that is an extremely problematic thing and
there has been evidence of considerable underreporting in New
York. There is evidence of considerable variation regionally in
delays in reporting. But I think that those kinds of problems
are inevitable and I think compensation can be made for them. I
have seen some good analyses done particularly by Jeff Harris at
M.I.T. on reporting delays and how data can be used, the data
that are available, to adjust for reporting delays and get a much
more precise estimate of incidence of AIDS itself. So I think
that although we know there are reliability problems with the
reporting of AIDS, I think they are probably inevitable and

288

 

 
 

 

those data are still highly useful. I think the problem with
aggregation is one of release of information to people. I
worked with Professor Steve Lagakos on a problem of estimating
the average latency period. We know how important that is from
transfusion related data. We believed several years ago that
there was a problem with the analytical approach that people at
cDc had taken but were not able to get the actual data.

In fact, we had to wait until an article was published
and then make use of the published data which was not precise
enough for the purposes that we needed so there is a separate
problem in release of data. The third problem is in projections |
and both for estimates of prevalence of infection right now and
for projections of AIDS into the future that isn't simply a
question of reporting or releasing the information. It is a
question of coming up with some kind of model and being able to
justify it. I think that the approaches that have been taken
both to extend the incidence curve into the future is one
approach that has to be looked at as long as there are sufficient
caveats that there may be important effects that are not now
distinguishable from time trends in incidence of AIDS but could
be crucial later on. For example, saturation of certain
populations like homosexually active men might be very hard to
observe in incidence of AIDS up until now but may be very
important in the future. So long as it is presented with the
right caveats of the limitations of the model, then I think that
those models are fine.

I think that what people who are interested in this
kind of data have to do is to encourage that there are many other
modeling approaches that are taken both by CDC and outside CDC
and that means making data, precise enough data, available to
different modeling groups so that we can see the full range of
possibilities. In terms of estimating prevalence of infection
right now, I think the approach that was taken at Coolfont to try
to estimate the size of the populations at risk and the
prevalence of infection in those populations, I think that
actually the data are just not there to give reasonable answers.

It happens that their estimates lie fairly close to the
ones that both I and Mr. Cowell came up with but I really think
that we have only a very crude idea of how many men are at risk
through homosexual contact and also what the prevalence is among
such men because it is going to be very, very different in San
Francisco than it is in the Midwest and it depends on ranges of
behavior and so on. So I am less confident of those, not the
figures themselves which may be okay but the method.

MR. CREEDON: Dr. Joseph who is the Commissioner of
Health in New York City was a witness this morning and he said
that they have done some work and we don't know exactly how they
did it but he thinks that there are 400,000 people in the City of

289

 

 
 

New York who have the virus right now, in one city, 400,000
people. Now what does that do to the nationwide figures if he is
correct in his judgment?

DR. DeGRUTTOLA: I would think that that would be a
very hard number to defend. I don't know the method that he used
but my estimate for New York City would be lower than that.

MR. CREEDON: Do you have an estimate?

DR. DeGRUTTOLA: I don't have an estimate for New York
City itself but I could say that there have been about 15,000
cases in New York City, less than that. If you had over 400,000
people infected, that would mean that you would have to have a
ratio of about 30 to one, 30 cases of infection to incidence of
AIDS. That would seem to be quite high in a city in which
several sub-populations of homosexually active men have already
been saturated.

In other words, to get to ratios that high what you
need is a lot of recent infection and unless there was a lot of
evidence of a lot of recent infection in young IV drug users and
their heterosexual partners, I think it would be tough to get up
to that figure and I am afraid people are developing those ratios
from data that were observed several years ago when there were
fewer cases of AIDS and consequently, a higher ratio of
prevalence of infection to incidence of AIDS and applying them
currently and I think they are getting inflated estimates.

MR. CREEDON: One of the things we found in the last
couple of days is that a good deal of the important data that is
available now stems from a study of the homosexuals in California
which was started because of Hepatitis B concern at that time and
I wonder how many studies of that type are currently underway or
whether you would have recommendations that we try now to
encourage such studies to be undertaken in various places around
the country on an ongoing basis and especially perhaps with
younger people, people who are teenagers now or groups of that
kind who ten years from now would be giving us this same kind of
data that we got from the San Francisco study. I wonder if you
would comment on that as to what is available now and what your
reaction would be to other things that might be done?

DR. DeGRUTTOLA: I think there are a number of studies
in place right now to estimate risk of AIDS after infection.
There is the cohort in San Francisco but there are also four
other studies of mostly homosexually active men, 1,000 at each
center, which means that there is a considerable effort to
looking at natural history of AIDS.

MR. CREEDON: But don't we have to do to it for other
than homosexual men?

290

 

 
 

DR. DeGRUTTOLA: Right. I think that there are fewer
studies in IV drug users although there are some but it is a much
more difficult population to get at. There are also studies in
hemophiliacs and there are ways of looking at that for
transfusion-related AIDS. I think that the second part of your
question, what about young people, I think that is the area where
we really know the least and there are two issues. One is, what
is the rate at which they are getting infected? That is
something that we clearly need to study much more and the other
is what will be their risk of AIDS after infection. But I think
that the most important piece of information that is lacking is
what is the rate at which young people are becoming infected.

MR. PASCAL: I am sorry I have to depart.
MR. CREEDON: Thank you very much, Mr. Pascal.

MR. PASCAL: I will leave the information of the group
with Ms. Knox.

MR. CREEDON: Thank you.
ADMIRAL WATKINS: Thank you very much, Mr. Pascal.

MR. COWELL: Maybe, Mr. Creedon, I could address
Admiral Watkins' question generally, the reliability of the CDC
data. I have come to rely heavily on the CDC data. I have come
to regard them as probably the most reliable in the aggregate. I
will get into the question of disaggregation in a minute. But I
might depart slightly from Dr. DeGruttola's point. I think it is
very well to look back two years or a year and a half and say
that the Coolfont numbers were really not well-based but I think
considering the limited amount of data the CDC had in Coolfont a
year and a half ago, I am amazed how closely they predicted the
overall, what I would call shape of the epidemic. Admittedly,
they may have been a little bit high on their estimates at that
time.

I totally agree with Dr. DeGruttola on the point of not
trying to impute more precision to these numbers than the data
that we have warrants and I so state that in page four of our
report, that while we recognize the ultimate goal of our study,
at least, was to attain credible numbers at this stage of our
analysis and I would say that that is as true now in December as
it was in the middle of the year or when we started, we place
more importance on validating the reasonableness of the processes
than on the absolute numbers they generate.

Now having said that, of course, everybody wants to
know what your numbers are and as I pointed out, we did in our
study take the epidemic on a logistic or a lightly damped

291

 

 
 

 

logistic curve that I will demonstrate in a moment in a picture
and link that through a fairly simply mathematical model to the
progression rates from San Francisco which are by the way and
there may be some other studies going on but the San Francisco
City Clinic CDC data are the longest longitudinal studies which
showed, for example, that in the highest risk group, male
homosexuals in the San Francisco study, four percent of the
unfrozen blood samples were seropositive in 1978 and
approximately 75 percent of that same cohort were infected in
1986. So that gives you some idea of the progression in a high
risk group.

The Frankfurt data are probably the most reliable. The
1986 study gave a three-year progression and Frankfurt has just
published the five-year progression and those data are currently
being analyzed by the Society of Actuaries group studying this
and that information will be published in April. Now as to the
absolute numbers, if, for example, 400,000 HIV infections in New
York City is reasonable, then based on a very crude grossing up
because New York City has about 28 percent of the total AIDS
cases in deaths to date, that would gross up to 1.4 million in
the United States.

It is a little on the high side of the average but
again, I will only say that the number is ten to the sixth power
which means that it could be anywhere from half a million up to
maybe two million and I would still say that it is a reasonable
number. Again, having said that I went out on a limb and
published my numbers that I believe to be reliable for insurance
purposes because they are largely male homosexuals. There is
very little heterosexual, well, there is some heterosexual, we
couldn't measure it but there is certainly minimal IV drug abuse
in those numbers, IV drug numbers in there, and if I were pinned
down to a tighter range somewhat reluctantly I would come up as
of about now, late 1987, with approximately a million, somewhere
between 900,000 and 1.1 million in the three highest risk groups
and another 125,000 in the three lowest risk groups, well, no,
one high risk group would be the female prostitutes and then the
others would be the heterosexuals and the hemophiliacs number
which probably combined, less than 50,000.

So I would produce a number somewhere between a million
and 1.2 million if I had to, if I were really forced to come in
with a tighter range but I am reluctant to do that because I keep
saying on the other hand I don't want to give this estimate an
‘aura of precision that I think it doesn't deserve. I really
don't think from the standpoint of your Commission, whether the
number is 900,000 or whether it is 1.2 million it really makes a
whole lot of difference because if it is 900,000 now, it will be
1.2 million a few years down the road and whatever we are doing
now that might affect those numbers, we will have to do in the
future. So I think to say that it is a million or thereabouts is

292

 

 
 

 

probably as much precision as anyone ought to try to give this.
I think that to the extent we go beyond that and I read a lot of
studies which is probably why my eyesight is so bad that I can't
find the front door of this place -- :

{[Laughter. ]

MR. COWELL: Maybe to redeem myself, I have tried to
draw this picture and it is supposed to be the one maybe not
worth a thousand words but maybe four or five hundred.

MR. CREEDON: You will have.to take the microphone, Mr.
Cowell.

MR. COWELL: Don't worry about the chart. I will
explain that. It is only for those in the group that are
mathematically oriented but I think it gets to this question of
aggregation and disaggregation.

ADMIRAL WATKINS: There will be a quiz for the
Commissioners on this slide. |

MR. COWELL: No, I won't give any exams. I was general
chair of the Society of Actuaries Education and Exam committee
for a couple of years so I know all about taking actuarial exams
and giving them and.I promise that there will be no quiz on this.
This first chart here shows in red a very steep progression that
might be thought of as typical of an extremely high risk group
that fills up very quickly such as, let's say, the highest risk,
San Francisco or the highest risk IV drug group. Now this is,
let's say, a less steep. They are both logistic curves and it is
just a matter, this is a generalized equation and just to
explain this, what this means is that if "P" is the population
that is ultimately infected, then this is one of the factors that
we don't quite know. This is the population infected.

This is the population uninfected and you sort of get a
sense of what the rate of change of the disease over an
infinitesimal fraction of time is proportional: to this percent
infected and the percent non-infected as the disease passes from
this group to this group and over a period of time you solve for
this and you come out with a differential equation or you can do
it in continuous Calculus and trust me, the numbers have been all
tested out and the actuaries all agree with my formulation.

MR. CREEDON: I am going to re-explain this after he
gets finished. |

{Laughter. ]

MR. COWELL: The population infected at any given time.
is the reciprocal of "1" plus "E" to the minus "a-t." Well, you.

293

 

 
 

 

can see very readily even if you don't understand much about
mathematics that if I just simply depressed this "1" which is not
a very large number, I get "1" over "E" to the minus "a-t" which
is "E" to the "a-t" which is your exponential curve. That is the
reason that during an early stage of the disease, the logistic
curve and the exponential are virtually one in the same thing so
whether people talk about logistic curves or damped logistic or
exponentials or damped exponentials, they are really all talking
about the same sort of thing.

The point I think that Mr. Robertson was making and
upon which I totally agree is when you are down on this end of
the curve, it looks very flat and you may indeed be on the
beginning of what is going to become an exponential curve but
let's suppose that this is the highest risk group and this is a
relatively low risk group, it may take you many, many, many more
years because your "a" value is a lot lower, that is, the
infectivity is a lot lower for this to turn into a classic "S-
shaped" curve. What we are seeing and the very oversimplified
model you see in our report shows really an aggregation of
hundreds and hundreds of sub-groups, some very steep like the San
Francisco City Clinic that went from four to 75 percent in eight
years and others like the heterosexual population that will go
from 0.0001 to 0.0002 in many years and you put all these
tegether and you get this aggregation which is what the cDC is
giving us.

So that may explain some of the problems and maybe
explain what seemed to be a lot of different explanations really
when ‘they are put under the microscope are really essentially all
the same explanation given by different groups of people. I
still, in summary, feel that the CDC data are as reliable as you
are going to get.

MR. CREEDON: One of the questions that we were
considering during these meetings is whether we should suggest
other studies to produce data other than the data that the CDC is
presently gathering so that we would have better data. Now I
think what you are saying is that knowing what we know now, the
CDC estimates are probably as good as you can get but one of the
questions we have is should we try to suggest that other data be
gathered, whether it would be a nationwide test for the virus or
whatever.

MR. COWELL: While I have the microphone still, I will
turn it over to Dr. DeGruttola in a moment, I would say clearly
the answer, Mr. Creedon, to your question is clearly "yes." In
the prologue in the forward of our study, I went back a little
bit in history. In 1662 a gentleman by the name of John Graunt
constructed the first English life tables from the crude records
of births and deaths that he found in Paris churches. These were
the only data then available. When his work was reviewed at its

294

 

 

 
 

 

tercentenary in 1962 Lilienfeld, a modern epidemiologist, |
observed that Graunt did not wait for better statistics. He did
what he could with what was available to him and by so doing, he
also produced a much stronger case for supplying better data and
I believe that that is probably the best that we can do, by doing
the best we can with the statistics that we have make a stronger
case for better data. So certainly, I would say, "yes." Any
effort that you and your Commission can bring to bear to supply
better data will be greatly welcomed, I think, in every
community of public health groups and epidemiologists, doctors
and certainly actuaries in if not across North America and the
entire world.

MR. CREEDON: We would appreciate receiving specific
recommendations from the three of you if you would as to what
better data we might suggest be gathered. Yes, Dr. DeGruttola.

DR. DeGRUTTOLA: When you ask the question, are you
referring only to the CDC because I think actually one of the
most important studies that should be done right now is of
female-to-male transmission.

MR. CREEDON: No, we are not referring only to CDC. We
are referring to just data generally that would be helpful in
trying to determine what is going to happen to this disease and
how we have to get ready for it. |

DR. DeGRUTTOLA: If I were to prioritize, I would say
one of the most important pieces of information that we don't
have is on the nature of the infectivity and the efficiency of
infection through heterosexual transmission. Why that is
important is that if there are long delays between infection and
the onset of infectivity, then you would expect a long period of
time for spread from let's say IV drug users to their
heterosexual partners and then a long period of time before you:
get the next generation which would explain why heterosexual
transmission has not been observed very generally but only very
near where there are IV drug users.

I think that those kinds of studies can be done now.
One of the best sources of such information would be on
heterosexual partners of hemophiliacs because hemophiliacs often
have stored blood so that the times of infection for them can be
estimated which means that the time from infection of the
hemophiliac to let's say infection of the spouse can be measured.

The opportunity to do those studies could be lost as
time goes on because hemophiliacs are fortunately no longer being
infected. So the early part of the tail would be permanently
lost unless it is done now. Dr. James Goedert whose work has
been referred to is currently doing such a study but the sample
size is too small. There is not a large national study in place

295

 

 
 

 

and I think that there “Should be.- Secondly, I..think studying |
female-to-male transmission is extremely important. As I am sure
the Commission has heard, there is a very great range of opinion
right now. You can hear everything from the fact that a self-
sustaining heterosexual epidemic is already underway to: the fact
that it is impossible. I think that the real answer is that we
just don't know enough about female-to-male. We know that
male-to-female is efficient enough to produce a high number of
cases. What we don't know is whether the reverse transmission is
efficient enough for there to be general spread.

Although that would make very little difference in what
we have observed right now, we would observe the same thing right
now whether it was possible to have a lot of heterosexual
transmission in the future or not, it is going to make a great
deal of difference for the future. One thing about interpreting
the incidence of AIDS among heterosexuals right now, although
the incidence is growing rapidly, in fact, among American-born -
women growing more rapidly than it is in other age groups, that
doesn't necessarily imply that there will be a self-sustaining
heterosexual epidemic because if you compare it to homosexual
men, the homosexual epidemic was clearly self-sustaining. Men
infected other people in the same way that they became infected,
through homosexual contact.

That is not true of the heterosexual epidemic. Most of
the cases observed right now are from IV drug users to women. So
it doesn't really tell you whether that is going to, in fact, be
not a total dead end but whether the epidemic will keep
propagating from that point or not and once again, knowing more
about infectivity would be really crucial. I think another very
important piece of information that we don't have is what is
happening among young people and that point has come out several
times but I think the kind of data that could be collected are
let's say from Sentinel high schools providing, of course, that
there was anonymity and protection because obviously you don't
want to stigmatize teenagers but provided that it could be done
without any fear of identification of individuals, I think !
getting information from high schools as well as from neonatal
studies or studies of newborns would provide very important
sources of information.

MR. CREEDON: Do you want ' to add to that, Mr.
Robertson?

MR. ROBERTSON: Yes. All those studies should be done
and they should be done as rapidly as possible. One other
example, of course, is to ao more work on transmission between
spouses after infection by transfusion. There is a fairly large
data base there. We don't know how many people were infected but
it is 20,000 to 30,000, something like that. That would provide
very valuable information. That should be done now in both

296

 

 
 

directions. The data so far are not inconsistent and I can't be
any less mealy-mouth than that, are not inconsistent with
transmission from man-to-woman and woman-to-man being roughly of
the same efficiency. We just don't have enough data to know, but
I think it is important to remember that this would be a rather
unusual sexually transmitted disease if there was a great
difference. We might expect a range of two to one or something
like that, but it would be unusual if it was ten to one. There
aren't other sexually transmitted diseases like that. So the
sort of baseline expectation would be that they were rather
similar.

MR. CREEDON: We had testimony from Dr. Wiley this
morning that suggested it was a ten-to-one difference.

MR. ROBERTSON: I find that very hard to believe and
the data, I don't think, are consistent with that but the
numbers are too sparse.

MR. CREEDON: Yes.

MR. ROBERTSON: It is ten-to-one in terms of total
numbers right now, yes, but that is confusing time with risk.

MR. CREEDON: No, I think he was talking in terms of
the likelihood of transmission.

DR. DeGRUTTOLA: I believe that is correct because I
have seen his presentation as well and I think that those are the
preliminary estimates. The problem is as Mr. Robertson pointed
out, that the numbers are very small there particularly for the
female-to-male transmission. It is based on a study, I think, of
12 male partners of infected women and another small study in
Miami. I would like to second Mr. Robertson's point that the
most important population studied for this question would be
male partners of women infected through transfusion because since
it is removed in general from the IV drug using community, you
have a much better chance of identifying how people were infected
and of getting rates and the numbers are there if you could find
a way to do the study.

MR. ROBERTSON: And we have the time of infection much
more precisely, too.

MS. PULLEN: Mr. Robertson, are you acquainted with the
cDC's plan that they have recently filed for doing a Family of
Surveys and then potentially a national seroprevalence study?

MR. ROBERTSON: Yes, I an.

297

 

 
 

MS. PULLEN: Could you comment, please, on whether you
think that the plan that they have filed is appropriate in the
way they are preparing to go about this and also the time line on
it, whether you believe that it could be done in a shorter period
of time or whether it needs a longer period of time?

MR. ROBERTSON: Yes. In talking about the Family of
Surveys as a whole, there are many individual studies in there
which I think are very valuable that wouldn't be very expensive
like the neonatal one and the Sentinel Hospitals one which should
be done and I think should be expedited. The national random
survey on the time scale which is in the CDC report, I think is
too slow and not very valuable.

MS. PULLEN: Do you think the time on that could be
enhanced?

MR. ROBERTSON: Yes. It is a question of money.
MS. PULLEN: Do you think it is impossible to do it?

MR. ROBERTSON: It is possible to do it. There are
very great difficulties. There is no question about that of
getting a good sample. If it is done, it should be repeated. It
shouldn't just be done once so that we can get both prevalence
and incidence data and if that were done, then even if it weren't
a perfect sample, at least we would be able to see change from
year to year or whatever the sampling period was.

MS. PULLEN: In your contacts with the CDC, did you
provide any input in the design of that?

MR. ROBERTSON: I have discussed these issues with them
and by the way, I would like to say that on the matter of
supplying data, I have always found them very helpful.

MS. PULLEN: Maybe we should call you for the
information.

(Laughter. }

MR. ROBERTSON: Perhaps. Yes, I have discussed the
sampling, the national random sample with people at the CDC and
have had many arguments. My impression is that they really don't
want to do it but that is an impression. It is not a statement
of fact and that they see the difficulties rather than the
information it would give. On the other hand, they are right in
the sense that other samples from their family of surveys if done
rapidly and repeatedly would provide valuable information with
less disruption. But a true random sample would be really nice
to have. I know everybody at this table would like it and
without that, we will be guessing.

298

 

 
 

 

MS. PULLEN: Thank you.
MR. CREEDON: Dr. Crenshaw.

DR. CRENSHAW: I have two questions and one is for Dr.
DeGruttola. I would really appreciate if you would elaborate on
a couple of statements you made in your paper in 1986 that I
found valuable but took me quite a while to understand and I want
to be sure I understand them correctly. At the beginning of an
epidemic a high odds ratio associated with any given behavior
implies only that the disease may be more efficiently transmitted
through that behavior than others. As the epidemic progresses,
however, and virtually all of those in the category of highest

risk have become infected, then less efficient means of
transmission begin to account for an increasingly large
proportion of the total number of cases.

Consider the case in which investigators at an early
phase of the epidemic find an odds ratio or risk ratio of
approximately ten to one, oddly enough, if the organism is
persistent, eventually every member of the lower risk group
could develop the disease. Then you have a statement that
follows, "On the other hand, if a chronic infectious carrier's
state characterizes the disease, it will be subject to
transmission even by infrequent contacts and inefficient
practices." If you elaborate on this, it might help clarify some
of the issues about the focus on spread through vaginal
intercourse, anal intercourse and various others you may wish to
identify.

DR. DeGRUTTOLA: The only point that I was making
there was that if early on in an epidemic you did a case control
study and observed an odds ratio and found that the odds of
having been infected through some means were much higher than
through some other means, whatever they were, it wouldn't
necessarily mean that the less efficient means was going to be
unimportant throughout the entire course of the epidemic.

I think that is fairly well accepted right now. It may
be true that an inefficient, relatively inefficient means like
female-to-male transmission or heterosexual contact or whatever
never will account for a lot of cases. I didn't mean to imply
that inefficient means would ultimately be very important.

I just meant to imply that you couldn't infer from a
case control study early on whether or not they would become
important simply by saying that you had an odds ratio early in
the epidemic of ten-to-one. I think it is intuitively obvious
that as people who are practicing very risky behaviors tend to
become saturated, the less risky behaviors can become more
important as time goes on.

299

 

 
 

I should point out that although it is easy enough to
make a statement like that, when you get down to actually trying
to predict what proportion of people will become infected by a
different means and what the future incidence will be, it is
complicated by the fact that you have people entering and leaving
pools or starting certain practices and ending them. So it is
hard to be much more specific than that.

DR. CRENSHAW: Aren't you also saying that inefficient
practices repeated frequently accumulate and become a rather
significant mode?

DR. DeGRUTTOLA: They could very well even though one
might hear about risks of vaginal intercourse being less than one
in a hundred or less than one in two hundred obviously. If the
risk really is constant over time, they can become significant.
So it is related to the question of infectivity which we still
don't know. If people are really only infectious during short
periods of time, then a very inefficient means of transmission
might not become very important because the effective number of
contacts would be small. On the other hand, if people were
infectious over long periods of time through sexual contact,
let's say, then an inefficient means could become very important.
So what it means is that it is not enough to say this mode of
transmission is inefficient, let's forget about it. You have to
say that we need to be very precise about how efficient it is
which means large studies and we need to be very precise about
what is the nature of infectivity before we conclude that a
given form of contact is or is not safe.

DR. CRENSHAW: Thank you. My other question is open to
anyone who would like to answer it but I found it interesting
that two days apart, November 13th and November 15th, two
different major papers reported the following headlines: "AIDS
up 50 percent in Eight Months Agency Says" and this comes out of
the World Health Organization where they upped their statistics
to revise them upwards from 100,000 to 150,000 people world wide
and two days later, "White House AIDS figures will show
overestimates Slow Down Sources Say." So within two days of each
other, the United States downgraded their estimates and the World
Health Organization upgraded their estimates. Can you help me
reconcile this or are you as lost as I am?

MR. ROBERTSON: I am sure the world wide estimate is
very low, is too low.

DR. CRENSHAW: I would agree.

MR. ROBERTSON: I would say that the estimates in this
country and in some European countries, Australia and so on, are
relatively good. They are not going to be out by a factor of

300

 

 
 

 

more than two, let's say. In Central Africa or in Egypt or in
China, we really have no idea. We know that in some countries,
the disease is very prevalent. Many countries didn't report the
disease at all to the WHO two years ago even though they had
delegates giving talks about it at international meetings. So we
know that there is going to be a continuing dramatic increase in
cases that were there all the time but are only just now being
reported.

It is very unlikely that the number of cases in any
African country is less than the number of cases here using
Central Africa, the AIDS Belt. But again, no individual country,
I think, reports more than 2,000 or 3,000 yet.

MR. CREEDON: In response to Dr. Crenshaw's question,
having studied mortality of a lot of things, not just AIDS, thank
goodness, over most of my professional lifetime, I would make the
observation that the reliability of data from a country is almost
directly proportional to its literacy rate. That is, the more
highly literate the population, the more reliable are any kind of
mortality data, so that the only highly reliable data are those
coming from North American, northwestern Europe or I guess you
could say the western industrialized world as we know it, and
that would be true of any kind of data, AIDS or any other kind of
mortality or morbidity. So I totally concur in Mr. Robertson's
observation that any estimates based on other than what we call
the western world where data are freely exchanged and are not
considered a state secret or part of the national pride should be
taken very, very cautiously. Now, as to estimates within in the
United States, and there has been over the last two months a
very wide range of numbers that have been thrown around, again, I
would revert to the most solid being those coming from the cDc.

MR. ROBERTSON: I've actually read the way in which the
White House estimate was prepared, and whatever the truth of the
actual number, the assumptions on which it was based, the two
most important ones were that the Army data showed no increase,
no new incidence. I think nobody here agrees with that. The
second assumption was that, given that the Army data showed no
new incidence, the infection stopped two years ago. So whatever
the value of the actual figure, it might by coincidence be right.
I don't think the method of estimation was useful at all.

DR. CRENSHAW: Thank you for clarifying that.
MR. COWELL: Just to put my number in, I think it's got
to be at least a million, and I'm higher than these two

gentlemen, but, really, I don't think it's more than 3 million.
I think it's more like 1.5 to 2.5 million. We really don't know.

301

 

 
 

 

DR. PRIMM: The population sampled by the Army or by
the military, the population is a cream population. It would =.
Seem to me that it should be going down in that population rather
than even to remain stable.

MR. ROBERTSON: Indeed, there's evidence of self-
deferral; but the part of the Army population in which the
incidence or the apparent prevalence went down was the better
educated white older male potential recruit who, clearly, is
self-deferring and no doubt tests himself before he goes.

DR. CRENSHAW: Mr. Robertson, it was explained to me
that what appears like a decrease is actually an increase because
it's like a college. If you keep saying that your rates of
gonorrhea in your entering freshmen are somewhat decreased, then
it's assumed that these people are actually lower in numbers
rather than a new population that you're measuring. I'm not
making myself very clear. I hope you know what I'm getting at.
Can you make it clear so that --

MR. ROBERTSON: . Yes. I'll give you a few examples, and
Dr. DeGruttola, I'm sure, will, too. Let's take the blood
banks. Now, when we first started -- when the blood banks were
first tested, I can't remember the actual numbers but something
like 4 or 5 per 10,000 was the apparent prevalence in that
population of people. Well, people who donate blood in general
' do it many times. They are repeat donors. So as testing was
done, two things happened. First of all, people who might think
that they were at risk. for being infected finally got tested and
were informed if they were and undoubtedly deferred, and their
friends and colleagues would have deferred.

The group that was infected was excluded very rapidly,
so the numbers which we're getting now, though smaller, don't
imply that the prevalence in that population of blood donors has
decreased at all. All they're doing is giving you a filtered
measurement of the current incidence in that group of people. We
don't really know whether it's a monthly incidence, an annual
incidence or what, but we do know that it shows increasing
infection, even though it appears to show a decrease in total
prevalence, the total number. The Army data is rather like that.
There are many forces which tend to diminish the actual raw
numbers coming in. Yet, if you sought the Army data, for
example, by age cohort, you find that there is a new incidence.
In fact, it's quite rapid, quite a rapid growth in some age
groups. But, again, that's probably a poorer --

DR. CRENSHAW: Can you be specific?

MR. ROBERTSON: Well, if you look at people who are 20
this year and who were 19 last year and compare them with the
people who were 19 last year, you'll find there's a higher

302

 

 
 

 

 

infection rate amongst the 20 year olds this year who would have
been from the same group as the 19 year olds who applied last
year. The same is found in people actually in the Army who were
tested and found to be negative but then have been retested
because they had some medical problem or were going overseas.
Again, there's a current incidence amongst those people, even
though if you take the recruit data as a whole, it appears that
the prevalence is constant.

MR. CREEDON: Dr. Lee.

DR. LEE: I want to commend this panel. It's really an
extraordinary hour or two we've had from you. All your curves,
as far as I can understand them, show a plateau. They are
plateauing; yet, the numbers we very commonly are exposed to
sort of continue to exponentially rise. I mean, it just sort of
goes like this (indicating). Now, from just life experience, one
would think it's not going to do this, that something is going to
happen. Now, all of your curves plateau. Do you think this
disease is going to plateau? Do you -- why are Mr. Cowell's
numbers continuing to go up when your curve was plateauing like
that? I'd like to hear from anyone on that.

MR. ROBERTSON: His numbers are going up because they
are the integration and the sum of every year. So even if the
curve plateaus eventually, he's still adding up each annual
component to get his total.

MR. CREEDON: Cumulative; he has a cumulative.
MR. ROBERTSON: Yes.
DR. LEE: So you were all plateauing?

MR. ROBERTSON: Well, we know that the curve will
plateau at 250 million at the very worst.

{Laughter. ]
DR. LEE: But those'curves have already plateaued.

MR. ROBERTSON: The question is where will it plateau,
of course. Any epidemic must plateau. There are various very
important issues. For an epidemic to be sustained, each person
infected must in his lifetime infect somebody else. We know
that's theoretically possible for various risk groups with the
HIV epidemic, and it's certainly possible for some heterosexual
risk groups. It's clearly impossible unless the virus finds new
modes of transmission for the monogamous couple who aren't
infected now. But somewhere in between we will plateau, and
none of the models we have can really tell us. We know in Africa
it's going to be quite high just from raw data.

303

 

 
 

 

MR. CREEDON: Any other questions from the panel?
MR. ROBERTSON: That was in the vernacular, too.
MR. CREEDON: Mr. Chairman?

DR. DeGRUTTOLA: I would second that. I think you can
go even stronger and say that by modeling, you can show that you
cannot know from current data when and at what level it will
plateau. The fact that the doubling time is increasing, that it
is plateauing, is a feature of any infectious disease epidemic.
The only way you would see an epidemic where that didn't happen
is if it kept moving from a lower risk to higher risk groups; in
other words, if it kept moving from less promiscuous to more
promiscuous individuals. But clearly, that can't go on for a
very long period of time. So the fact that it's slowing down
doesn't really give you very much information about when and at
what level.

One point I would like to reiterate is that in my
estimate of prevalence, my range would go from -- basing it on
information that's available, it would go from half a million to
2 million or even above 2 million. I don't think that we can be
more precise than that. If I'm asked for my own subjective
opinion based on a lot of sources of information that we've
talked about today which are not precise enough to give an
estimate that I think is absolute, but my own subjective opinion
is that it's probably around a million or slightly below a
million, but I would not rule out higher numbers, and I would say
that it depends on much more study of the dynamics of the
epidemic.

DR. PRIMM: I just want to ask one question just for my
own edification. That is, you mentioned you would study the
newborn. What would that yield be for you? Because I can't see
it.

DR. DeGRUTTOLA: What information would it give you?
DR. PRIMM: Yes.

DR. DeGRUTTOLA: Well, it would give you information
about prevalence of infection of women at childbearing age. That
could be very useful for finding out, first of all, to try and
estimate the age in which they became infected. Now, you only
know -- you don't know the age at which they became infected.

You simply know an age at which they are infected; but by some
kind of modeling procedures you may be able to work back to
estimate the age at which they're becoming infected. It would
also give you some sense of what the prevalence is for a group
right now for which prevalence is very difficult to estimate

304

 

 
 

 

because the latency is long and we haven't seen many cases of
AIDS yet and we wouldn't expect to see many cases of AIDS yet
among those women. So it would give you much more precise
information; in a sense, a kind of leading indicator, to borrow a
term from economics, on where the epidemic is headed.

DR. PRIMM: We get the same information just from
testing the mother, though.

DR. DeGRUTTOLA: You could get it from testing the
mother as well. The reason that people talk about doing
neonates is that there's already neonatal screening for PKU and
in Massachusetts, for example, the bloods are already being
collected. The added expense for doing the study was really
very, very small.

DR. PRIMM: Thank you.

MR. COWELL: I have drawn another picture, and this one
isn't nearly as complicated as the last. The question was asked
about a change in the prevalence of HIV antibody in military
recruits. Now, I'll hold up this chart. This is from the CDC
data, and I have simply reproduced this on a little larger scale.
Here is the actual pattern of the incidence among new military
recruits. This is the percentage testing HIV positive for the
last -- since they've been testing HIV positives. This is ona
very large sample. It's a total of over a million, about
1,075,000. This is the aggregate prevalence rate. You can Say,
well, it's declined, but the decline is somewhat slight. You can
see it jumps up and down every month. These are monthly
intervals, so there is some decline. The lower level is the rate
among whites, and the rate among blacks and Hispanics is
somewhat higher. They show the same parallel trends. While
we're on the subject of military recruits, if your panel, ladies
and gentlemen, has not already heard from Dr. Robert Redfield of
the Walter Reed Army Institute --

DR. PRIMM: We did, yes.

MR. COWELL: You have, fine. I would say that we had
Dr. Redfield at our Society of Actuaries Panel in Montreal, and I
will say now what I said then. I thought in one-half hour he
gave us more information on the impact of this disease in
speaking as a doctor who was on the ward. He was telling us, I
think, with a lot more conviction than most of us had heard in 18
months. Of course, you then are familiar with his thesis of the
spread is a function of the promiscuity of the group. Getting
back to the question Dr. Lee raised about where these curves will
plateau out, again I have to echo everything that's been said
here. We really don't know, but the model that is demonstrated
in our report suggests that -- again, I don't have a time frame
on here, but if this is the late '70s when the epidemic entered

305

 

 
 

the population and this, let's say, is around today, somewhere
here, that somewhere between there on into the future, that curve
will flatten out.

Now, if the epidemic stops spreading today, if
everybody with the disease stopped doing what it was they were
doing to spread it, then the thing would flatten out at about a
million, and I won't try to give it any more precision than that.
Our particular model shows it growing. This obviously is not to
scale, but the charts in our report are to scale. It suggest
that it might peak out at around 2.5 million sometime around the
late ‘90s. So even I predict an eventual peaking out on the
assumption that there is not a massive spread into the
heterosexual population. I think the numbers that we have seen
to date and I think that we're all in concurrence with why we
feel or why I feel this is so. Out of my estimate of about a
million, somewhere between half and three-quarters of those are
from the highest risk group, the male homosexual/bisexual
population, and the highest of the high risk group, the highly
promiscuous male homosexual population numbers at most 2.5
million. Yet, that has generated maybe 500,000 or 1 million or
one-half of the total number of HIV infections.

Contrast that to the population 50 times its size, the
heterosexual population of 140 million adults that probably have,
at most, 50,000 to 100,000. So you have a population that. is
1/50th the size of the adult heterosexual population, that has
probably generated 10 times as many cases. In other words, the
incidence is 500 times in the highest risk groups what it is in
the heterosexual population. That tells you the kind of problem
that our society would be faced with if public education and
public health measures to stop the spread of this disease were
not being taken.

I guess I have some difficulty in discussions where --
and I speak as an actuary who loves precision in numbers, but
from my perspective and I'm speaking now not as an actuary but as
a concerned citizen, it's difficult for me to understand why our
public response to this would be vastly different if the number
were 2 million than if it were 1 million. In either event, there
is a sufficiently large pool of infected there to create problems
for all of us at every level of society.

MR. CREEDON: Thank you, Mr. Cowell.

ADMIRAL WATKINS: Let me close out the questioning just
with one observation. We've heard a great deal about
mathematical modeling and bloodletting to obtain really good
seroprevalence data, the latter having some difficulties in
getting true data across the country for a variety of reasons.
We've heard very little about any imaginative efforts to look at
analogues, a family of surrogate studies, for example, analyses

306

 

 
 

 

that seems to me might be an interim approach to try to do some
validation of data. We had experience in the military,
significant experience, where initially rejected behavioral
analyses came in on drug abuse, only to find out they were
accurate within a few percentile points of the urinalysis data we

received.

It just seems to me that while we're going through the
trauma of dealing with confidentiality and discrimination and
other things that may be a bar to openness on the part of certain
individuals in any kind of a seroprevalence survey that's run,
that there ought to be other kinds of approaches taken. For
example, we heard presentations here on the direct correlation in
rectal gonorrhea with the homosexual high risk group and
AIDS-related activities in the San Francisco area. Does that
pertain elsewhere, and is that a possible surrogate? Is there an
approach that we could take on a behavioral analysis, for
example, relative to the military proclivity to come in that
would give us some kind of a factor that says we're going to weed
out those that will not apply for the military for the variety of
reasons and run a behavioral anonymous survey in metropolitan New
York or other large areas where we have good data on the
proclivity of people to come in and their HIV infection rate.

It seems to me those kind of things should also be
ongoing in an integrated way and continue the validation of your
modeling while we're waiting to solve some other very serious
problems that we have in getting on with the kind of hard data
that would make all of you happy, as you mentioned on the panel
today. So what is the opinion of you analysts about doing some
alternative family of surrogate surveys that may get you closer?

MR. ROBERTSON: We certainly want to do surrogate
surveys for sexually transmitted diseases, but the example you
gave, the rectal gonorrhea one, shows how dangerous it can be,
and I'll explain that. Gonorrhea is a disease which can be
cured chemically, and it has a very short infectious period, a
matter of days. So as far as passage goes, it's more like a cold
than like AIDS. In other words, if somebody with gonorrhea wants
to avoid infecting somebody else, he only has to keep apart for a
couple of weeks and be cured.

ADMIRAL WATKINS: I'm not talking about the disease.
I'm talking about the correlation of the rate decline, the
incidence rate on --

MR. ROBERTSON: I haven't finished my answer.
Therefore, if you use the rate of gonorrhea as a surrogate for
behaviors that might give you AIDS, you'll be misled. While that
rate of infection was declining drastically in the city clinic
cohort for gonorrhea, the rate of new AIDS infection was going up
equally fast because of the time courses, the two diseases being

307

 

 
 

 

so different. But, yes, in general, I think the more surrogates
that one could find, the better, but we don't really have enough
information yet to know what they are.

ADMIRAL WATKINS: How about behavioral surveys? Why
are they so inaccurate or why aren't they being used more
aggressively?

DR. DeGRUTTOLA: I think that's a good question. I
think that we need much more information. Clearly, if you want
to do any kind of modeling, we need much more information about
standard sexual practice. The kind of logistic formulations that
Mr. Cowell presented are very appropriate in certain kinds of
settings, but they're not appropriate for all settings. I think
that we'd all be in agreement, we need much more information in
order to get more precision on those curves, what the shape of
those curves should be. I think that it's difficult to do such
surveys. The one place where they have been done is San
Francisco, where there's a population-based survey under way
which is collecting information about behavior as well as
infection with HIV and other viruses. The problem is that San
Francisco isn't generalizable to the rest of the country, and
it's harder to do those surveys in other places.

ADMIRAL WATKINS: Do you mean they have tried and
failed?

MR. ROBERTSON: I don't know --

ADMIRAL WATKINS: We haven't had any presentations to
that.

MR. ROBERTSON: No. I believe the only population-
based survey where they asked information about sexual activity
is in San Francisco. It's the only one to my knowledge.

ADMIRAL WATKINS: Is that good, or would you like to
see more; and if so, how do you inspire those? What do we have
to do as a commission to inspire those, and are they technically
useful to the analysts to try to find a better bridge between
modeling and reality?

DR. DeGRUTTOLA: They would be very useful. Clearly,
what is limiting all of us is not knowing what the sizes of these
populations are. If we did know the size of the IV drug using
and the homosexually active population and something about how
they mix, we could do much better in our modeling than we've
done today.

ADMIRAL WATKINS: What recommendation would you make to
the Commission?

308

 

 
 

 

DR. DeGRUTTOLA: There would be two. One is that the
information provided by demographers, which is only on single
men, marital rates and in-migration and out-migration, can be
very useful. It was useful in San Francisco to get that one
estimate of the size of the homosexually active population and
useful to show why the ratio of prevalence into incidence must be
much lower than had otherwise been thought. So demographic
analysis can provide us with important information cheaply. Now,
the more expensive way is to take that information and use it to
develop population-based surveys to ask questions about sexual
behavior. I think that's going to be crucial if we're ever
going to do much better in terms of precision of modeling, and I
think it would be a great recommendation to come from the
Commission, just recognizing that they are very difficult surveys
to do. For a variety of reasons, I think it's hard to find
people who have the competence and the willingness to do then.

DR. CRENSHAW: Could I underscore a point? I am
totally in favor of behavioral tracks. I think this is where
we're lacking the most information of all, and we need to do
effective research there. But it is exceedingly labor intensive
and cost intensive, and there isn't the manpower. Fewer than 10
percent of physicians, according to surveys, can even take a
sexual history, much less a competent one. Nonphysicians have -
even lower aptitudes and abilities. So before you could even do
that, you'd have to train a large manpower, personpower,
womanpower to take an adequate sexual history. It's delicate;
it's complicated. It can be done. We don't have the resources
in place. We should get then.

MR. COWELL: For specific recommendations, Admiral
Watkins, I would just endorse what has been said and would just.
extend it to say that I think we do have agencies that are in
place whose responsibility it is to do what you're asking;
namely, the CDC. The National Academy of Sciences, which is just
a few blocks from here, their institute of medicine did a study
that they published last year confronting AIDS. That described
the magnitude of these at-risk groups, the size of them, and that
was the one that we used for our report.

The University of Frankfurt in West Germany is
continuing to develop studies of progression rates from HIV
infection to AIDS under very carefully controlled clinical
conditions. My recommendation would be to assemble
representatives from each of these groups together and have an
organization that was able to do on a massive scale what we in
the Society of Actuaries and the American Council of Life
Insurance are trying to do in a sort of microcosmic sense,
looking at only the insured population, to do this on a massive
North American scale or U.S. scale and to put together a model
that reflects all of these things, including transmission factors
from groups so that it would, indeed, be a model that people

309

 

 
 

could identify with as not being just a bunch of theory but,
really, that reasonably closely represented reality.

As I say, we are attempting to do this to our model.
By the end of April, we will have another iteration, if you will,
of our model, and we've got a lot of people working on it in the
United States and Canada. I'm reasonably confident that the work
of the Society of Actuaries and the ACLI is going to produce as
believable a model in 1988 as will exist. I just encourage
people who are already working in this area seriously to continue
their efforts.

ADMIRAL WATKINS: Thank you, Mr. Cowell. Before we
close here, I want to make sure the panelists are aware that we
would like to keep the dialogue open with every panelist that has
come before this Commission. There will be additional
questions. We've had some already prepared by the Commissioners
today to get back to you expeditiously. We're asking the
Commissioners here today to please get their additional
questions in for any of the last two days of panelists’
presentations not later than next Wednesday. We do not want to
keep stumbling over ourselves as we move to the next important
hearing on drug abuse and want to move these questions
expeditiously to the panel members. So please get them in by
next Wednesday. We will include them as they are written to each
of the panelists, either all of you or singulary as the
Commissioners so desire. If you would be willing to get back
within a couple of weeks to us, it would be very helpful in this
area. So as we close this hearing today, I'd like to take a few
minutes to make some brief remarks about what I feel has been an
extremely beneficial exercise to date by this Commission.

We've moved very quickly in a relatively short period
of time to establish our agenda and then to lay out the
groundwork for accomplishing our task. These particular two
days of hearings provided us with a substantial amount of
information on an issue we have all agreed is one of our
near-term priorities: determining the incidence and prevalence
of the HIV infection in the U.S. population. While to some that
may not be important, to us, we must have it to carry out our
task to the President, which requires a lot of projections on
health care, delivery projections, budgeting and so forth.

So it is germane, and we must at least establish a
baseline that is clear when we move in with our final report in
June and, hopefully are moving in a direction to sharpen the
baseline of data at that same time. A meaningful and properly
integrated response to this epidemic cannot be mounted if we do
not have this solid baseline. It's incumbent upon us to search
out all of our national resources, experts in all the fields
relating to epidemiology, and find out what obstacles stand in
the way of our ability to state with considerable accuracy how

310

 

 
 

 

many of our own citizens are currently infected with this disease
and how many we can expect to be infected in the future. It's
my feeling that we have made some headway, and this was enhanced
by these two days of hearings. The testimony that we've heard
here has been from a diverse and respected group of witnesses and
has hopefully provoked some new thoughts and insight into this
very important issue. We'll be following up with more specific
questions, as I mentioned, and we hope to have the opportunity to
meet further with some of you. We must find a solution to the
problem. We know that there are no easy answers to many of the
challenges that confront us, but we have already committed
ourselves to providing our President with the best answers
available. This is our charge, and I believe we've taken a giant
step forward to this end.

Next week, we'll be continuing to move ahead with our
agenda and under the leadership of Commissioner Dr. Primm we will
be holding hearings on the multitude of issues related to IV drug
abuse. We have a very impressive group of witnesses coming
before this Commission at that time. So I want to thank you for
your commitment and support. Your presentations have been
extremely professional and informative today. You're a very
impressive set of witnesses, and we've learned a great deal.
Thank you for coming. With that, I will adjourn the Commission
for this set of hearings and will look forward to next week's
hearings. Thank you very much.

(Whereupon, at 4:40 p.m., the hearing in the above
entitled matter adjourned. ]

311

 

 
 

 

APPENDIX

 

 
 

THE CLINICAL SPECTRUM OF INFECTION WITH THE HUMAN
IMMUNODEFICIENCY VIRUS: FROM ASYMPTOMATIC INFECTION
TO THE DEVELOPMENT OF AIDS

Statement of
Anthony S. Fauci, M.D.
Director
National Institute of Allergy and Infectious Diseases
National Institutes of Health

and

AIDS Research Coordinator
National Institutes of Health

presented before

THE PRESIDENTIAL COMMISSION ON THE
HUMAN IMMUNODEFICIENCY VIRUS EPIDEMIC

December 10, 1987

 

 
 

The common denominator of infection with the human immunodeficiency
virus (HIV) is the destruction of the body’s specific defense mechanisms,
namely, the immune system. The immune system is responsible for protecting
the body against microbial invaders as well as the development of certain
types of neoplasms. When the immme system is compromised for whatever
, Feason, depending on the extent of the immune defect, the body is
susceptible to serious illness caused by microorganisms which would not
usually be able to gain a foothold in the body. However, because of the
weakened state of the body’s defenses, these microorganisms seize the
opportunity to invade; hence, they are referred to as “opportunistic”
infections. The same holds true for the development of unusual or
opportunistic neoplasms which are usually kept in check by an intact immune
system. The immune defect that is ultimately caused by HIV infection is so
profound that opportunistic infections and neoplasms occur readily and
repeatedly, constituting the condition known as the acquired

immunodeficiency syndrome (AIDS).

HIV enters the body either as cell-free or cell-associated virus and
binds specifically to cells in the body expressing the CD4 molecule, which
is the cellular receptor for the virus. This molecule is heavily expressed
on human T4 lymphocytes which comprise the helper/inducer subset of cells
responsible for the induction of most, if not all, of the functions of the
human immune system. If there is one cell type which is critical to the
effective orchestration of the entire immmne system, it is the T4
lymphocyte. Its selective destruction can thus explain the global nature of

the immune defect which occurs with HIV infection. Other cells, such as

 

 
 

 

those of the monocyte/macrophage lineage and certain cells of the central
nervous system, may also express variable amounts of the CD4 molecule and
thus may be susceptible to infection with HIV.

Upon entering the body, the virus can exist either in a latent form or
in a form which is actively replicative. Scientists are beginning to
' delineate the mechanisns whereby a cell that is latently infected with HIV
and appears normal can be induced to actively produce virus and ultimately
be destroyed. This process may take place gradually or in intermittent
spurts over a prolonged period of time which may explain how a person can be
infected with HIV for years before the virus destroys enough of the CD4-
positive lymphocytes to result in clinically significant immunosuppression

and disease.

It is important to appreciate and distinguish the differences. among the
various phases of HIV infection. The first phase is asymptomatic. infection
in which virus is present in the body but the host has no symptoms. The
virus can be transmitted from one individual to another during this
asymptomatic phase as well as when the person has symptoms and/or disease.
The second phase is the phase of infection in which the individual shows
some signs and symptoms but does not meet the empiric criteria for the
diagnosis of AIDS that have been established by the Centers for Disease
Control (CDC). This type of symptomatic infection with HIV is generally
referred to as the AIDS-Related Complex (ARC) and may be manifested as
fever, weight loss, diarrhea, fatigue, night sweats, lymphadenopathy, and/or

variable degrees of immunologic abnormalities. Finally, those individuals

 

 
 

meeting the CDC criteria for the diagnosis of AIDS comprise the third

category of infection with HIV -- full blown AIDS.

As of early December, 1987, approximately 47,000 people in the United
States have been diagnosed with AIDS. The analogy of an iceberg is often
used to explain the spectrum of infection with HIV with the tip of the
' iceberg representing the 47,000 people who have now developed full-blown
AIDS. In addition, approximately 150,000 people are symptomatically
infected -- have ARC or persistent generalized lymphadenopathy. Finally,
between 1 and 1.5 million infected individuals are completely asymptomatic.
At this time, we know that approximately 20 - 30% of individuals who are
infected and asymptomatic will develop full-blown AIDS within five years.
Whether this pattern will continue in a linear fashion or whether it will
plateau is currently unclear. In other words, it is uncertain whether or
not after twenty, thirty, or forty years, virtually everyone who is infected
with HIV will develop full-blown AIDS if effective treatment is not
developed. On the other hand, the relative proportion of infected people
who develop full-blown AIDS may level off over a period of time. Recent
evidence from the Walter Reed Army Medical Center indicates that if the
immunologic parameters of HIV-infected individuals are followed over time,
80 - 90% will demonstrate some level of deterioration of immunological
function over a few years. This suggests that the vast majority of infected

individuals will be adversely affected by the virus over time.

A more recent CDC classification scheme for HIV infection divides the

infection into four major groups. Group I, acute infection, encompasses

 

 
 

 

-

those individuals who, shortly following infection with HIV, develop an
acute viral-like illness. Most often, this illness resolves spontaneously
followed by the appearance of antibodies to the virus. These individuals
may then remain asymptomatic for years or may ultimately go on to develop
symptoms. Group II includes individuals who are infected with the virus but
show no symptoms of disease. Group III, persistent generalized
lymphadenopathy, includes relatively asymptomatic persons who carry the
virus but manifest generalized lymphadenopathy, which persists. Finally,
Group IV is given the categorization of "Other Diseases." This group is
divided into several subgroups. The first subgroup (A) is constitutional
disease, which is also known as ARC. Subgroup 5 denotes the presence of
neurologic disease in the absence of either secondary infection or neoplasm.
This category was only recently added when it was determined that HIV could
directly infect the nervous system. Subgroup C is full-blown AIDS with
secondary opportunistic infection. Subgroup D is AIDS with secondary
cancers such as Kaposi’s sarcoma or B-cell lymphoma. Finally, subgroup E
includes other conditions not classifiable above that may be attributed to
HIV infection or that may be indicative of a defect in cell-mediated

immunity.

There are three major components of full-blown AIDS: opportunistic
infections, unusual neoplasms, and central nervous system disease. The most
common of the opportunistic infectious agents is Pneumocystis carinii, which
causes pneumonia. However, unusual microorganisms such as Mycobacterium

avium-intracellulare, Cryptococeug neoformans, disseminated Cytomegalovirus,
Toxoplasma gondii, and others are also important causes of serious

 

 

 
 

infections in full-blown AIDS. The mean survival time for an individual
presenting with the first episode of an opportunistic infection is between

30 - 40 weeks.

The next important category of full-blown AIDS is unusual neoplasms.
The most common of these is Kaposi’s sarcoma. The immunologic function of
individuals with AIDS manifested by Kaposi's sarcoma is not nearly as
severely impaired as that of individuals who develop opportunistic
infections. This underlies the fact that the survival for individuals who
develop Kaposi's sarcoma is usually longer than those who present with

' Opportunistic infection, and may be as long as a few years.

The final major category of AIDS is infection of the human brain with
HIV which results in conditions from asymptomatic infection to subtle
cognitive and psychiatric abnormalities, frank dementia,
meningoencephalitis, and localized or diffuse neurologic abnormalities. It
is estimated that greater than 30% of individuals with AIDS will develop
clinically significant central nervous system disease directly related to
infection with the virus. In addition, it has now been well established
that virus can be isolated from the cerebrospinal fluid of greater than 503%

of infected individuals, even among those who are asymptomatic.

A recent report from New York City established the survival rate in
AIDS at approximately 49% one year following diagnosis and approximately 15%
five years following diagnosis. Clearly, individuals who present with

Kaposi’s sarcoma alone have a more favorable survival rate than individuals

 

 
 

 

who present with opportunistic infection. As of November 23, 1987, the

cumulative case fatality rate was 57%. However, this includes individuals
who have been only very recently diagnosed with AIDS. Examining the total
number of individuals diagnosed with AIDS since 1981, between 80 - 90% of

those diagnosed in 1981 and 1982 have died.

Thus, the clinical spectrum of AIDS ranges from an asymptomatic to a
symptomatic state to a state of full-blown AIDS. Understanding the
mechanisms responsible for the progression from a latently infected state to
one of active viral production and ultimate immunosuppression will have
major implications in the design of treatment strategies for the virus

itself as well as for the reconstitution of the immune response.

 

 
 

 

TESTIMONY OF DONALD P. FRANCIS, M.D., D.Sc.
DECEMBER 10, 1987

First, I would like to thank you for allowing me to testify before
this important committee.

Second, I want to make it clear that what testimony I give here is
my own opinion and does not necessarily represent the Opinions of
the Centers for Disease Control or the California Department of
Health Services.

And last, I would like to express my admiration of the committee
for its Interim Report. The recognition of the need for more
societal commitment was refreshing. But, the statement,

"that too much time has elapsed and too many people have

become afflicted while questions remain unanswered." was
spectacular. And the "call for collective dedication" was
magnificent.

Since I am sure that many in the next two days testimony will fill
in many of the details of AIDS prevention, including prevalence
and incidence estimating, I, as a young but well bruised veteran
of serious epidemics, would like, if I may, give some insights of
a broader nature.

The overall gist of what I will say is that: 1) we know a great
deal about HIV, its transmission, and how to prevent it; 2) we in
public health want to and should move ahead with a scientifically
designed aggressive prevention program; 3) there are some
obstacles that are inhibiting that movement; and 4) most of these
obstacles are readily removable.

First, let me outline what we know:
1. HIV is a very virulent virus, perhaps the most dangerous
virus of man.
2. It has a long and silent incubation period during which
time infected persons remain infectious.
3. It is transmitted exclusively through sexual intercourse
(both heterosexual and homosexual), through Sharing of blood
(especially between intravenous drug users), and from
infected mothers to their infants.

The combination of high virulence, long, silent incubation period,
and sexual transmission make this a virus of major concern for
those of us having experience with these situations.

Indeed evidence to date indicates that the virus has already
extensively invaded the United States and, regardless of what
estimate one takes, we have a major tragedy on our hands.

The good news is that we as a society and as individuals in the
society have it in our power to stop the transmission of HIV
today. Preliminary evidence from both the homosexual male

 

 
 

 

 

community and the intravenous drug using community suggest
strongly that well designed intervention programs can effectively
reduce the transmission of HIV and, as a result, the rates of
AIDS.

Given this information a broad consensus has been reached in the
pubic health sector and reports from the Surgeon General, the
National Academy of Science, the Association of State Health
Officers and the Conference of State and Territorial
Epidemiologists have all agreed on the approach to be taken.

Yet, despite this major problem and the consensus on what needs to
be done there is, at best, only a skeleton of an AIDS prevention
program in the field.

Why is this? I see several obstacles to fielding a successful
AIDS prevention program:
The first is the perception that no one really cares. Using
your own words, it appears that instead of having a solid
prevention program based on modern science, we have, etresteck
wf a fragmented program peeet—om prejudice and fear.
SOMO rian Con Sasees by
Second is the confusion centering around the question of
whether AIDS prevention should be a police action of the
government against the people or a cooperative endeavor of
the people with their government.

The consequences of these obstacles have been absence of necessary
resources and effective leadership, slow delivery of owen basic
prevention, and an absence of the usual positive American can-do
Spirit.

We can, and should, rebuild that spirit, use modern science to
generate policy, supply the resources and quickly mount an
aggressive progran.

We, in the public health sector can do that. We want to. I am
confident that given the spirit, the leadership, and the resources
we could launch a AIDS prevention program which could severely
inhibit the further extension of this virus into the American
public.

It is now six and a half years since the discovery of AIDS. It is
time to act maturely, ignore all of the peripheral distractions,
and mount an effective AIDS prevention program.

 

 
 

 

REFORT TO THE FRESIDENTIAL COMMISSION

ON THE HUMAN IMMUNODEFICIENCY VIRUS EPIDEMIC

PREPARED BY:STEVEN L. SIVAK, M.D. and GARY F. WOSMSER, ".D.
DEPARTMENT OF MEDICINE,
SECTION OF GENERAL INTERNAL MEDICINE
ana
DIVISION OF INFECTIOUS DISEASES
NEW YORE MEDICAL COLLEGE

VALHALLA, MW. Y. 195895

Since widespread testing for human immunodqeficiency
Yarce CHIM) infection as not performed in large groure of
Anericans at different risk for HIV anfection and since thre
conmorts wnich have been tested may not accurately represent
the groupe to which they belong, it is impossibie to
dererwir precise figures for the s-evelence of HIV intectiun
rm the United States. This matter is compounded by the “face
that HIV anfect:on, except in the form of the Acquired
imnunodeticiency Syndrome .AIDS) is not a reportable
condition. Furthermore, obtaining exact figures as to the
total population of risk group members is difficult.
Determination of prevalence must therefore at this «ime

he considered an estimate based upon inferences from

twlemtially biased data,

 

 
 

 

Using the numoers of the known cases of SIDS based upon
sucmveillance cata collected by the Centers for Disease
Contreal (CDC) as well as reported data regarding prevalence
of HIV infection in several risk groups, we believe that one
cam estimate the prevalence of HIV infection in the general

L.S. population as well as in identified subgroups.

The critical assumption in our calculations is thet tre
ratio of the nuuter of living patients with AIDE te the
number of iandivicuals infected with HIV 1s constanc withir:
coifferent subqraups. We believe that this assumotion
we re@eveconable and has been used by others Cid. Sznce the
epidemic of HIV int-ecticon in this country 1s probably in a
dymemic slate, we chose to wee the ratic employirag tre miumber
qf cuerentiy Jaivang petientis as orpesecd te the tetai iuember
ey nmieiremte with AILS c:nce this mav gave a more aeturate
Crease e2ttion avueravimation af ianformatien relevant a. a
Toy Tachudb ar pmark aim came, am LYSE we demansur ated. weang tne
bits’, gvallabie anformation regarding the prevalerce or 41?
rnrectson im two well stuoied groups as well as the
eurveiliance data from the CDC, that for everv imtravenocus

reg abucer CIVDA) or hemophiliac alive with AIDS at that
Lime there were approximately 200 IVDA and Foc
rFeranhiliacs, respectively infected with the virus. We
applied the 1:2300 ratio to each of the then identitied rigs

groupe and obtained a total estimate of apreresaimately 41.7

Madlivws Anericans anfected with HIV of which eboaul 64,90)

 

 
 

-

individuals were at no known risk (see table). Usine census
data for adults in the United States and subtracting
approximate numbers of known high risk group members we
estimated & prevalence of HIV infection for U.S. adults at no
kriown risk to be 45/100,000. At the same time, the United
States Red Cross reperted that the prevalence of HIV
infection among blood donors at presumably no known ris! for
HIN anfection wes 78/100,000. We felt that since the ratics
of those alive with AIDS to those infected in two weli
sludied groups were similar and that an estimated U.S. adulic
prevalence af thuece et na known risi was similar to thet
eituaily een, that our assumptions ana data were reesonaoly

Pace ete, These results were reported im November 1985 (24.

Other pvetimates for that same period of time of the

yee alence Of si] infection range from 500,020 to greater
Chat 2,009,000 CL ,0]. We recently re-examined the cata in a
tiiiiar fa5hicn. We again determined the estimated

prevalence of infection in high risk groups where data was
available aim an attempt to obtain the current ratio of those
alive with AIDS to those infected. It is clear that this
figure if subject to change due to many potential factors.
As those at high risk modify their risky behavior, the
prevalence of infection in that risk group is likely to
change. As more infected individuals develop AIDS and
~wneofar as the survival may be different mow thar it wee in

the pact, this azgpect of the ratic may change at well, In

 

 
 

one well studied cohort of homose::ual men the reported ratio
of those with AIDS to those infected chanced from 1:825 to
1:28 over a four year period C1]. Another study revealed in
1982 that the ratio of those with AIDS to those infected with
the virus was 1:30 [4]. This group of homosexual men
attended a sexually trarsmitted disease clinic and may not
have been representative of e111 homosexual men because of a
sotential bias towerds individuals with early manifestations

of IDS.

At this time we chose to determine the ratio in question
in well studied pepuiations of New York City. Here
reasonably accurate figures for prevalence of HIV infection
ei1st and estimates of the total numbers of members af high

rigélt populatione are available.

Tt 25 estimated that in New York City there are up to
TOO hamesexual men cf whom approximately fifty per cent
er Th0,00 are infected with HIV (5,63. We obtained data
from tne New York City Department of Health AIDS Surveiilance
lindate of 10/78/87, which reports the cummulative casese of
S:DS in N.Y.C. by risk group. According to this report 43%
of all patients with AIDS since the epidemic began remain
alive. KHecause survival data for each risk group was not
available at the time of this report, we assumed that this
figure applies to all ris! groups. Since there are an

—_~

esbtimsted 3,92] homosexual men in New Yorn City alive with

 

 
 

AIDS then it follows that for every homosexual man in New
York City alive with AIDS there are approximately i135
homosexual men in New York City infected with HIV. It is
also estimated that there are approximately 290,000 IVDA in
New York City of which an estimated sixty-five per cent or
120,000 are infected with HIV €7,8]. It is further estimated
that there are approximately 1,531 New York City IVDA alive
with AIDS and hence, for each New York City IVDA alive with
RiDS, there are approximately erghty-fave New York Caty iVD4

infected with HIV.

If then am average ratio of lsice of thase alive with
RiDS to those infected is applied to all risi: groups, one may
orctain & current estimetion of the prevalence of infection,

Fithouch we have assumed for the purpose of calculation

Tt
“—T
rt

fu

the ratio is the same for all groups, thie obvicusly
need not be the case. For example, the ratio for

tnose who receive blood or blood products is prebably lower
Since neat treatment of factors along with screening af
donated Slood has resulted in a decline in HIV infecticn.

In fact our calculation of the ratio in hemophiliacs is 1:57.
The ratio of those at no known risk or heterosexual partners
of high rist group members may be higher since the epidemic
of HI anfection in these groups is probably not as "cid" as
2m ather risk groups. The ratio appears to tend to secline

e the epidemic ages since there would be more cases of A#IDa.

Ty

 
 

 

prolonged survival of these individuals and hopefully a
decrease in incidence of infection.

If one applies the ratio to N.Y.C. cases of AIDS at no
known risk one can estimate the prevalence of HIV infection
in this group. Since there are approximately 34 N.Y.C.
adults at no known risk for HIV infection with Aids based
Lipon data from the N.Y.C. Depatment of Health, we estimate
that 3400 New Yorkers are infected. The U.S. census reports
that there are 3,206,172 adults living in N.Y.C. By
subtracting estimated numbers of high risk group members one
can estimate that there are approximately 4,161,172 New Yori
City adults at no inown risk ¢or HIV infection. This
translates to & prevalence cf @9/100,000. The actual
prevalence of HIV infection in Blood donors in N.Y.C. if
So/iod, 300 C9]. The similarity here appears to suprart the
accuracy Of Sur calculations and assumptions.

Tf the ratios of 1:190 of those alive with AIDS to thoss
rorectec with HIV is applied to the U.S. population then tre
new approximation for aduits infected in the Unitec Statet 15
then somewhat less than 2,900,000 (see table). This
represents an overall increase of eleven per cent from tne
data reported in 1985 and is probably censistent with the
reported incidence of HIV infection in high risk group
members of two to twenty per cent of the remaining uninfected
pepulation per vear £19,111]. There also appears to be
2upronimetely 25,0290 chialdren infected with thie virus if the

ratic holds true im this group (see table).

 

 
 

 

In summary then, the ratio of cases of AIDS to those
infected with HIV appears to have changed from 1:200 to 1:100
from 1985 to 1987. It does appear, however, to remain
relatively constant in two high risk groups in which
reasonably accurate figures concerning the prevalence of HIV
infection and the total population exists. The total number
of Americans infected is estimated to be just less than
-9200,000 and has increased by eleven per cent since 1985.
More accurate estimations of the prevalence of infection
could Be made if more data concerning HIV infection in other
groups was available. This perhaps, could te accompliched
threugh the reporting to lecal public health departments
indaivaduals who are infected with HIV an addition to those
who have developed AIDS. We recognize, however, that this is
¢ controversial issue and must anclude protection for the
rignts of those infectec. If such safeguards are not
emacted, high risk andividuals and others may refrain from
being tested thus potentially increasing the risk cf
transmission to others. Aggressive attempts to educate the
American public about HIV infection would be an effective
means to decrease the incidence and prevalence of this

infection.

 

 
 

 

Table

ESTIMATED NUMBER OF INFECTED INDIVIDUALS WITH HIV INFECTION

IN THE UNITED STATES

”
ts

NO. LIVING WITH

ESTIMATED NO.

WITH

HIV INFECTION

Nov.1957

1,467,100

a7 p40

*11,974

7a, Se

7. WOU

1,707,274

ey

anv g et ee

Ge cree ORSe sides ome cee a ee ome ee

AIDS
July 1983 Nov.1987 July 1°85

GROQUF

i

Homosexual or 4,485 14,67) 1,345,000
Bisexual Men

1VDA gos welt 74 279,000
Hemcphiliacs Zl 196 8,970
Heterasexual

Cantacts of 187 783 36,100
Ferecons at High

Fa st

Recipients of

Blocd Products oF ~1e a0, 7
Sersors at No

Lrownr Rast ~14 5o8 a4, 20
TOTAL ADULTS T9077) ))—”~<C~«SNS HL dG VOSS 1G PDT EPG
FEDIATRIC CASES 23o

: Includes Homoesexual IVDA

=~ Includes andaividuals fram countries

where heterosexual transmission is

common
2 CoC AIDS surveillance data 7/22/85, 11/16/87

*Based on aratio of 1:57 ‘(see text)

 

 
 

BIELOGRAPHY

it.

li.

Curran JW, et al. Science 1985:229: 1352-6
Sivak SL, Wormser GP. N Engl J Med 1985:713:1252
Rees M. Nature 19873226:
Jaffe HW, et al. Ann Intern Med 19853103: 210-4
New York City Fublic Health Press Release Division,
Fesonal communication.
Weinberg DS, Murray HW. N Engl J Med 1987:717:1469
Leib L, et al. Froceedings of the third international
conference on AIDS 1987:172
Lange WR, et al. Froceedings of the third international
conference on AIDS 1987:71
Fersonal communication, Celso Bianco, M.D. N.Y.C. Blood
Center

Schoenbaum EE, et al. Proceedings of the third
internetlonai conference on AIDS 1987:1:7

Martin ¢C. Proceedings of the third international

conference on AIDS 1987:72

 

 
 

 

HOW COMMON IS HTLV-III INFECTION IN THE
UNITED STATES?

To the Editor: Fairly precise figures for the number of cases of the
acquired immunodefciency syndrome (AIDS) in the United States
are based on surveillance data gathered by the Centers for Disease
Control. Since the presence.of antibodies to human T-cell lympho-
trophic virus Type ITI (HTLV-III) is itself not a reportable condi-
tion, and since little information is available on tesung for HTLV-
III antibody in the general population, the number of residents of
this councry who have been infected with the virus is unknown. We
suggest thar the prevalence of this infection in the general U.S.
populacon as well as in identified subgroups may be estimated on
the basis of existing data. The critical assumption in our calcula-
ions is that the ratio of the number of living patients with AIDS to
the number of padents with HTLV-III is constant within different
pacent groups. To test this assumption we compared the number of
living patients with AIDS with the estimated number of patents
with HTLV-III in two well-studied populadons — intravenous-
drug abusers and hemophiliacs. 1: .

The total. estimate of. hemophiliacs in this country’ is 14,367, of
whom at least 52 per cent, or 8970, may be assumed to be infected
with HILV-III according to Western biot electrophoresis.* It is

known that there are 31 living hemophiliacs with AIDS (Centers for
Disease Control, AIDS surveillance, July 22, 1985). Therefore, for
every living hemophiliac with AIDS, there are approximately 289
hemophiliacs with HTLV-III. Similarly, the total number of intra-
venous-drug abusers infected with HTLV-II] in this country may .
be estimated to be 36 per cent? of 750,000,' or 270,000. Since there
are 925 living intravenous-drug abusers with AIDS, it follows that
for every intravenous-drug abuser with AIDS there are 293 abusers
with andbodies to HTLV-III. A recent study revealed thac this
rato was smaller in a group of homosexual men — 1:30.* However,
this group was selected on the basis of attendance at a clinic for
sexually transmitted diseases and voluntary participation in the
study, and may not be representative of all homosexual men:

If the ratio of living patients with AIDS to those with HTLV-II
is approximately 1:300, then the total number of U.S. residents who
are infected with HTLV-III can be estimated to be 1,765,470 (Ta-
ble 1). This figure includes 64,200 persons who do not belong to any
identified risk group. Using currently available population dara,
one can estimate that in the U.S. adult population at no known risk
for AIDS, the prevalence of HTLV-II! infection is 45 per 100,000.

A recent report from the American Red Cross? indicates that the
actual prevalence of HTLV-III infection, calculated by screening
more than | million units of donated blood in the general U.S.
population is 38 per 100,000. We believe that the similarity of the
calculated and actual prevalence figures supports the accuracy of
our assumpuons and calculations.

STEVEN L. Srvax, M.D.
. | Gary P. Wormser, M.D.
-Valhalla, NY 10595 ee _ New York Medical College

 

 
 

 

DRAFT RRR

TESTIMONY PROVIDED BY ROBERT R. REDFIELD MD

PRESIDENTIAL COMMISSION ON THE HUMAN IMMUNODEFICIENCY VIRUS
EPIDEMIC DECEMBER 10 1987

I appreciate this opportunity to share my perspectives of
the AIDS epidemic with this committee. The views presented
here are my own and do not purport to reflect the position

of the Depaartment of the Army or the Department of Defense.

My perspective on the AIDS epidemic is the product of the
opportunity that I have had to practice medicine in the
military in the early years of the AIDS epidemic. As a
physician, scientist and military medical officer concerned
with public health, I have been placed in a unique position
to intregrate these disciplines to comprehend the magnitude
of the problem that the AIDS virus is causing and will

continue to cause our Nation.

First recognized in 1981, AIDS was a new and mysterious
Syndrome. Yet today, AIDS is no longer a mystery. [It is an
infectious disease caused by a retrovirus, HIV.

Within 3 years, science had identified the etiology, defined
the methods of transmission, and developed accurate
diagnotic tests of the AIDS virus infection. Yet despite
these advances, the AIDS epidemic has escalated from a
medical curiosity, to an Isolated public health problem, to

to what tn my opinion, !s a worldwide crisis of potential

 

 

 
 

 

catastrophic proportion requiring an urgent coordinated
response. And yet the magnitude of the problem remains a
subject of controversy; definable yet undefined, resolvable

yet unresolved.

WHY? Unfortunately the orlginal case definition of AIDS
identified only the late clinical stages of the disease, and
thereby severely underrepresented the magnitude of the
epidemic from the beginning. The clinical spectrum of HIV
infection ranges from an acute retroviral syndrome, to
chronic lymphadenopathy, to subclinical T cell deficiency
and finally to systemic T cell deficiency opportunistic
infection defined AIDS). Thls spectrum of disease can be
reduced to 6 stages of infection which each advancing stage
representing a more severe stage immunological dysfunction.
Although early studies reported progression rates of
asymptomatic seropositive individuals to AIDS of only 5-10%,
the natural history of the AIDS virus has been clarified.
Today multiple cohort studies demonstrate progression rates
to AIDS in excess of 30%. Our own cohort study demonstrates
greater than 75% of patients followed for greater than 18
months has demonstrated progressive immunological disease as
accessed by the Walter Reed Staging Scheme. For example
among patients evaluated for greater than 18 months, 10% of
WR2 developed AIDS (WR6); WR3, 29% AIDS: WR4, 71% AIDS: and
WRS 100% developed AIDS (WRE). When followed for 18-42
months 5% of WR2 dled; WR3,14% dleds WR4 57%, dleds WRS, 87%

died; and WR6, 100% dled. My conclusion from the available

 

 

 
 

 

data is that, In the absence of a scientific solution, HIV
infection is a progressive infectious disease (in a majority
and possibly all individuals ) characterized by progressive
immunological dysfunction requiring 5-10 years (or nore)

from time of infection to result in death.

Initially, our understanding of the epidemiology of AIDS was
also incomplete, contributing to the underestimation of the
problem. Today scientific investigations have unequivocally
documented the following modes of transmission: efficient
bidirectional sexual transmission ¢ male to male, male to
female and female to male); parenteral transmission (blood
and blood products, and sharing of drug paraphernalia) and
perinatal transmission. SEXUAL TRANSMISSION IS, AND WILL
REMAIN, THE MAJOR MODE OF HIV TRANSMISSION. I[t Is
critically important to recognize that the sexual
transmission of HIV is not dependent on a specific risk
behavior. For HIV transmission to occur from one human
being to another, all that is required is that one human is.
infected and one human being is not and they communicate by
a method efficient in the transmission of HIV. HIV isa
virus which does not discrimate by age, sex, race, ethnic
group, social status, sexual preference, or risk behavior.
If sexual, parenteral of perinatal contact with the virus
occurs an individual is at risk for infection. THIS IS THE
MESSAGE WE MUST COMMUNICATE!! Despite continued scientific
understanding of the methods of transmission some still

continue to focus on AIDS as a high risk group disease.

 

 
 

 

This 1s re-enforced by the continuation of hierarchical
categorization of AIDS cases by risk groups. Certainly we
all recognize that homosexuality does not cause AIDS; and
heterosexuality does not protect one from HIV infection.
RELIANCE ON RISK BEHAVIOR IS SUBOPTIMAL. KNOWLEDGE OF
INFECTION STATUS IS CRUCIAL. Today, there is only one
common risk group for HIV infection, that is a human being
who has been sexually, parenterally or perinatally exposed

to HIV.

Despite knowledge some still underestimate the ultimate
importance of heterosexual transmission of HIV. These
individuals speculate that HIV will. be unique among STD’s,
rather then assume that the Laws of Nature be as described
until definitively proven otherwise. Worldwide
heterosexual transmission is the major mode of transmission.
Some individuals hope that the United States will be unique.
Unfortunately, in the absence of a scientific solution over:
the next decade, heterosexual transmission will become the
major mode of transmission is the United States. We have an
opportunity to challenge this deadly virus, but we must
recognize our common enemy for what it is, a deadly sexually

transmitted disease.

Today’s AIDS cases represent an historical account of the
magnitude of this epidemic in the late 1970's; yet fails to

define our problem today. If one continues to focus on AIDS

 

 
 

as the magnitude of our problem, we will recognize the

magnitude of today’s problem in 1997.

Likewise, we now appreciate that the epidemiology of AIDS in
1987 merely provides a historical perspective of HIV human
to human transmission nearly a decade ago; yet fails to
inform us about the epidemiology of viral transmission

today.

If we mount our National response to meet the challenge of
AIDS in 1987, we are doomed to failure, because the

magnitude of our response {is a decade out of synch.

It is critical that we change our focus from AIDS and the
problem in the late 1!970’s towards the virus and the problem
now in 1987. Let us embrace the scienctific advance of
1981-1984. Rather than relying on end stage clinical
criteria, we must focus on the VIRUS. Serological testing
for HIV gives us that opportunity to define today’s
problem-TODAY, rather then debate the issue (i.e. what is
the magnitude of HIV infection in America today and what
will it be next year). I believe we should define it today
and redefine it again next year. The logic that elects to
confront a problem undefined when definable, should not be

embraced by our society.

WHAT do we Know about the extent of HIV infection today?

 

 
 

 

Serological studies of the prevalence of HIV infection have
been almost exclusively confined to traditional “high risk”

groups.

One exception is the American Red Cross blood screening data
which demonstrates a rate of 1-4 per 10,000. The limitation
of all these studies is that none reflects the current
status of HIV infection among the general population of

sexually active young people.

It is with pride as a military physician, that I reflect on
the approach that the leadership of the Department of
Defense has executed in response to the AIDS epidemic. HIV
infection has been confronted by the military as the
important medical, public health and military problem that
it is. The DOD policy has been guided by science coupled
with important medical and public health issues. It is
also with pride as a military researcher, that I reflect on
the foresight exhibited by the leadership of the Department
of Defense in developing a system to collect HIV screening
data which enables DOD to provide critically needed
information to our Nation as to the extent of the AIDS virus

infection in our Nation’s young sexually active people.

One DOD screening program providing valuable data is the
civilian applicant screening program. All civilian
applicants for the Armed Forces must undergo a medical
examination with includes along with other medical testing,

screening for the AIDS virus as part of the medical

 

 
 

 

evaluation for fittness for military service. It should be
noted that test results are available within 24 hours for
those negative by screening ELISA tests, and within 72 hours
for those requiring a confirmatory Western Blot. This time
schedule was met independent of where the sample was drawn
throughout the United States. Some have critized that
testing for HIV is extremely expensive. It should be noted
in rebuttal that the DOD testing program, including
confirmatory testing, costs less than: $5.00 per individual
tested. Some speculate that serological testing in low
prevalence population would generate an a prohibitive number
of false postive results; in rebuttal (as reported by COL
Donald Burke in his recent congressional testimony) the
scientifically documented false positive occurrence
following the DOO HIV testing algorithm was less than ! per
100,000 individuals tested. In addition, it should be noted
that the DOD has provided test linked education to over 4
million individuals to date, without precipitating a social
crisis. Large scale timely accurate inexpensive HIV
screening can be done; it is being done every day by both

the American Red Cross and the Department of Defense.

Between October 1985 and September 1987 over 1,250,000
civilian applicants for military service were tested for
presence of antibody to HIV. The overall prevalence of HIV
infection was 1.5 per 1000 (data provided by MAJ John

Brundage and COL Donald Burke WRAIR). To put this in the

 

 
 

 

proper prespective, one should recall that in 1975 less than

1 per 1000 homosexual males in San Francisco were Infected
with the AIDS virus, today over 500/1000 had become
infected. In light of the demonstrated efficient sexual
transmission of HIV, the fact that 1.5/1000 of American
youths desiring to serve their country In the Armed Forces
are already Infected with this deadly sexually transmitted

disease is a tragedy.

The prevalence for males was 1.6/1000 and among females
0.7/1000, for a male female ratio of 2.4 to 1. This ratio
is In sharp contrast to the male to female ratio of 13 to 1

of reported AIDS cases.

HIV prevalence increased almost linearly with age:

age 17, Q0.16/10003 age 18, 0.22/1000; age 19, 0.60/1000;
age 20, 0.98/1000; age 21, 1.42/1000: age 22, 2.17/1000;
age 23, 2.45/1000; age 24, 3.25/1000;

and age 25, 3.36/1000.

Grouped by age prevalence rose from 0.5/1000 for 17-20 years
olds, to 2.4/1000 for 21-25 year olds, to 4.4/1000 for 26-30

years olds, and 3.8/1000 for applicants over the age of 30.

It is of interest that among 17-20 year olds, the male to
female ratio was 1.6 to 1 ¢ 0.5/1000 for males and 0.3/1000

for females).

Prevalence rates also varied by geographic locale (18 month

data). For example, New York State had a prevalence of

 

 
 

 

4.2/1000, Maryland 3.7/1000, New Jersy 3.3/1000, Delaware
2.3/1000, California 2.1/1000, Florida 2.0/1000, and Tezas
1.8/1000.

In addition, the metropolitan areas of New York City,
Newark, San Francisco and Washinton DC all included counties
with prevalence rates of greater than 1%. in both male and
female applicants with a male to female ratio of approaching
bt to l, For example, in Manhattan the male and female
prevalence was 2.0% and 1.7% respectively. In metropolitan
areas such as NYC, gender and race were no longer predictors
of HIV tnfection. The major predictive factor for HIV
infection was age. Individuals 18-21 had prevalence rate of
3-0/1000 for males and 3.8/1000 for female; for applicants
22-25 the rate was 1.3% (13/1000) for males and 1.4%
€14/1000) for females; and for individuals 26 or older, the
rate was 3.4% €34/1000) for males and 1.3% ¢13/1000) for

females.

Analysis of overall temporal trends of military applicants
by birth year cohorts (¢ MAJ Brundage WRAIR), unfortunately
documents a substantial increases in HIV infection among
clvillan applicants from birth year cohorts between 1962 and
1969. For example male applicants between the age of 18 and
25 had a 30% greater llkelyhood of HIV infection between
over a one year period. This was true for both white and
black applicants. Estimated doubling times of the HIV

epidemic In this population was less than 3 years.

 

 
 

 

These data demonstrate that a substantial proportion of the
America’s youth are alreadly infected with the AIDS virus
and capable of unknowingly transmitting this deadly virus to

others.

Another source of information to provide a glimspe of the
extend of HIV in a population Is the result of the US ARMY
HIV program. The preliminary analysis provided by MAJ
Patrick Kelly, WRAIR demonstrated the overall prevalence of
HIV infection in the US ARmy for males is approximately 2
per 1000 and for females nearly 1 per 1000 (0.9/1000).

Again age is an important variable with prevalence rates
ranging from 0.4/1000 for individuals under 20 years of age,
to 1.6/1000 for 20-24 year olds, 2.7/1000 for 25-29 year
olds, 3.0/1000 for 30-34 year olds, and 1.9/1000 for
soldiers over 34 year of age. Analysis by race/ethnic group
demonstrated prevalence rates of 1.1/1000 for caucasians,
2.6/1000 for Blacks, and 2.5/1000 for hispanics. Bivariant
analysis by race and sex demonstrated that prevalence rate
ranged from 0.4/1000 for caucasian females, to 4.7/1000 for
BLack males. Notable however is that the prevalence of HIV
infection black and hispanic females (1.6/1000) is greater
than that of white males ¢1.1/1000). Also of interest ina
the fact that in 1983 and 1984 as a physician at WRAMC,
approximately one half of the patients that I[ cared for with
symptomatic retrovirus infection were married men and women.
I was forced to confront the grim reallty then of the Impact

that HIV infection will have on the American family.

 

 
 

 

Although many critics challenged our findings of
heterosexual transmission of HIV, few acknowledged the
uniqueness of WRAMC patients in 1984 with HIV Infection (le
50% married, over 20% females ect). As a result of the HIV
program, it should come as no susprise that over 40% of the

soldiers infected with HIV are married.

The philosophy behind the HIV program simply put its that
knowledge of HIV infection is better than Ignorance of HIV
infection. Each member In the US ARMY !s provided several
{important opportunities. The opportunity [f infected with
HIV to recieve medical care based on all the medical
knowlegde available {tn 1987. The opportunity to face the
future without blinders. The opportunity to no longer
unknowingly transmit this deadly virus to another human
belng. The opportunity to be informed if they have been
exposed to this virus. The HIV screening programs were
responsible for providing this knowledge to the individual
soldier over 90% of the time. Less than 3% of total HIV
infection were established as a result of voluntary self
referral and less than 5% were established because of a
Clinical evaluation. Therefore greater than 90% of all HIV
infected Individuals in the US Army, only became aware of
their HIV infection as a consequence of one of the

military’s screening programs.

The DOD has a firm grasp on the extendof HIV in the DOD.

The US ARMY also has in place procedures, not only to insure

 

 
 

 

earlier accurate diagnosis of HIV infection in its members,
but also to define the dynamics of infection In the ARMY
over time. Quess work will not be required. Preliminary
analysis of direct measurements of the incidence of HIV
infection In the US Army, (a population provided tested
linked education) is in excess of 0.5/1000 (data provided
by MAJ John McNeil, WRAIR>. I believe that the data
provided by the US military provides America with Its best
quess. However because of selection blases, it should be
recognized that the milltary data in likely to underestimate
the HIV infection rates in the * general population “° of

young sexual active Americans. |

BUT WHAT IS THE MAGNITUDE OF HIV IN YOUNG AMERICANS
THROUGHOUT THE US TODAY? WHAT WILL IT BE NEXT YEAR? WHAT
WILL BE THE CHANGE? IS OUR CURRENT POLICY EFFECTIVE? HOW
WILL WE KNOW? WHAT SHOULD WE D0?

Never before in the history of the human race, has a
soclety been given the opportunity to have the sclentific
knowledge at Its disposal, at a time of the emergence of a |
new Infectious agent. We have been given a rare and
treasure opportunity. It is our responsibility not to waste
it. In the year 2000, neither you nor I will be held
accountable for whether or not an AIDS vaccine is developed
or curative therapy discovered; these sclentific
advancements will occur with time, maybe in 10 years maybe

sooner, or maybe later. No one Knows. However, we will be

 

 
 

 

 

accountable to the American of the twenty-first century Cour
children and thiers), for our utilization and Implementation

of the Knowledge available to us in 1987.

No matter how excellent the allblas for why we do not have a
firm grasp on the magnitude of HIV infection in 1987 in our
Nation, we can obtain this critical information. I believe
this to be crucial. We must define the extent of HIV
infection. If we continue to mount our Nationa oan: to
foight Hi cr “ha i
meet the challenge of AIDS in 1987A4°We our dooméd to
failure. National policy should be based on the knowledge
of HIV Infection in our Nation today; and not limited to
quesses, models or opinions. We must ensure that we don’t
loss the war just because we underestimated the enemy; or
because having committed the necessary resourses, applied
them to the wrong Front. The only way I know to ensure the
generations of the twenty-first century (my children, our
children) and beyond, that we confronted this virus to the
best of our abllity is to first be sure we accurately
define the problem. This is the only way we can optimize
our National response, providing all the resources needed,
where thier needed, when their needed. This virus will
steal the lives of more young Americans over the next
decade, than those who gave thier lifes to defend our
Country in the past 4 wars. We can hope for an early
sclentific breakthrough, otherwise, unfortunately, the AIDS
epldemic of the late 1990’s has already occurred. However

the AIDS epidemic of the twenty-first century is

 

 
 

 

preventable. We have at our disposal all the tools necessary
to eliminate ignorant transmission of this deady virus.

Yet, to date, we have falled to take charge and accept our |
resposnsibility to challenge this virus with the vigor,
courage, and committment that has made our Nation so great.
We need ocnly to use our knowledge, to recognize our
compassion, to explo!t our ingenuity, to demonstrate our
courage , and persevere. Then (to paraphase Theodore
Roosevelt) at best we will know the triumph of high
achievement measured in the reduction of human suffering, or
at worst with fallure, know we falled while daring greatly.
It is Important that we all recognize the problen for the
crisls 1t 18s so we can begin to develop the solution for a
better tomorrow. Define the problem, remove the
uncertainty. Let’s change the debate from “Is thier a
problem? Where? How great?, to, now that we know the

extent of the problem *° What can we do?*,

In closing, 1!t is my opinion that the AIDS virus is no
longer an isolated pulbic health problem. It has escalated
to a National problem capable of affecting all governmental -
departments and altering our society and the World as we
Know it today. MORE AMERICANS WILL DIE OVER THE NEXT 10
YEARS FROM THIS VIRUS THAN LOSS THEIR LIVES IN THE LAST 4
WARS. WE SHOULD ASK OURSELVES= BEYOND THE LOSSES WILL BE
EVEN GREATER. WHAT IS AN APPROPRIATE RESPONSE TO THE - |
REALITY OF SUCH A PREVENTABLE LOSS OF AMERICAN LIFE? We

must recognize it as such now desplte the fact that the

 

 
 

 

ultimate impact of today’s problem will not be upon us for
3710 years. The difficulty, as I see it, is that some
Americans can not foresee the grim reality of the next
decade. They will only understand when forced to confront
human suffering in the flesh. We must provide them the
vision of the future, so that we have the necessary resolve

to respond fully as a Nation now.

No matter now excellent the alibias, we know enough now to
ifmtt the impact that this virus will have on us, our
children and our Nation. THE EPIDEMIC OF THE TWENTY-FIRST
CENTURY IS PREVENTABLE. Yet we have been slow to take
charge and accept our responsibly to challenge this virus
with the vigor and committment that has made our Nation so
great. Public health Is a responsiblity of governement.
Cooperation to maintain the public health is the
responsibity of each of the eltizens of that government. In
any infectious disease knowledge of the infection is
Paramount in its control. I would hope that soon every
Sexually active citizen would have the same opportunities
provided the members of the Armed Forces. Vigorous
leadership, coupled with accurate education and classical
public health measures will Limit the spread of this deadly
virus in our Nation. But first, we must, as a Nation, define
and recognize this AIDS virus as the national crisis that it
is. Then we must have the courage to develop an effective
national pollcy and the resolution to mobilize the resources

necessary to combat and defeat our common enenmy, this

 

 
 

 

virus. THE EPIDEMIC OF THE TWENTY-FIRST CENTURY CAN BE
PREVENTED. WE CAN DO IT.

Thank you for the opportunity to both share this information

with you and to express my opinion.

 

 
 

PREPARED STATEMENT TO THE PRESIDENTIAL COMMISSION ON AIDS

Richard L. Vogt, M.D., F.A.C.E.

State Epidemiologist

Vermont State Health Department

Clinical Associate Professor

Departments of Family Practice
and Medicine

University of Vermont

College of Medicine

Dec. 10, 1987

 

 
 

 

Aids Virus Testing

I have been asked to provide you information about the
accuracy of HIV antibody testing in both high and low
prevalence populations. As has been discussed, there are two
testing procedures used to attempt to identify persons
infected with the AIDS virus. The first procedure is the
ELISA test which measures human immunodeficiency virus
antibody levels in serum. The ELISA test was originally
designed to be as "sensitive" as possible, to pick up all
those persons who may have antibodies to the virus so that
they could be removed from the blood donor pool. As a
result, this test creates a significant pool of individuals
who may not have the antibody, but test positive, anyway.
Therefore, it is recommended that the blood sample be re-
analyzed with the ELISA technique to determine if the sample
is repeatedly positive. If a sample is repeatedly positive,
through two analyses, the level of suspicion is raised that

this may be a positive test.

If the blood sample is repeatedly positive on ELISA, the
blood sample is subjected to another test, the Western blot
method. This is a very laborious and costly procedure which
identifies specific particles of the HIV antibody in a
person's blood as shown in the next slide (SLIDE 1). Each

band shows a reaction to the antibodies to the specific

 

 
 

 

components of the HIV virus. Researchers currently feel that
positive tests to both ELISA and Western blot should be
considered as probable laboratory evidence for infection with
the virus. However, as we shall shortly see, the accuracy of
that determination depends on who is tested. If the test is
considered positive by the testing laboratory, persons
submitting the sample would be notified of a positive test
result and the person is counseled on the likelihood for

exposure to the virus.

Accuracy of the HIV Tests

All laboratory tests need to be evaluated for their accuracy
and are usually measured against a comparison called a "gold"
Standard. When the ELISA tests were originally licensed,
they were evaluated against patients who had AIDS and
evaluated against the Western blot test. Most determinations
about accuracy of HIV antibody tests are made by comparing
ELISA with Western blot procedures. Test comparisons with
the recovery of the virus, the true evidence of infection,
are technically difficult and usually do not enter into the

discussion about test accuracy.

One of the dilemmas that confuses all discussions about HIV
test accuracy concerns the disagreement within the scientific
community about what constitutes a positive Western blot

test. The next slide (SLIDE 2) shows the differing criteria

 

 
 

 

for test positivity by different scientific groups using this
test around the country. Laboratories also use different
preparations for conducting their Western blot procedures.

It is no wonder that there has been a great deal of
discussion about the accuracy of HIV antibody testing when we
have disagreement over the criteria for a positive Western

blot test.

Most statements about accuracy of laboratory tests are made
in terms of two determinations: sensitivity and specificity
as shown in the next slide (SLIDE 3). The "gold" standard is
listed at the top of the 2 X 2 square and the test result is
listed on the left hand side. The published sensitivities
(C/A+C or the chance of correctly identifying a positive test
in those who have AIDS) and the specificities (D/B+D or the
chance of correctly identifying a negative test in those who
do not have AIDS) of the ELISA test look fairly good. The
range of published ELISA sensitivities is 93.4% to 100% and
the range of specificities ranges from 99.2% to 99.8%. There
were similar agreements when comparing the ELISA test with

the Western blot analysis.

When considering whether a medical test is meaningful in
high or low prevalence populations, you need to turn to two
different measures of accuracy as depicted in the next slide
(SLIDE 4). The comparison cells are the same as those used

for sensitivity and specificity determinations, but different

 

 
 

 

equations are used to determine a more important measure:
predictive values. Predictive values answer the following
questions: What is the chance that a person with a positive
test is truly infected with the virus (A/A+B on the slide)?
Likewise, what is the chance that a person with a negative
test is truly free from the virus (D/C+D on the slide)?
These are very important questions to be answered when a
patient wants to know the meaning of his test result.
Whereas sensitivities and specificities remain relatively
constant with both low and high prevalence populations,
predictive values are widely different depending on the
characteristics of the population screened.

This stands to reason, because if you screen a population
that has no chance of being infected or afflicted with a
disease, all of the positives generated from a test will be
false positives, since no laboratory test or series of tests
are perfectly accurate. For example, you could test 1000 men
to see if they are pregnant. The laboratory test will
identify a few that will be test-positive, even though there
is no possibility that the positive results will be accurate.
In this situation, we would refrain from testing an
inappropriate population (men), so we would fail to generate

erroneous information.

We need to consider the test accuracy with the combined

procedures of both the ELISA and Western blot tests.

:
i

 

 
 

 

Sequential testing will tend to further limit the number of
false-positive tests but will not eliminate them. The number
of false-positve tests for both ELISA and Western blot tests
will still escalate when you screen populations without much
infection. This is an exaggeration of testing errors that
occurs with any laboratory test or sequence of tests which
tends to be magnified in populations that are sparingly

infected.

The next slide (SLIDE 5) helps to describe the value of the
combined ELISA and Western blot analyses when they are used
to screen different groups. The likelihood of combined
correctly positive tests is depicted on the vertical axis.
The horizontal axis is the estimated infection rate in
different populations. Populations with a greater likelihood
of infection will tend to be further to the right on the
horizontal axis. Two assumptions for the sensitivities and
specificities of the two tests are given in the legend. The
exact values do not have that much importance. The curves
that they generate help demonstrate the principle that the
value of combined positive test results will deteriorate if
they are used to screen low-risk populations, such as those

seeking marriage licenses or routinely seeking medical care.

Looking at this situation in another way by using data in an
article pubiished by Meyer in the New England Journal of

Medicine, the overall best estimate of the false-positive

 

 
 

 

rate for the combined tests: first ELISA, second ELISA and
confirmatory Western blot test is estimated to be l per
20,000 samples in that article. The article also states that
the seroprevalence of a low prevalence population (such as
female blood donors) is approximately 1 per 20,000. That
means that for every 20,000 tests run, there will be a total
of one false positive test and one true positive test.
Therefore, half of all positive tests will be inaccurate when
one screens this population. That is, the positive

predictive values of the test is 50%.

Sequential testing presents another dilemma that. often fails
to be addressed when HIV testing is discussed. Persons may
be ELISA test positive but Western blot test negative. It is
difficult to sort out whether the patient is infected and the
result is frequently considered "indeterminate". The
likelihood of an “indeterminate” result increases
dramatically when populations who are unlikely to be infected
are screened. Using strict criteria for Western blot
positivity has the appeal of minimizing the number of false
positive results. However, this will increase the number of

“indeterminate"™ test results, as well.

One can see that one runs the real risk of generating
inaccurate seroprevalence data. when low-risk populations are
screened. Even more importantly, one also runs the risk of

labelling someone positive who is not really positive if

 

 
 

widespread screening of low risk populations are undertaken.

The tests, as developed, are extremely useful when used to
screen appropriate populations, most notably those persons
who are engaging in high-risk behaviors such as intravenous
drug users and persons who are having unprotected sex with
multiple different sexual partners. But the tests can and
will fail when applied to incorrect populations. Tests have
to be close to 100% sensitive and 100% specific to be
applicable for all groups. The ELISA and Western blot tests

are not that accurate for universal applicability.

 
 

 

p120

p66
p51

p41

p31

p24

SLIDE 1

 

 

 
 

 

SLIDE 2

 

 

 

 

 

 

 

 

 

 

Weste ot e t a Fo os Vv est
m
a @ a gp160
@ i !|lea. Qf ep110120
a | | . od = p65/66
= p55
a p51/52
oe on = a gpél
8 wg p31/32
4a eB || ai ap p24/25
am
aad = pi7/18
@
4 b P ”
Red Cross‘ DuPont$ ASTHO® = NIH? Positive
DoD / for HIV

1 Only one possible combination {is shown.
2 Presence of either one, plus p24/25 and p31/32, is positive.

3 Presence of any two is interpreted as positive: gp110/120 and gp160 count
as one. !

4 Positive is either pl7/18, p24/2S, or p55, plus gp41, gp110/120, or gpl60,
plus p31/32, p51/52, or p65/66.

S Positive is p24/25, p31/32, and either gp41 or gp110/120.
6 Positive is any two of p24/25, gp4l, and gpll0/120-gp160.

7 Positive is p24/25 plus gp4l.

 

 
 

SLIDE 3

 

 

 

 

 

 

INFECTION NO INFECTION

TEST

POSITIVE A="TRUE POSITIVE" B="FALSE POSITIVE"
TEST

NEGATIVE C="FALSE NEGATIVE" D="TRUE NEGATIVE"
SENSITIVITY = PROBABILITY THAT THE TEST WILL BE POSITIVE

WHEN INFECTION IS PRESENT = A/ (A+C)

SPECIFICITY = PROBABILITY THAT THE TEST WILL BE NEGATIVE

WHEN INFECTION IS NOT PRESENT = D/ (B+D)

 

 
 

SLIDE 4

NO INFECTION

 

B="FALSE POSITIVE"

 

 

 

D="TRUE NEGATIVE"

 

 

INFECTION
TEST
POSITIVE A="TRUE POSITIVE”
TEST
NEGATIVE C="FALSE NEGATIVE"
PREDICTIVE VALUE POSITIVE = PROBABILITY
A POSITIVE TEST TRULY IS
PREDICTIVE VALUE NEGATIVE = PROBABILITY

A NEGATIVE TEST TRULY IS

THAT A PERSON WITH
INFECTED = A/ (A+B)

THAT A PERSON WITH

NON-INFECTED

D/ (C+D)

 

 
 

 

SLIDE 5

Figure 1. Predictive Value of Combined Positive ELISA
and Western Blot Using Different Estimates
of Sensitivity and Specificity

100- a

 

 

rv
de
VS
we)
o
ast [
© |
2 |
abu!
9 80+ |
oO | x ELISA sensitivity = .97
“ 75+ / Specificity = .99
0 | WB Sensitivity = .92
© | specificity = .995
S
o ar | © ELISA sensitivity = .96
> | specificity = .98
© | WB sensitivity = .90
> ST | specificity = .98
J
O |
a
:
0
|
§5+ ]
50 .

 

05 30 55 80 1.05 1.30 1.55 1.80 2.05
Infection Prevalence (% of population)

 

 
 

 

November 23, 1987

Farr's Law Applied to AIDS Projections

by

Alexander D, Langmuir (1)

Prepared for a panel on the Epidemiology of AIDS at
hearings of the Presidential Commission on the Human

Immunodeficiency Virus Epidemic, Washington, DC, December
10, 1987.

 

Based-on a paper submitted to the New England Journal of
Medicine by Dennis J. Bregman and Alexander D. Langmuir.

1. Epidemiologist, retired, Chilmark, MA 02535

 

 
 

 

The current AIDS epidemic presents a major challenge to
theoretical epidemiologists to find a rational basis for
projecting the future incidence of disease. Purely empirical
approaches cannot be expected to provide as relevant a projection
as one based on epidemiologic inference. The epidemic theory of
AIDS must have a biological basis supported by epidemiologic
precedent. In October 1985, one of us proposed that Farr’s Law of
Epidemic Curve be invoked (1). In this paper we extend this
concept by applying it to the most recently available data. Our
analysis indicates that the homo/bisexual male and IV drug abuser
epidemics will crest in 1988 and then decline progressively

through 1995.

FARR'’S LAW

In 1840 in his second annual report as secretary to the Registrar
General of England and Wales, William Farr observed that epidemics
tended to rise and fall. This was not a particularly profound
statement, but he went on to suggest that the curve followed a
reproducible mathematical function. As Brownlee pointed out some
70 years later Farr did not seem to be conversant with Gaussian
statistics, but his proposed method of fitting a curve to the
smallpox data in 1840 was essentially a normal curve (2). In its
simplest form Farr’s Law of the Epidemic Curve may be stated

“epidemics rise and fall in a normal curve.”

Farr did not pursue these ideas further, at least not in his

extant “ritings, until 1865, when a severe epizootic of cattle

 

 
 

 

plague struck in England. The epizootic began in late summer and
increased alarmingly to a total of approximately 50,000 eases
during a four week period at the beginning of the year 1866. On
February 16, 1867 Farr wrote to the Daily News of:London as
follows; (2)
“Sir,
The following passage occurs in the report of Mr. Lowe’s
last speech in the House of Commons;
"If we do not get the disease under by the middle of
April, prepare yourself for a calamity beyond all
calculations. You have seen the thing in its infancy.
Wait, and you will see the averages, which have been
thousands, grow to tens of thousands, for there is no
reason why the same terrible law of increase which has
prevailed hitherto should not prevail henceforth.’
No one can express a proposition more clearly than Mr.
Lowe; but the clearness of a proposition is no evidence
of its truth. And in the present instances I hope to be
able to convince Mr. Lowe himself that the proposition
which he has propounded is founded on a
misconception...”
Farr then proceeded to show that the figures in successive four
week periods did not continue on the principle of doubling but
rather that the rate of increase was progressively declining:
“Thus the increase in the first interval was at the rate
of 96.07 percent; in the second interval it was 79.81
percent; and in the third or last interval under
observation it was only 39.47 percent. Now here is a

complicated law of decrease in these rates; and the

"terrible law’ is such that - if it is any law at all -

 

 
 

 

the number of attacks in the next four weeks will be
less.”
Farr predicted an early crest for the epizootic followed by a
rapid decline. His published prediction was amazingly close to

the actual course of the epizootic.

There is a certain resonance in the public statement of a high
ranking member of parliament commenting on the cattle plague with
those of some high ranking officials in Washington and elsewhere
commenting on the current AIDS epidemic. Wade Hampton Frost
during the 1930s is reported to have commented that Farr’s
published statement on the cattle plague was the ‘most courageous’
prediction ever made by a responsible public health official (3).

It’ deserves respect today.

THE AIDS EPIDEMIC

The data we present here are derived solely from the official
reports of the CDC AIDS Program released on October 5, 1987.

Table 1 shows the semiannual incidence of AIDs cases in the United
States from 1982 to 1986 by transmission category and date of
diagnosis. These data have been corrected for the estimated
number of diagnosed cases not yet reported to the CDC AIDS
Program. The data for the first Pive transmission categories are

graphed On a semilogarithmic scale in figure 1.

For 1982 and 1983 all of the curves show a steep linear increase

reflecting the well known logarithmic or geometric rise in

 

 
 

 

incidence. By 1984 the curves begin to veer off to the right.
This tendency increases in 1985 and 1986. This is merely a
graphic presentation of the long recognized and fully publicized
increase known as the doubling time used to describe the early
stages of the epidemic (4,5). It is indeed reminiscent of Farr’s

observations on the cattle plague.

Extending the curves backward in time and downward to an arbitrary
epidemic threshold of less than ten cases per year it is clear
that the epidemic among homosexuals began in the late 1970s; that
among the IV drug abusers appears to have started in 1979; and
those among the transfusion recipients and hemophiliacs clearly

later, in 1981 or possibly not until 1982.

The data for heterosexual and unknown transmission categories are
deemed too insufficient to warrant projecting their curves. We
note that these transmission categories have consistently produced
about 7 percent of all AIDS cases. No evidence suggests their

trends differ markedly from the total.

In table 2 we have consolidated the data for homosexuals and IV
drug abusers into one group comprising about 90 percent of the
AIDS cases and tabulated the data on an annual basis. The column
of first-ratios is calculated by dividing each annual total by the
total of the preceding year. It reflects the annual rate of
increase. This declines from 187 percent in 1983 to 49 percent in

1986. The second column of ratios reflect the acceleration in the

 

 
 

 

 

rates of increase. In this instance the acceleration is negative,
the second ratio being less than unity.

Actually these second order ratios are quite stable, particularly
those for 1985 and 1986 which are nearly identical. Taking the
mean of these two ratios, namely .8490, and assuming it to remain
constant we have fitted a normal curve to the data by simple

arithmetic as shown.

These projections are shown in figure 2. The fitted curve
characterizing the homo/bisexual and IV drug abusers crests in mid
1988 and declines symmetrically thereafter. The total cases for
these combined transmission categories to 1995 is 113,588.
Inasmuch as these comprise 90 percent of the total cases, and this
percentage has not changed appreciably over time one can arrive at
a rough but reasonable estimate for total cases of AIDS in the

United States to be reported by 1995 somewhat less than 130,000.

We do not believe it would be productive to calculate confidence
limits for this very simplified analysis of an intrinsically
complex situation. To do so would imply a greater degree of
confidence in the precision of our data than is justified.

Perhaps it would be more prudent for us to state that our
projection for total cases falls short of the projection accepted
by the National Academy of Science Commission on AIDS of 270,000
cases by the end of 1991 (6,7). The basic conclusion of our study

is that using elementary mathematical concepts and a hypothesis

 

 
 

with biological basis (8) supported by epidemiological precedent
the data suggest that the AIDS epidemic will crest soon and then

decline.

 

 
 

References

1.

3.
4.

Langmuir, AD. Opening comments at a panel discussion on the
@pidemiology of AIDS. Annual Meeting, Institute of Medicine,
Washington, D.C. October 1985.

Brownlee J. Historial note on Farr’s theory of the epidemic
curve. British Medical Journal, August 14, 1915.

Merrell, M. Personal communication.
Jaffe HW, Bregman DJ, Selik RM. Acquired Immunodeficiency

Syndrome in the United States: AIDS - the First Thousand Cases.
Journal of Infectious Diseases vol 148(2) 339-45 Aug 1983.

- Morbidity - Mortality Weekly Report, Centers for Disease

Control, Atlanta, GA. Update: Acquired Immunodeficiency
Disease - United States. 1986 35: 757-766.

Baltimore D, Wolff S. Confronting AIDS =- Directions for Public

Health Care and Research. National Commission on AIDS, National
Academy of Sciences and Institute of Medicine, National Academy
Press, Wash DC, 1986.

Macdonald DI. Coolfont Report: a PHS Plan for Prevention and
Control of AIDS and the AIDS Virus. Public Health Reports
101(4) pp 341-348 July-August 1986.

Brownlee J. On the Curve of the Epidemic. British Medical
Journal, May 8, 1915.

 

 
 

 

TABLE 1

Semiannual Incidence of AIDS in the United States :
from 1982 to 1986 by year of diagnosis and transmission category

Year and Homosexual IV-Drug

 

 

half year Bisexual Abuser Both Transfusion Hemophiliac Heterosexualt Unitnown Total
1982 ist 223 56 23 1 2 34 14 353
2nd 399 109 71 7 4 33 20 643
1983 Ist 754 216 117 13 8 66 33 1207
2nd 982 296 160 27 4 65 46 1580
1984 Ist 1618 410 214 40 24 96 72 2474
and 2118 553 300 49 25 120 75 3240
1985 Ist 2982 755 343 103 34 153 rT 4461
2nd 3735 951 378 106 68 193 154 5585
1986 Ist 4570 1142 528 183 60 278 190 6951
2nd 5496 1263 610 202 82 315 259 8227
Total 22877 5751 2744 731 311 1353 954 J4721
Percent 65.9 16.6 7.9 2.1 0.9 3.9 2.7 100.0
t Figures corrected for delayed reporting due to the lag from date of diagnosis to date the

report is received by the CDC. The corrections were made specific. for each transmission
category and month of diagnosis

+ Heterosexual category consists of contacts of AIDS infected persons plus those exposed
overseas.

 

 

 
 

 

Figure 1
Semiannual Incidence of AIDS in the USA
by Year of Diagnosis 1982 - 1086

Number of Cases

      

10000 =

= Homo/Bisexual

1000 = IV-Drug Abuser
= : Both Homo/Bisexual
= and iV-Drug Abuser
a Tranefusion Reoipient

100 = Hemophilieo
10

   

1

q
1978 1980 1982 1984 1986 1988 - 1990 1992
Semiannual Period of Diagnosis

PRE peppeeppriptiptipugprss
0 v v v rF.)—COUYti<‘“<—Cir

 

 
Table 2
Annual incidence of AIDS in the United States

among homosexual/bisexual and IV drug abuser cases combined

by date of diagnosis, 1982 to 1986 with projections to 1995. *

 

Number of First Second
Year Cases Ratio Ratio
1982 881
1983 2525 2.8661
1984 5213 2.0646 0.7203
1985 9144 1.7541 0.8496
1986 13609 1.4883 0.8485
1987 (17197) (1.2636) (0.8490)
1988 (18450) (1.0729) (0.8490)
1989 (16807) (0.9109) (0.8490)
1990 (12999) (0.7734) (0.8490)
1991 ( 8536) (0.6567) (0.8490)
1992 ( 4759) (0.3575) (0.8490)
1993 ( 2253) (0.4734) (0.8490)
1994 (| 979) (0.4019) (0.8490)
1995 ( 340) (0.3413) (0.8490)

x Projections, shown in parenthesis, for 1987 and later are

based on the assumption of a constant second ratio in 1985

and 1986

 

 
 

—_

 

 

Figure 2
AIDS In the USA
Homo/bisexual and IV Drug Abuser Cases

by Year of Diagnosis 1082-1086 with Projections to 19906
Thousands of Cases

 

 

 

 

== Projected from Data
% Observed Number
16F
10°
6 bas
or—t }——_}—_—_}—_+—_+—_+—__+_+_+—_ +_ “4
1980 1984 1988 1992

Year of Diagnosis
Corrected for delayed reports

 

 

 

 

 
 

AIDS Projections are too High

 

Prepared for presentation by Alexander D. Langmuir at
the Hearing on Incidence and Prevalence, of the Presidential
Commission on the Immunodeficiency Virus Epidemic, in

Washington, DC, Thursday, December 10, 1987

 

 
 

I deeply appreciate the honor of being invited
from an active retirement on Martha's Vineyard to testify
along with my colleagues on this panel. They are now on the
firing line in this serious epidemic crisis. I once was
there too when chief epidemiologist at the CDC. They have
my warm sympathy and best wishes.

My long held view, which is now becoming widely
known is that the projections of the incidence of AIDS in
the USA are too high. I presume this is the reason I have
been invited to testify. I am happy to do so.

The active practice of epidemiology has been my
professional career for 50 years. During this time I have
become increasingly intrigued with Epidemic Theory, namely
the effort to divine the laws governing the occurrence and
course of epidemics and to express these in mathematical terms.
Progress in this field over a century and a half has been
disappointingly slow. The factors involved are too complex
varied and intangible, the measurements are too imprecise
to be amenable to mathematical expression even with the aid
of modern computers.

Back in 1840, however, a great epidemiologist named
William Farr in London made an observation that in a very
general way had stood the test of time. He noted that epidemic
small pox seemed to follow an orderly path. He fitted a
curve to his data which we now recognize to be a normal curve,
the simple "“cocked-hat" curve that is the basis of classical
statistics. His later admirers have promulgated Farr's Law
which simply: states that epidemics rise .and fall .in -a -pormal

curve. During an epizootic of cattle plague in London in 1865-

 

 
 

 

- 2-

h

1866, when the incidence was seemingly disaster-

ously increasing he predicted that the epizootic would soon
crest and rapidly decline. He was right. Thus began the
hazardous game of predicting the course of epidemics.

I am known among my colleagues as temerous
Alex, foolishly courageous in making predictions. I have been
right some of the time, and wrong on many occasions. I persist
in this trait for several reasons. First. of all it is exciting
and fun to be at the cutting edge of an unresolved epidemic
problem and sometimes even over the edge. More seriously
to make reasonably responsible predictions demands at least
the beginning of an understanding of the underlying theory
of the disease. When predictions are fulfilled one gains
confidence in going forward toward a more complete theory.
When predictions fail you pick up the pieces, reevaluate
your basic premises and start over.

I first indulged my obsession concerning AIDS
in October 1985. I was “drafted" by Dr. Fred Robbins, then
president of the Institute of Medicine to open the discussion
@t @ panel on the subject during the annual meeting that was
devoted to AIDS. He instructed me, “Be provocative.”

I quoted William Farr. I then went on to challenge
the then prevalent and still widely believed concent thar
AIDS would continue to spread geometrically and extend
widely into the heterosexual population. I stated categor-
ically on the basis of my professional judgement that "No

biological system, surely not an epidemic. can increase geo-.
metrically for long. I argued that. among the four principal

transmission categories, the homosexuals, the IV drug abusers,

 

 
 

-3-

the transmission recipients and the hemophiliacs I could

find no theoretical basis for a continuing geometric increase.
In fact the then already well known “increase in the

doubling time" precluded such a conclusion. I even hazarded

a forecast, the word prediction is too strong, that the
epidemic would crest in “mid summer of 1986." This caused

no ripple of interest in the panel. They went on to discuss
other matters. No mention was made in the book summarizing
the whole meeting.

It is just as well. The forecast missed. The
incidence continued to increase, to some alarmingly, although
the rate of increase was dampened and the doubling time length-
ened. I was wrong.

With my colleague Dennis Bregman we persisted.

In May we submitted a brief manuscript to the CDC with proposed
publication in the Morbidity-Mortality Weekly Report. We pre-
dicted a crest to the epidemic in December 1987. The paper

was respectfully declined. Although all the evidence is

not yet in it appears that we were wrong again

Reassessment of our failures led to what we think
are clear explanations. The first time, in 1985, -I had
grossly underestimated the incubation period. I assumed it
to be two to three years. In this error I had ample company.
We now believe it is more nearly six to seven years and quite
variable. The second failure was an artifact of reporting.

During the late months of 1986 and early months of 1987

 
 

reporting from the states to CDC lagged perceptibly only

to be corrected by an increase in reporting during the late
winter of 1987. This gave, for a short period, a false sense
of an impending turnover in the curve.

Now Bregman and I are persisting in our view. A
manuscript that has been submitted to a peer review scientific
journal has already been introduced in evidence to the panel.
I would like to examine the two tables and figures in that
paper.

Table 1 presents the incidence data of AIDS in
the USA from 1982 through 1986 by transmission category.

This comes directly from the official CDC AIDS surveillance
data and is or at least should be familiar to all students
of the problem. It has been published in essentially this
form regularly in the Morbidity-Mortality Weekly Report.
There are certain crucial differences. These data are by date
of diagnosis. The are limited to cases diagnosed in 1986 or
earlier. They are drawn from the October quarterly report
for October 1987. Thus a nine month period has elapsed to
correct for the lag between diagnosis and report to CDC.
The data have been further corrected for longer delayed
reports. The table has been submitted to CDC, not for

approval but for criticism. None has been offered.

 

 
 

 

-5-

These data are graphed in figure 1, using the
standard semilogarithmic scale which shows relative change
and properly reflects comparative trends. The straight
line steep upward trends in 1982, 83, and 84 are evident
reflecting the logarithmic phase of the epidemic. Then
all of the curves veer off to the right with increasing
speed, which is a graphic portrayal of the familiar increase
in the doubling time.

The curves are astonishingly parallel, a phenom-
enon I did not expect to find, but which I believe to be
of great significance. The essential congruity of the curves
must mean that some overriding force, or rather a composite
of many forces, in mathematical terms a vector or resultant,
is exerting approximately equal effect on all four transmission
categories in spite of the wide divergence among these groups.

Any epidemiologist who accepts William Farr as a
role ‘model cannot resist the temptation to apply his law
to these data. To achieve the stablest data for curve fitting
we have combined the semi-annual figures in table 1 to annual
data and merged the homosexual data with the IV drug abuser
figures in Table 2. Using the simplest arithmetic procedures
we have fitted a normal curve. The fit is astonishingly good.
The crest occurs in mid summer 1988. The decline is sym-
metric to a low point in 1995. The total projected cases of
AIDS in the USA is approximately 130,000.

We make no claim to great precision in this pro=
jection, We rather expect the curve to decline somewhat more

slowly and thus be skewed to the right.-It should fall- off

to an endemic level yet to be accurateliy estimated.

 

 
 

-6-

The important part of the projection is that the
epidemic will crest soon. The total projected cases will be
in the range of one half the presently widely accepted figure.

The policy implications of this projection are
enormous and serious. I would be happy to offer my views if

the panel is interested.

 

 
 

 

TABLE 1

Semiannual Incidence of AIDS in the United States
from 1982 to 1986 by year of diagnosis and transmission category*

Year and HNomosexual IV-Drug
half year Bisexual Abuser Both Transfusion Memophiliac licterosexual* Unknow; Total

 

1982 Ist 223 56 23 i 2 34 14 353
2nd 399 109 71 7 4 33 20 643
1983 ist 754 216 117 13 8 66 33 1207
2nd 982 296 160 27 4 65 46 1580
1984 Ist 1618 410 2i4 40 24 96 72 2474
2nd 2118 §53 300 49 25 120 75 3240
1985 ist 2982 755 343 103 34 153 91 4461
2nd 3735 951 378 106 68 193 164 5585
1986 ist 4570 1142 528 183 60 278 190 6951
2nd 5496 1263 610 202 82 315 259 8227
Total 22877 5751 2744 731 311 1353 954 34721
Percent 65.9 16.6 7.9 Ze1 0.9 3.9 2.7 100.0
t Figures corrected for delayed reporting due to the lag from date of diagnosis to date the

report is received by the CDC, The corrections were made specific.for each transmission
category and month of diagnosis:

+ ileterosexual category consists of contacts of AIDS infected persons plus those exposed
overseas.

 

 
 

Figure 1
Semiannual Incidence of AIDS in the USA
by Year of Diagnosis 1982 - 1086

Number of Cases

 

 

10000 =
= Homo/Bisexual
- IV-Drug Abuser
1000 =
= Both Homo/Bisexual
= and |V-Drug Abuser
_ Tranafusion Recipient
100 = Hemophiliao
10=
phe ppt pt pt pt pt pt ft tt
1978 1980 1982 1984 1986 1988 1990 1992

Semiannual Period of Diagnosis

 

 
Table 2

Annual incidence of AIDS in the United States
among homosexual/bisexual and IV drug abuser cases combined

by date of diagnosis, 1982 to 1986 with projections to 1995."

 

and 1986

 

Number of First Second
Year Cases Ratio Ratio
1982 881
1983 2525 2.8661
1984 5213 2.0646 0.7203
1985 9144 1.7541 0.8496
1986 13609 1.4883 0.8485
1987 (17197) (1.2636) (0.8490)
1988 (18450) (1.0729) (0.8490)
1989 (15807) (0.9109) (0.8490)
1990 (12999) (0.7734) (0.8490)
1991 ( 8536) (0.6567) (0.8490)
1992 { 4759) (0.5575) (0.8490)
1993 { 2253) (0.4734) (0.8490)
1994 ( 979) (0.4019) (0.8490)
1995 { 340) (0.3413) (0.8490)

x Projections, showm in parenthesis, for 1987 and later are
based on the assumption of a constant second ratio in 1985

 

 
 

 

 

Figure 2
AIDS In the USA
Homo/blisexual and {V Drug Abuser Cases

by Year of Diagnosis 1082-19886 with Projections to 1085
Thousands of Cases

 

 

 

20
-——= Projected from Data
% Observed Number
16F
10r
5 =
Oo -}—_f——_}—_+_ +_+_+_ +—_ + +—_ + + + _ +
1980 1984 1988 1992

Year of Diagnosle
Correoted for delayed reports

 

 

 
 

poder nel!

r

REVISED OBJECTIVES OF THE STATEWIDE HIV RISK REDUCTION
AND DISEASE PREVENTION PLAN, MINNESOTA
DECEMBER 1987

The AIDS epidemic continues to grow in Minnesota. As of November 23,
1987, 279 persons had been reported to the Minnesota Department of Health
(MDH) and 159 had died. The MDH estimates that an additional 15,000-20,000
persons in Minnesota are infected with the human immunodeficiency virus (HIV).
These persons are capable of transmitting HIV, even though they may not be
aware of their own infection status. Approximately 1,500 cases of AIDS are
expected to occur in Minnesota by the end of 1990 (1), and it is likely that
every community in the State will eventually be affected. Many of the AIDS
cases that will occur by 1990 are already infected with HIV; however, it is
still possible to impact on future trends in the HIV epidemic. An effectively
implemented risk reduction and disease prevention plan that has broad support
and involvement by all segments of the statewide community is essential to
reduce the tragic morbidity and mortality caused by this disease in Minnesota.

The Commissioner’s AIDS Task Force approved a statewide human
immunodeficiency virus (HIV) risk reduction plan in the spring of 1986. The
original plan has been updated to incorporate the following objectives for

1988.

I. Objective 1: To study the prevalence and incidence of HIV infection in
Minnesota.

This objective will be met by continuing the current system of
ongoing tabulation and follow-up of AIDS cases and patients with positive

test results for HIV infection (repeatedly reactive EIA with a positive

 

 

 
 

 

 

Il.

Western blot, positive HIV antigen test, or positive culture for HIV),
regardless of symptoms. Ongoing tabulation of seroprevalence rates from
the blood banks and the state-sponsored counseling and testing sites can
also provide such information. During 1988, seroprevalence data will be
obtained from patients attending sexually transmitted disease (STD)
clinics in the Twin Cities (the Red Door Clinic and Room 111 Clinic).
Such data will provide the basis for recommendations about HIV screening
among sexually active persons (including gay and bisexual men, and
heterosexuals). In addition, during this time period, seroprevalence
data will be obtained from family planning clinics (testing of high-risk
women), drug abuse treatment facilities and patients with tuberculosis.
In addition, data may be obtained from one metropolitan hospital
(inpatients and outpatients) and neonates born in Minnesota. Such data
will be used to evaluate the existing HIV surveillance system and to make
appropriate recommendations for risk assessment and screening among
persons potentially at risk of acquiring infection. Screening in these
programs will follow guidelines set forth in the document entitled, "The
Use of Human Immunodeficiency Virus (HIV) Antibody Testing in Public
Health," which was submitted by the Task Force for the Commissioner’s

subsequent approval on October 20, 1986. |

Objective 2: To study and evaluate knowledge, attitudes and behavior of
persons at risk for acquiring HIV infection.

High-risk persons include those likely to be exposed to HIV through

sexual contact, and persons who share needles when using intravenous

drugs. To obtain these data, persons attending counseling and testing

sites (CTS’s), family planning clinics, sexually transmitted disease

 

 
 

IIl.

IV.

 

(STD) clinics, and chemically dependent treatment centers can be
surveyed. Also, at-risk participants of AIDS education/safer sex
programs can be given pre- and post-tests to assess these issues. In
addition, studies can be designed through MDH-funded programs to assess
knowledge, attitudes and behavior of other persons at high-risk for HIV
infection (such as persons in correctional facilities who may have

engaged in high-risk activities).

Objective 3; To conduct outreach programs.

Outreach programs are intended to access high-risk persons so that:
1) they perceive their risk of acquiring HIV, 2) persons who are infected
with HIV know ways to limit transmission, and 3) persons who are not
infected can remain so. These programs are specifically intended to
reach persons who may not be identified through .community-based
organizations. Outreach programs include the following: 1) media
campaigns; 2) third-party contact notification; 3) partner outreach
services (a program aimed at helping HIV-infected persons notify their
own partners); and 4) follow-up of persons identified to be HIV-infected
through physicians’ offices, hospitals, clinics and laboratories

(including blood banks). Those programs that are in place will continue.

Objective 4: Conduct programs leading to risk elimination/reduction
through behavior change.

The major mechanism for meeting this objective is through MDH
funding of community-based organizations. | Such organizations should
provide services to persons at increased risk, such as gay or bisexual

men, persons who engage in prostitution, intravenous drug users, persons

 

 
 

 

_ with hemophilia, and sexual partners of persons at risk. On October 1,
1987, the MDH awarded funding to eight agencies in Minnesota to conduct
community-based risk reduction activities. These include: Health Start
(high-risk teenagers), Hennepin County Community Health Department
(training of correctional facility staff and inmates), Lutheran Social
Services (youth in prostitution), the Minneapolis Health Department
(women of childbearing years at high risk), the Minnesota Institute for
Black Chemical Abuse, the Minnesota AIDS Project, St. Paul Division of
Public Health (HIV-antibody-negative clients who persistently engage in
high-risk behaviors), the University of Minnesota Comprehensive
Hemophilia Center, and Women Helping Offenders (women in the correctional
system). Funding for these agencies will be for a 15-month period
(October 1987 through December 1989). Also, ‘the outreach programs
mentioned above (Objective 3) provide counseling to assist in behavior

change.

Objective 5: To provide adult public education.

This objective is ongoing and is achieved predominately through
media efforts, public lectures and seminars, education in the workplace,
and the Minnesota AIDSLINE. Public education is aimed at the prevention
of HIV transmission, and the prevention of “hysteria” related to AIDS and
discrimination against HIV-infected persons. Since some. persons engaging
in high-risk activities do not have access to information directed at
high-risk groups, educational programs for the general public must also

contain information about specific risk activities.

 

 
 

 

 

VI.

VII.

Objective 6: To provide youth education.

Providing education to youth allows impact on behavior formulation,
so that young people in the State can be educated to develop healthy
behaviors and avoid developing behaviors that put them at risk for
acquiring HIV infection. In addition, the MDH will work with other
professional organizations to provide such education. The MDH will
continue to work with the Minnesota Department of Education to assure
that all Minnesota students know how to protect themselves from becoming
infected with HIV. Currently, the MDH is also working to assure that
local community health services (CHS) agencies can respond to the HIV
informational needs of school administrators and educators. The MDH is
also contracting with the Minnesota AIDS Project to develop and pilot

test a peer education approach for youth.

Objective 7: To provide professional education to health-care
providers.

Professional education can be provided through the MDH Disease
Control Newsletter (DCN), and seminars and lectures intended for health-
care professionals. In addition, the MDH will work with other
professional organizations to provide such education, including the
Minnesota Medical Association (MMA), the Minnesota Nursing Association
(MNA), the Minnesota Hospital Association (MHA), the Minnesota Dental
Association (MDA), and the Minnesota Chapter of the Association of
Practitioners in Infection Control (APIC). Professional education should
focus on routine HIV risk assessment and, where appropriate, risk
reduction counseling. Professional education should also emphasize

infection control practices aimed at limited transmission of HIV to

 

 
 

VII.

IX.

health-care providers.

Objective 8: To develop a plan for addressing the disproportionate risk
of acquiring HIV infection among Minnesota's communities
of color.

Currently, a disproportionate number of AIDS cases has occurred
among Minnesota’s communities of color, indicating that such persons are
at increased risk of acquiring HIV infection. Therefore, the MDH is
currently assessing the level of risk and evaluating HIV antibody
seroprevalence in persons of color. In addition, the MDH is developing a
survey to determine current knowledge, attitudes and behaviors regarding
HIV infection in persons of color. Also, programs aimed at risk
reduction and behavior change for communities of color in Minnesota will

be developed. Such programs need to be focused around culturally-based

service organizations.

Objective 9. To assist local public health agencies in developing
community-based plans for dealing with HIV infection in
their area and providing education to their local
populations.

The MOH recently awarded funding to eight local public health
agencies to enable them to develop mode} community education programs.
In addition, the MDH intends to work with other CHS agencies throughout
Minnesota to help them develop community-based approaches for providing
HIV education and services. These approaches need to: 1) develop
consensus in the community for dealing locally with issues related to HIV
infection; 2) provide education to local populations through the use of
media, schools, social and religious organizations, and education in the

workplace; 3) assure that adequate services are available to persons

 

 
 

infected with HIV and to persons with AIDS.

Objective 10: To evaluate the efficacy of risk reduction programs.

It is important to assure that risk reduction efforts reach persons
at highest risk for acquiring or transmitting HIV infection. Efficacy of
risk elimination/reduction programs can be measured through specific and
general outcome, impact and process objectives outlined by the MDH.
Outcome objectives focus on measurable data (i.e., seroprevalence rates
in high-risk populations) for Minnesota in 1991. Process objectives
identify activities to be conducted by the MDH during the current one-
year budget period. Impact objectives specify anticipated results of
targeted educational efforts during the corresponding budget period. All
such objectives involve timed and measurable outcomes. Such objectives
have been determined for current programs, and this evaluation is

ongoing. These objectives are available from the MDH.

 

 

Shultz JM, Danila RN, MacDonald KL, Osterholm MT. The predicted disease
impact of acquired immunodeficiency syndrome in Minnesota: five-year
projection models for a low-incidence state. Minnesota Medicine
1987 ; 70: 203-209.

 

 
 

TESTIMONY PREPARED FOR PRESIDENTIAL COMMISSION ON

Presented by:

Prepared by:

HUMAN IMMUNODEFICIENCY VIRUS EPIDEMIC

DECEMBER 11, 1987

Topic: Heterosexual Transmission

Sheldon H. Landesman, M.D.

Associate Professor of Medicine

Director, AIDS Study Group

State University of New York - Health Science Center at
Brooklyn

(718) 270-2275

Sheldon H. Landesman and the Brooklyn AIDS Clinical Care
Initiative

 

 
 

INTRODUCTION

The focus of my testimony is heterosexual transmission of Human
Immunodeficiency Virus (HIV). This topic has generated much discussion,
many conferences and voluminous literature. Unfortunately, little
Clarity has been shed on the problem. There is widespread confusion in
the mind of the public, governmental and political figures and even
health officials as to the true nature and magnitude of the problen.
During my brief time with you, I will attempt to clarify some of the

confusing or contradictory concepts.

I will discuss nine specific points:

1) Phases of disease spread - heterosexual spread of HIV occurs in

the secondary phase of HIV transmission.

2) There already exists a large pool of infected heterosexual persons

in the country that exceeds a quarter million people.

3) 39,000 - 115,000 persons are already been infected through hetero=

sexual contact.

4) The terminology we use to classify people with HIV disease masks

the potential for heterosexual transmission.

 

 
 

5)

6)

7)

8)

9)

3
The burden of heterosexual transmission of HIV will fall nost
heavily on the poor, and the urban inner city minority women

Heterosexually transmitted HIV is a family disease.

Care, tact, boldness and sense of urgency must imbue our own
attempts to control the epidenic.

AIDS is already in the "general population"

Recommendations

 

 
 

 

ases o isease ead:
1) HIV spread can be divided into three phases - primary, secondary
and tertiary. Heterosexual spread is currently going on in the

secondary phase of disease transmission.

HIV transmission can be conceptualized as occurring in three
phases. Phase one was the introduction and spread of the virus into gay
men and intravenous drug users. The phase may be considered the
primary (1°) phase with gay men and drug users constituting the 1°
infected populations. The secondary phase (2°) of spread is from
infected drug users and bisexual men into their non-drug using female
partners. This 2° phase is characterized by HETEROSEXUAL SPREAD of the
virus from a large population of infected men into their non-drug using
females sexual partners. Although female to male spread of HIV will
also occur in this phase, it is less likely to do so Simply because the
reservoir of infection is largely male (100% of bisexual men and 75% of
drug users are men). The tertiary phase (3°) or "spread out of the risk
groups" occurs when there are a sufficient number of non-drug using
infected females (infected from the secondary phase) to infect or
feedback the virus to non-drug using heterosexual men. These men would
obviously then infect more women and the process could become self-

sustaining.

Confusion arises because of the inadvertent blurring together of
the secondary and tertiary phases of epidemic spread of HIV. What much

of the population and many governmental and public officials mean when

 

 
 

 

5

they refer to heterosexual transmission of HIV or "spread of the virus
out of the risk groups" is the tertiary phase of spread. Tertiary
Spread of HIV cannot proceed until there is a significant amount of 2°
heterosexual spread. Currently, there is little evidence that indicates
wide tertiary spread of HIV. However, the secondary stage of hetero-
sexual transmission e.g. spread from drug users and to a lesser extent
bisexual men into non drug using women and their offspring is occurring
in certain areas of the country and in certain populations. It is a
significant problem now and will be a more serious problem in the near
future. If we hope to prevent areas of the country from being overrun
by the tertiary or third phase of the epidemic we must aggressively

address the ongoing secondary stage of heterosexual HIV transmission.

2) There already exists a large pool of infected heterosexual persons
in the country. The number is in excess of one~quarter million and

they are principally drug addicts.

Cases of AIDS and HIV infection are classified by risk group (e.g.
gay men, drug users, hemophiliacs, heterosexual contact, no identified
risks, etc.) This classification is a means of identifying how persons
acquired the virus. If we look at these risk groups carefully, we see
that approximately 25% of all AIDS cases occur among persons who
themselves are heterosexual, the single largest group being drug users
(19%). In some areas of the country drug users with AIDS account for

more cases than do gay men.

 

 
 

 

6

The point I wish to make here is that when we look at persons with
AIDS or HIV infection we must not only look at how they acquired the
virus but also at how they spread the virus. We de-emphasize the poten=-
tial for heterosexual transmission by focusing on the mechanisn by
which drug users acquired their infection (e.g. - sharing of needles).
We must remember that most drug users are male heterosexuals and can
thus spread the virus two ways - through heterosexual contact and by

sharing of needles.

If 25% of the AIDS cases are in persons who are heterosexual, then
at least 25% of the reservoir of HIV infected are heterosexual. The
CDC report to the Domestic Policy Council estimates that there 900,000
~ 1.4 million infected persons in the U.S. Thus, approximately 25% of
this pool of infected persons are heterosexual (225,000 to 420,000
persons Table 14, CDC report). This is large reservoir of heterosexual
HIV infected persone matter how you look at it. The majority of this
pool is composed of drug users or persons infected heterosexually from
drug addicts. Keeping in mind that infection is by and large a per-~
manent condition, we are currently faced with a large reservoir, equal
to or exceeding 250,000 chronically infected heterosexuals. The impact
of this population on future generations of women and children is just

starting to be felt.

For the rest of this presentation, I will focus on drug addicts
since they are the largest reservoir of heterosexuals with HIV infec-

tion. The majority of drug uses are male and the majority (70%) of

 

 
 

7
their partners are non druq_using women. Thus, the infected male drug ,
user, is and will continue to be in the foreseeable future, the
principal vector for spread of the virus into non-drug using women. In
its November 30th report to the Domestic Policy Council, the cpc
estimated that there are 900,000 regular drug users of whom 25% are
infected with HIV (225,000) and 200,000 occasional drug users of who 5%

are infected (10,000).

3) Current estimates are that 30,000 - 115,000 persons have already
been infected through heterosexual contact.

What is the consequence of having such a large reservoir of
infected heterosexually active persons? A natural consequence is the
infection of women who live in close social proximity to these men and
who interact sexually with them. This brings us to the problem of HIV
infection in women and children. In our own hospital, Kings County
Hospital Center, in Brooklyn, New York, 2% (1 in 50) of all women who
give birth are infected. Approximately half of these women are infected
via heterosexual activity, the remainder being drug users themselves.
At other hospitals in New York City, the rate is even higher. At Emory
Hospital in Atlanta, GA, more than 13% of women who deliver are in-
fected. In a recent study in Massachusetts, the rate of infection
among women who recently delivered ranged from approximately 1/1000 in
rural areas to 1/100 in inner city areas. In San Juan, Puerto Rico,

1.7% of women coming to prenatal clinics are seropositive, while in

 

 
 

 

- 8
Alameda, California, 0.5% of women tested positive. Other studies of a
revealing nature include a nearly 1% rate of HIV positivity in a
Baltimore Family Planning Clinic and a 0.7% rate of positivity in

prenatal women at a university hospital in Jacksonville, Florida.

These numbers reflect a mixed population of women, some of whon
acquired their infection via drugs, some of whom were sex partners of
infected men. It is important to realize that infected drug using women
will comprise a decreasing proportion of infected women over tine while
the number of women infected through heterosexual transmission will
gradually and inexorably increase as a consequence of the sexual
activity of the large number of infected males. Estimates derived from
CDC document indicate the numbers of persons infected through hetero-
sexual contact is already significant. These estimates are in the
range of 30-115,000 persons. This number thus represents the current
extent of the secondary phase of HIV Spread and defines the current

magnitude of the heterosexual transmission problen.

t

4) The terminology we use to classify people with HIV infection and
AIDS masks the potential for heterosexual transmission.

The existence of more than 250,000 people infected heterosexuals
and 30-115,000 persons infected through heterosexual contact is
partially masked by the terminology with which we describe the epi-

demic.

 

 
 

9

For example, it is often stated that HIV disease is not spreading
outside of the risk groups. While this statement is technically
correct, it contains a basic flav. Specifically, persons who are
heterosexually active with another risk group member (e.g. bisexual man
or drug users) and acquire the disease are themselves defined as being
in a risk group. Thus, partners of drug users, female sex partners of
bisexual men, female sex partners of hemophiliacs, etc. constitute a

risk group.

Our terminology then, hides the concept of heterosexual transnis-
sion. By classifying sex partners of bisexual men and drug users as a
risk group, we count the cases that occur in this manner, state that
they still are in a risk group and then assure ourselves that the
disease is staying within our self-defined risk groups. This may calm
the population but does little to slow the heterosexual spread of the

virus.

5) The burden of heterosexual transmission of HIV will fall most
heavily on the poor, the indigent and the urban inner city

minority women and their children.

As mentioned earlier, heterosexual transmission is inextricably
linked to drug users. The majority of the drug using population reside
in the large inner city areas of this country. This population is
largely composed of Black and Hispanic people. Thus, the problem of

heterosexual transmission of HIV is largely but not exclusively a

 

 
 

 

10

problem of urban minority populations. The population at greatest risk
for heterosexual transmission of HIV is the young inner city minority
women who live in close social and sexual contact with the infected
male drug users. Infection to this population is a terrifying addition
to equally depressing already existing conditions of poverty, unemploy-

ment, discrimination and drug abuse.

The threat to this population is large. As noted earlier 30-
115,000 persons are estimated to have been infected through heterosex-
ual transmission the largest number being people of color. As the AIDS
epidemic continues its spread, it is this population that will suffer.
The rates of seropositivity in pregnant women quoted earlier gives
stark evidence to the fact; most of the studies cited earlier are from
inner city populations are composed almost exclusively of people of
color. Blacks and Hispanics clearly represent a disproportionate number
of AIDS cases. Excluding homosexual men, the likelihood of acquiring
AIDS if you are Black is 12-13 times that of a white person while

eighty percent of all children with AIDS come from minority families.

In our own city of New York, the problem is most extreme. In the
South Bronx area of the city where drug use is common, Dr. Ernest
Drucker from Montefiore Hospital has performed studies and has calcu-
lated that 8-20% of the sexually active males in that area are in-
fected. The risk to the sexually active woman in this area is immense.
Areas of northern New Jersey have comparable figures. In other cities,

the problem is less severe, but nonetheless present. Every city that

 

 
 

 

11
has a serious drug use problem is under siege by HIV. The situation in
NYC should give solace to no one. Rather, it is a message that efforts
must be redoubled elsewhere so as to prevent the problem from reaching

a comparable order of magnitude.

6) Heterosexually transmitted HIV is a family disease.

Heterosexual transmission of HIV has many severe derivative
consequences; the most important being the creation of what I call a
"family disease". Preeminently among diseases HIV infection within the
heterosexual community is a family affair. The usual scenario is that
of an ailing or dead infected male drug user who has infected his non-
drug using wife or girlfriend. The wife or girlfriend may have already
given birth to an infected child or may give birth to one in the
future. Often, the infected woman is already clinically ill and usually
has two or three young children at hone. Thus, we are faced with a
seriously ill male, largely unable to care for himself, an infected or
ill partner with young children at home who. If the children are lucky,
they might be free if HIV infection, but they still suffer the serious
effects of having ill and or absent parents. All of this generally
occurs within an environment and social circumstance of extreme
poverty, general poor health, inadequate housing, a fragile family
structure and the near total absence of Significant resources or

Support systems.

 

 
 

 

12

It cannot be emphasized too strongly that HIV destroys families,
father, mother and children. The children caught in this situation lose
on two counts; either they are born infected (1/3 to 2/3 of children
born to an infected mother will themselves become infected) or their
infected parents will get ill and the children often end up in foster
care institutions, as boarder babies in the hospital or some other

secondary home setting.

Added to all of the above is the misery and pain associated with
HIV infected children. The Coolfant report, published in 1986, es-
timated that there would be 3,000 cases of pediatric AIDS in 1991.
Since there are 3-4 HIV infected infants for every case of AIDS, we can
anticipate 10,000 cases of perinatal HIV disease by 1991. The problem
of perinatal or pediatric AIDS, already severe will only get worse with
time. The reservoir of infected heterosexual males and the fact that i1-
3% of pregnant women in many inner city locations are currently
infected tells us that pediatric and perinatally acquired HIV disease

will be an enormous problem in the upcoming decade.

7) Care, tact, boldness and sense of urgency must imbue our own
attempts to control the problem if we are to have even a nininal

chance at success.

Reaching the communities most affected by heterosexual trans-
mission will not be easy. The minority communities are justifiably

concerned about their being labeled and stigmatized as another "risk

 

 
 

13
group". They fear that any campaign directed towards HIV counseling and
testing could turn into a eugenics program with forced sterilization or
coerced abortion. Both of these fears have a strong basis in reality;
the discrimination and stigmatization surrounding populations at risk
for HIV is well known; less well known but well documented is the
history of 100,000 forced sterilizations imposed on minority popula-
tions in earlier years. Further, these communities are concerned that
the condom or sexual abstinence campaigns, designed to slow transmis-
Sion represent subtle or not so subtle campaign of racial genocide and

denials of procreative rights.

A total lack of understanding of the social and sexual mores of
minority populations increases our difficulties. We know little enough
about the sexuality of white middle class teenagers and young adults
(e.g. in the recent CDC document the sexual study most often cited was
the Kinsey Report of 40 years ago); we know less about the sexuality

of the urban inner city populations.

In approaching the problem of heterosexual transmission of HIV
infection in the ghetto areas of our country, understanding and
boldness will be required by all concerned. The power structures and
governmental authorities who allocate money and set policy for this
problem must understand that the Black and Hispanic population will
approach any governmental solution with caution if not outright

suspicion and hostility. Patience will be needed.

 

 

 
 

 

1

. 14

The social and political leaders of these communities must not
underestimate the threat that HIV poses. to their communities nor must
they shy away from meeting the problem head on. Once the level of HIV
infection in females reaches a significant but as undefined level,
tertiary spread of the virus can occur. We already have 1-3% of females
of child bearing age being infected in certain cities. If the level
continues to rise, the consequences are dire. If HIV infection progres-
ses much further, the issue of altering sexual behavior and use of
condoms will not be one of racial genocide, rather, it will be one of
racial survival. Thus, empathetic and sensitive leadership is required
by government as they fashion their programs but bold forward planning

4

is also required by the key leaders of these affected communities.
8) AIDS is already in the “general population*®

All persons involved in the AIDS epidemic talk of "risk groups"
(gay men, drug users, hemophiliacs, etc.) as if they are separate and
apart from the general population. Commonly, I have heard government
and public health officials state that AIDS is not spreading to the
general population (general population refers to white middle class
heterosexual individuals) but in staying within the risk groups. These
statements, put forth with the best of intentions, mislead the public

in several ways.

First, all persons who are infected with HIV are part of the

general population. Is there any doubt that the estimated 900,000 to

 

 
 

 

15

1,400,000 infected persons are part of the everyday ordinary population

of the country? They may differ from us by race, sexual orientation or
economic status, but they are the general population. They may not be
your next door neighbor or previous sex partner but they are part of

this country.

Second, the concept that HIV disease is still within risk groups
and not yet in the general population fosters the concept of the
infected versus the uninfected. It fosters within the mind of the
public the concept that they are infecting us, it increases the
public's concern that sometimes turns into hysteria; most of all it
prevents rational discourse about the problem and its proposed solu-
tions. The concept of the general population versus risk groups fosters
divisiveness within our society and divides us at a time when we must

unite.

Divisiveness is especially troublesome when discussing hetero-
sexual transmission. We must recognize that HIV disease already exists
in the general population, not in the general population as defined as
white middle class, but the general population as defined by persons
who live in this country. Failure to correct this misuse of language
can only result in increased fear on the part of white middle class and
increasing severe coercive attempts as they, the uninfected seek to
protect themselves from us, the infected. As we attempt to slow the
transmission of HIV further into the heterosexual population, we must

realize that we are all in the same boat and will either sink or sail

 

 
 

 

together.

16

 

 
 

 

17

Recomendations can be placed in three categories: Provision of care,

research, and education.

9)

Recommendations:

Provision of Care

1.

Additional resources must be targeted toward the drug users: this
must be our first priority. We can never hope to stop the epidemic
until we slow transmission among and by drug users. We must place
addicts into counselling and drug treatment centers where educa-
tion concerning needles and heterosexual transmission can be
intensely and repeatedly applied. Other speakers will address in

greater detail the needs related to drug users.

One point to remember within the context of resource allocation
for drug users is that they are not the popular persons in this
country. Nonetheless, we treat and care for them because our
societal ethic demands that we care for all our citizens. Further,
it is critical to understand that we must provide resources for
the drug users if for no other reason than our own narrow self=

interest as we attempt to stop heterosexual transmission of HIV.

Regional comprehensive care programs for women and children
infected with HIV should be created. These programs would: (1)
educate women at risk, (2) provide linkage between community based

support programs and hospital based prograns, thereby lessening

 
 

 

18
the fragmentation of health care so common in the inner cities,
(3) stabilize the fragmented and (4) decrease the number of
homeless infants (boarder babies) and children born to HIV

infected mothers.

Education

3. Culturally appropriate educational programs should be targeted to
the female populations at highest risk of HIV disease. Including
widespread availability of volunatry HIV counseling and testing.
Such programs must be accompanied by additional support services;
you cannot tell a 23 year old woman she is pregnant and infected
and just leave the office. Campaigns advocating increased condom
use are useful; but a broader approach is needed. Especially in
the inner city areas women must be made aware of the high risk
attendant to sexual activity, particularly with a IVDA. A message
of sexual selectivity must be incorporated into educational
programs. The recent debate of condom promotion versus promoting
abstinence is a false issue. It is possible to steer a middle

course.
4. Educational campaigns and grants should be ‘subject to critical
analysis as to their effect and impact. Exhortation, history
tells, rarely changes behavior. Behavioral scientists, including
anthropologists and sociologists should be drawn into the creation
and evaluation of education programs. We must learn if the message

sent is the message received and if received, is it acted upon?

 

 
 

5.

 

19

More information is needed about heterosexual transmission of HIV

~ its incidence, prevalence, causative factors (behavioral

and

biological) and prevention. Additional behavioral, biologic and

epidemiologic research in this area should be a priority. As an

example, studies on the seroprevalence of HIV infection in women

should be continued. The currently planned national study of HIV

infection in new mothers (done by testing a drop of blood from

newborns) uses a relatively unbiased methed to give a true picture

of HIV infection in childbearing women. This type of study will be

very useful to public health management and HIV surveillance.

 

 
 

-~

Original Contributions ——

Serosurvey of Human Immunodeficiency

Virus Infection in Parturients

Implications for Human Immunodeficiency Virus

Testing Programs of Pregnant Women

Sheldon Landesman, MO; Howard Minkoff, MD; Susan Holman, RN, MS; Sandra McCalla, MD; Odalis Sijin, MD

Although perinatal transmission of the human immunodeficiency virus (HIV) is
well documented, seroprevalence rates of HIV in populations of women of
reproductive age have not yet been reported. To determine the seroprevalence
0” \childbearing women from a population with a high incidence of acquired
it eficiency syndrome, cord blood samples were collected from 602
infants delivered at an inner-city municipal hospital in New York. Demographic
and HIV risk factor information was also collected from mothers of these infants.
Twelve (2%) of 602 samples (95% confidence interval, 1% to 4%) were positive
for HIV on enzyme-linked immunosorbent assay and Wester blot analysis. In
interviews, seven of 12 seropositive women had risk factors as defined by the
Centers for Disease Control, Atlanta; the remaining five seropositive women
had no self-identified risk factors. The HIV seroprevalence rate in our hospital
(2.0%) is several times higher than that of many other diseases for which
- screening is already routine. This serosurvey indicates that HIV infection of
inner-city parturients is a significant problem that warrants broadly implemented
health strategies. Furthermore, the data also suggest that if risk factor informa-
tion elicited by physicians is used to initiate HIV antibody counseling and testing
of pregnant women, a significant number of seropositive parturients is missed.
In areas with significant seroprevalence rates of HIV infection, a broader
counseling and testing program may be needed.

(JAMA 1987;258:2701-2703)

 

RECOMMENDED strategies to re-
duce perinatal transmission of the
human immunodeficiency virus (HIV)
depend on counseling and antibody test-
ing programs directed toward par-

 

From the Deparunents cf Meciene (Or Landesman
and Ms Honan) and Oosteines and Gynecology (Drs
Minee and McCailla), State University of New York
re nce Center at Brooklyn. and the Department
c 23 and Gynecology. New York University (Or
S

fey.nt requests to Box 56. SUNY-Hearth Scence
Center at Brooktyn, 450 Clarkson Ave, Brookiyn. NY
11203 (Or Landesman). :

" JAMA, Nov 20, 1987—Vo! 258, No 19

turients who are willing and able to
identify themselves as “at risk.” The
scope and focus of these programs, in

 

See also pp 2693 and 2736.

 

turn, must be predicated on accurate
seroprevalence and demographic data.
There are few data extant, however, on
the seroprevalence of HIV infection
in childbearing women. Furthermore,
there are no data on the seroprevalence
of HIV among sexually active women in

inner cities, where a large number of
men who are intravenous drug abusers
live in close sexual and social contact
with a pool of women of reproductive
age. These men, through heterosexual
encounters, could potentially infect
women who might not realize they have
engaged in risk behavior and, hence, .
would not report a risk factor at a prena-
tal or family planning clinic. Failure to
self-report could in turn reduce the effi-
cacy of counseling and testing programs
directed only toward self-identified
members of risk groups. Since public
health policies related to HIV testing of
parturients in the inner city depend on
seroprevalence rates and the demo-
graphics of infection, the following stud-
ies were conducted to collect this
information.

METHODS

Kings County Hospital Center
(KCHC) is a 1200-bed municipal hospi-
tal in Brooklyn, New York. It provides
medical care for a large inner-city
minority population, including a sizable
Haitian community (500 to 600 deli-
veries a year at KCHC). There are 5000
deliveries a year at KCHC and an equal
number of planned abortions. Current-
ly, HIV testing af pregnant females is
not available at the hospital.

Preliminary Study

In April and May of 1986, we under-
took an anonymous random serosurvey
of 359 cord blood samples obtained from
approximately 600 newly delivered in-

HIV Infecton—Landesmanetal 2701

~

 

 

 
 

 

Table 1.—Summary of HIV*® Cord Blood Subjects

 

Ethnic Group Na. (%) of Total
White 4 (0.7)
Hisparec 109 (18.1)
Black Amencan 217 (36.0)
West indian 138 (22.6)
Endemic arsat 69 (11.5)
Othert 49 (8.1)
Unicnown 18 (30)
Total 602 (100)
2

*HIV indicates human immunodeficiency virus.
Trait or black Africa,
$Tharty-eught of these 49 subjects were from Guyana.

fants. In this preliminary survey, 1.7%
of the samples were positive for HIV on
enzyme-linked immunosorbent assay
(ELISA) and Western blot analysis.
Three additional samples were con-
firmed to be positive but were not
counted because they were collected
from women known to have risk factors
and therefore represented an accrual
bias. Had these three samples been in-
cluded, the seroprevalence rate would
have been 2.5%. Because our initial
serosurvey was too small and lacked
prospectively collected risk factor data,
we performed the larger study de-
scribed below.

Current Study

Between Dec 8, 1986, and Jan 31,
1987, cord blood samples were obtained
at the delivery of children born at
KCHC. The serum from these samples
was stored at -70°C until antibody
testing was done.

Risk factor data for HIV infection
were obtained from all newly delivered
mothers by means of an interview.
House officers and senior medical
students specially trained to ensure
accurate acquisition of risk factor infor-
mation collected the data Training
included meetings and discussions with
our special perinatal study staff [asnes
concerning privacy, guult, definition of
risk factors, etc, were discussed and
reviewed. The interview took about ten
to 15 minutes; extensive counseling did
not occur. Women who expressed inter-
est in additional information concerning
HIV disease were referred to an HIV
counseling and testing site. After the
demographic and msk group data were
obtained, they were matched to the cord
blood sample. All identifiers were then
removed, and HIV testing was per-
formed.

We performed ELISA testing by
standard methods using a licensed
ELISA kit (Electro-Nucleonics, Co-
lumbus, Md). All intial positive resuits
were confirmed by Western biot ana-
lyses. A minimai positive blot consisted
of a p24 band with one additional band,
such as p55 or p41. All positive blots had
multiple positive bands.

2702 = JAMA, Nov 20, 1987—Vol 258. No 19

 

Table 2.—Fisk Factors
Le
No. af
Alek Fector Patients
Wiravenous drug abuse 23
Sex partner of intravenous
drug abuser a
Sex parmer bisexual,
oF sex partner received transfusion 6
From endertec area, or sax partner trom
endemic aren 76
Transfusion 37
Totai* 165
ee

“Seventeen women had two risk factors. One hun-
cred iorty-enght (24.6%) of 602 women had risk factors.

RESULTS

During the study period, 761 women
delivered. Cord samples for serologic
study were obtained from 673 patients.
Demographic data, risk factor informa-
tion, and HIV serologic findings are
available for 602 patients. The remain-
ing women either had no serum samples
or were not interviewed either because
they delivered on a weekend or because
the study personne! were on vacation.
No attempt was made to select or specif-
ically collect samples from parturients
at high risk. The 602 patients for whom
we have HIV serology results and inter-
view information constitute our study
population. The characteristics of the
population are listed in Table 1.' Eighty-
four percent (84%) of our study popula-
tion was black, Caribbean, or Hispanic.
The ethnic breakdown of our study pop-
ulation reflects the population seen at
the KCHC prenatal clinics.

Of the 602 women, 148 (24.6%) had a
self-identified msk factor (Table 2).'
Among the women with risk factors, 23
were past or current intravenous drug
users, and 23 had sex partners who used
intravenous drugs. Thus, 46 (7.6%) of
602 women had self-reported risk fac-
tors directly attributable to drug use.
Seventy-six women (12.6%) had a sexu-
al partner from or were themselves
from an area where the acquired immu-
nodeficiency syndrome (AIDS) is en-
demic. Many of these women came from
Haiti; a few came from Central Africa.

Of the 602 samples tested, 12 sera
(2.0%) were positive for HIV on both
ELISA and the Western blot analysis
(95% confidence interval, 1.0% to 4.0%).
Table 3’ summarizes the data for the 12
seropositive individuals. Seven of these
12 women had risk factors for HIV
acquisition elicited by intervew. In-
cluded among these seven were a wom-
an from Hait!, a woman with 30 sexual
partners, and a woman who received a
single blood transfusion in 1982. None of
the three perceived themselves to be at
risk, but all were classified as such
based on the categorization of risk
groups by the Centers for Disease Con-

Table 3 —Patients With Cord Blood Samples
Seropositive for HIV"

 

eum
Subject NeJ/
Age, y Ethinie Group Risk Factor
128 Fisparuc None
22s Hispanic None
¥25 Black American intravenous
drug abuser/sax
partner of
wiravenous
. drug abuser
4a West indian None
sx Glack American intravenous
drug abuser
6/32 Heltian Endeuc area
7/26 Black Amencan intravenous
Grug abuser/sex
partner of
intravenous
drug abuser
af21 Black Amencan Multole sax
partners
3/21 West indian Transfusion
10/22 Black Amencan None
1V27 Hisparuc intravenous
drug abuser
12/19 Black Amencan None
ee

“HIV inchcates human immunodeficency virus

trol (CDC), Atlanta.“ The rate of sero-
positivity for women with risk factors
was 4.7% (7/148). Five of the remaining
454 women were seropositive, indicat-
ing a seroprevalence rate of 1.1% in the
group with no identified risk factors. It
is important to note that while seven of .
the 12 seropositive patients were in self-
identified msk groups, the remaining
five (42%) were not.

COMMENT

In our hospital, 2.0% of women who
delivered were infected wth HIV.
Among Haitian women, one of 67 was
seropositive. Thus, if all Hartian women
were removed from the study popula-
tion, the seroprevalence rate would still
be 2.0%. To our knowledge, there are no
published rates of seropositivity from
comparable hospitals in the United
States. In Africa, however, 8% of births
in Kinshasa, Zaire, and 2.0% of births in
Nairobi, Kenya, are to seropositive
women.’

It is reasonable to assume that the
seroprevalence of HIV at KCHC is not
unique among certain hospitals in New
York or other areas where AIDS is
more commonly diagnosed in women
(ie, Newark, NJ, Miami). If cases of
pediatne AIDS are indicative of the
prevalence of HIV infection among
women in the reproductive age range,
then Brooklyn ranks third among the
boroughs of New York, since it trails
the Bronx and Manhattan in cases of
pediatric AIDS per 100000 persons.‘
Unpublished data from Bellevue Hospi-
tal in Manhattan show a 3.7% seropre-
valence rate among partunents (K.
Krasinski, MD, W. Borokovsky, MD,
oral communication, June 1987).

HIV Intection—Landesman et a!

 

 
 

It is probable that five of the seven
s°--sositive women with risk factors
‘d their infections nonsexually
‘ rug addicts and one blood trans-
fusion). The other two patients with risk
factors, one from Haiti and one with
multiple sex partners, probably ac-
quired their disease sexually. We are
less certain about the five seropositive
women who had no risk factors. They
may have chosen not to admit to a risk
factor (eg, intravenous drug use, a sex
partner who used drugs) because they
were suspicious of institutions or were
reluctant to admit to illegal or socially
unacceptable behavior. Alternatively,
they may simply have been unaware
that their sex partner was infected. In
either case, these data imply that a sub-
stantial number of infected women are
not identified by self-reporting or brief
interviews.

In our institution, self-reporting and
physician interviews only identified
seven (58%) of the 12 potentially sero-
positive patients. Thus, if our physi-
cians had followed the current recom-
mendations of the CDC,' asked patients
about risk factors, and recommended
that those at risk be tested, five (42%) of
the seropositive women would have
gone undetected. It is possible that a
r trained interviewer who spent a

cime with each patient might be
a... w elicit risks that were not detect-
ed by our clinicians. However, the logis-
ties of spending more than 15 minutes
with every prenatal patient to deter-
mine who has HIV risk factors and
needs further counseling and testing
may be prvhibitive, especially in set-
tings with a high patient volume, such
2s municipal hospitals. Furthermore, as
HIV continues its spread into the het-
erosexual population, a patient's aware-
ness of risk status may be reduced to the
point where even in-depth interviews
will not suffice.

We believe that HIV testing should
be routinely offered to all pregnant

JAMA, Nov 20. 1987—Vol 258, No 19

women in a situation where there is a
combination of high seroprevalence and
poor identification of people at risk.
Such testing must be accompanied by
counseling, consent, and appropriate
provisions for privacy.

The relevance of a 2.0% HIV antibody
prevalence rate to standards of prenatal
care and counseling can best be gauged
by comparing the frequency of perina-
tally transmitted AIDS with that of
other perinatal diseases for which
screening standards already exist. For
example, the incidence of perinatally
contracted herpes is approximately one
in 7500 to 20000 births,’ congenital
rubella occurs once in 300 000 births,‘
and neural tube defects occur once in
1000 births.’ Assuming, conservatively,
a 33% transmission rate of the HIV,*”
the incidence of perinatally acquired
HIV at our hospital is approximately
one in 150 births. Although $56 million is
spent annually to perform cultures for
herpes on women," rubella antibody
Studies are routine in many prenatal
clinics, and testing for maternal serum
a-fetoprotein in pregnancy is becoming
standard despite its 95% false-positive
rate, no comparable HIV counseling
and testing program currently exists in
any hospital despite the much greater
prevalence of perinatally acquired HIV.

Hospital facilities throughout the
country, in areas of low and high preva-
lence rates of HIV infection, should per-
form similar surveillance studies in
obstetric and family planning clinics,
sexually transmitted disease clinics,
and abortion clinics. A locally deter-
mined and continually reassessed sero-
prevalence rate can then be utilized for
the creation of an appropriate HIV
counseling and testing program direct-
ed at women of childbearing age. Such
programs will be useful in bringing
information about HIV serostatus to
women before they become pregnant or
early in their pregnancy.

Since this article was written, the

 

CDC has published revised guidelines
for counseling and antibody testing.
The new guidelines expand the defini-
tion of women at risk for HIV infection
to include persons who “are living in
communities or were born in countries
where there is a known or suspected
high prevalence of infection among
women.” According to the guidelines,
“these women should be routinely coun-
seled and tested for HIV antibody.” Our
study population, which had a 2% rate of
positivity, would appear to qualify
under this expanded definition.

We thank Anne Willoughby, MD, from the
National Institute of Child Health and Human
Development for her time and effort in reviewing
the manuseript.

References

i, Recommendations for assisting in the preven-

tion of perinatal tranamismon of human T-lympho-

trophic virus type (111) lymphadenopathy assoaat-

ed virus and the acquired immune deficiency

syndrome. MMWR 1985;34:721-731,

2. Heterosexual tranamission of human T-lympho-

trophic virus type [1] Aymphadenopathy-assonat-

ed virus. A/MWR 1985;4:561-563.

3. Quinn TC, Mann JM, Curran JW, et al: AIDS in

Africa: An epidemiologic paradigm. Science

1986;234:955.

Survellance Repricaord tear et AlDS

pru 1987

5. Moni G, Hardt W: Management af herpetic

vulvovaginitis in pregnancy. Semin Perinatol

1985;7:16-23,

@ Rubella and congenital rubells syndrome—

N.Y.C. MMWR 1886:35:770-778,

i. Main D, degees id oan oe
prenatal diagnosis and counseling. Obstet Gyne-

col 1986;67:1-16.

& Rubinstem A: Pediatric AIDS. Curr Prodi

Pediatr 1986;26:362-409.

9. Minkoff H, Nanda N, Mendez R, et al: Pregnan-

10. Parks WP, Scott GB: An overview of pediatric
AIDS: Approaches to diagnoms and outcome
assessment, in A/DS: Modern Concepts and Ther-
apeuhe Challenges, New York, Marcel Dekker
Ine, 1987, pp 98-121.

1. Binkin NJ, Koozrov JP, Cates W: Preventing
neonatal herpes. JAMA 1984:251:2829,

12. Public Health Service guidelines for counseling
and antibody testing to prevent HIV infection and
AIDS. MMWR 1987:36:509-515.

HIV Infecton—Lancesmanetal 2703

-

 

 
 

 

Fditotadls ——EEyEyEylEEEEEEEEEEEEEEE

Facing the Complex Issues of Pediatric AIDS:
A Public Health Perspective *

A simple child,
That lightly draws tts breath,
And feels its life in every limd,
What should tt know of death?
William Wordsworth, 1798

Almost seven years have passed since the first recognition of
the acquired immunodeficiency syndrome (AIDS) in this
country, and during the past four years we have been able to
identify many of the clinical and epidemiologic consequences
of human immunodeficiency virus (HIV) infection. Although
children with AIDS were recognized early in the epidemic,
the impact af pediatric AIDS on clinical medicine and society
in general is only now being fully appreciated. In this issue of
THE JOURNAL, Landesman et al’ provide us with important
insights into the pediatric HIV infection problem of today and
the AIDS qroblem of tomorrow through their survey of
women delivered of infants at King’s County Hospital Center
in Brooklyn, NY.

See also p 2701.

 

In their study of HIV seroprevalence in cord blood samples *

from 602 infants delivered in an inner-city municipal hospital,
Landesman and colleagues found a demographic and HIV risk
factor history similar to what is currently reflected in re-
ported cases of AIDS in this country. All of the mothers were
part of a large inner-city minority population (ie, persons of
color). Their finding of a 2% infection rate among the women
who gave birth is similar to that reported from some hospitals
in central Africa.’ In addition, in their study, 42% af the HIV-
seropositive women did not acknowledge or could not identify
a recognizable risk factor for their HIV infection, suggesting
unrecognized heterosexual transmission. These findings,
along with studies in other metropolitan areas of the country,
demonstrate an emerging shift in the sociodemographic
pattern of HIV infection. Although this study provides
information only on the number af infants born with HIV
antibody-positive cord blood, we can expect that between
30% and 65% of such children will be infected with HIV.“
The results of this study are sobering. Unfortunately, many
of us in public health may take comfort in recognizing that
these observations come from New York City and thus have
only limited implications for the remainder of the country.

2736 = JAMA, Nov 20, 1987—Vol 258, No. 19

'

However, we must remember that early in the HIV epidemic,
many practitioners in clinical medicine and public health
believed that AIDS in gay men would be an important prob-
lem only in New York City, San Francisco, and several other
metropolitan areas on the east or west coasts. Today, the
“Heartland” is repeating the experience of those coastal
metropolitan communities with regard to AIDS in gay men.
We need to recognize that HIV infection in women and the
resultant cases of pediatric AIDS will likely unfold as a similar
story in selected urban populations throughout the United
States.

As of Sept 30, 1987, five hundred eighty-four children 13
years of age or younger with pediatric AIDS as defined by the
Centers for Disease Control (CDC) case definition criteria
were reported in this country. Of these, 177 (30%) were re-
ported after Jan 1, 1987. As many as 2000 additional children
currently are known to have symptoms of HIV infection but do
not fit the specific CDC AIDS case definition.‘ The Public
Health Service estimates that by 1991, some 3000 children
will suffer from the disease, and without new efficacious drug
therapy virtually all will die.

Pediatric AIDS cases currently account for only 1.4%
(584/42 354) of AIDS cases reported in the United States.
While this is a relatively small proportion of all cases, we
believe that these numbers will grow. Also, control of HIV
infection in the pediatric population raises many difficult
issues that touch the core af the social fabric among America’s
urban poor, especially communities of color. These issues
include poverty; drug abuse; teenaged pregnancy; prostitu-
tion; and inadequate social services, education, and health
care systems. To address HIV infection in this population, we
must also address these complex social issues with a multidis-
ciplinary approach.

The first step in any program aimed at primary HIV pre-
vention must be to identify the population at risk. Seventy-
eight percent of all cases of AIDS in children younger than 13
years of age and reported to the CDC had a parent with, or at
risk for, AIDS. Also 78% of reported AIDS cases in children
younger than 6 years of age (457/584) were in children of color.
In contrast, children of color constitute only approximately
21% of our nation’s population in this age group. To understand
this striking disparity in the risk of pediatric AIDS by race,
one must only look at the risk factors for adult AIDS by race.,

Editonals

 

 
 

As of Sept 30, 1987, 80% (5511/6885) of all reported AIDS
cases in intravenous (IV) drug abusers have been in persons of
color; similarly, of all reported cases related to heterosexual
transmission, 83% (1386/1660) have occurred in persons of
color. These data do not represent “just” the risk of IV drug
abuse in selected metropolitan areas such as New York City or
Newark, NJ. For example, in Minnesota, part of the “Heart-
land,” we too have documented an emerging, but similar,
increased risk of AIDS in persons of color. Of 260 cases of
AIDS reported to the Minnesota Department of Health
through Sept 30, 1987, 12% (82 cases) have been reported in
persons of color; in contrast, persons of color account for only
3.5% of the state's population.

With this background, what are the implications for pris
mary prevention of pediatric AIDS? We must first realize in a
straightforward but compassionate manner that the great
weight of pediatric AIDS will continue to fall on our communi-
ties of color, particularly in the inner-city areas. It will be
important for these communities to take ownership in devel-
oping creative solutions to this problem, with public health
support, and to begin addressing its many complexities.
Thus, if we are to have any impact with HIV primary preven-
tion programs among potential mothers, we will also have to
address the issues of IV drug abuse, the high rates of unwant-
ed or teenaged pregnancies, poverty, and lack of education.
Particularly, these issues should be raised in younger age
groups where the developing behavior patterns can be influ-
enced, rather than attempting to devote all resources to
behavior change among adults and older adolescents.

These social problems are of dramatic proportion. After
more than 70 years of organized attempts by publicly and
privately sponsored intervention programs to address the
issue of IV drug abuse in this country, we find the estimated
number of individuals who share needles for purposes of
injection at an all-time high. At the same time, we continue to
lead the industrialized world in the rate of teenaged preg-
nancy and birth rates. In particular, the rate of black, never
married women aged 15 through 19 years who are sexually
active is almost 35% higher than that of white women. While
government, private foundations, and concerned citizens’
groups have made numerous efforts through the years to
address these issues, success has generally not been forth-
coming. It is likely that new strategies to address these
problems will be needed. Also, we recognize that programs
aimed at social change require extensive economic and human
resources. However, the added burden in cost and life lost to
our society as a result af high AIDS incidence rates in inner-
city populations should be taken into account when examining
the cost of programs for social change that may impact on HIV
transmission. Allocating resources now for such programs
may prevent a much greater cost in the future.

We acknowledge that the social and economic reality for
women af childbearing age in our inner cities cannot easily be
changed; therefore, we must also consider the more limited
traditional public health strategies of disease prevention and
control. The CD issued recommendations for assisting in the
prevention of perinatal transmission of HIV infection in De-
cember 1985.° Emphasis was placed on counseling and testing
for HIV antibody in those populations of women with in-
creased prevalence of HIV. These include women who have
used drugs intravenously for nonmedical purposes; women
who were born in Guntries where heterosexual transmission

JAMA, Now 20, 1987—Vol 258, No 19

 

is thought to play a major role; women who have engaged in
prostitution; and women who have been sexual partners of
men who abuse IV drugs, are bisexual, have hemophilia, or
have evidence of HIV infection. Landesman and colleagues
have concluded in their own area that these recommendations,
when applied to their hospital-based population, indicate that
counseling and HIV antibody testing should be routinely
offered to all pregnant women. We would agree with them and
also agree that public health officials in other areas need to
obtain similar data to assess the potential public health utility
of such programs in their local areas.

In counseling pregnant women, screening programs also
raise the need for appropriate counseling that focuses on
issues from a culturally specific approach. Counseling must
include options for limiting disease transmission to offspring.
One option, abortion, is already a volatile issue that will only
further increase the difficulty of dealing with HIV infection in
the population of pregnant women. The option of avoiding
future pregnancies also has cultural implications for the
women being counseled, and these need to be taken into
consideration. There are many difficult questions; there are
no easy answers.

In addition, despite primary and secondary prevention
efforts in the populations most likely to be hardest hit by
pediatric AIDS, we must also be prepared to deal with the
increasing pediatric AIDS caseload. Unanswered questions
include the following: “How do we diagnose HIV infection in
the newborn?” “How do we provide compassionate and appro-
priate care for the increasing number of HIV-infected infants
who will be left to foster homes and other sources of public
care?” And, “How do we deal with potential HIV transmission
associated with child sexual abuse?” The list goes on.

In April 1987, Surgeon General C. Everett Koop convened a
“Surgeon General's Workshop on Children With HIV Infec-
tion and Their Families.” The workshop provided a unique
opportunity “to summarize the current knowledge about
pediatric AIDS and to make recommendations about future
directions in research, prevention, and amelioration of the
effects of pediatric AIDS.” Recommendations from this work-
shop have been printed and should provide a viable com-
munity framework on which to begin dealing with this painful
problem.’ Now is the time that the clinical and public health
communities must begin to address pediatric AIDS with the
same energy and effort as other aspects of the AIDS epi-
demic. Programs aimed at social change and HIV-related
disease prevention may help limit the number of future
children who will “have to know death.”

Michael T. Osterhoim, PhD, MPH

Kristine L. MacDonald, MD
Minnesota Department af Health
Mi i

1. Landesman S, Minkoff H, Holman S, et al: Serosurvey of human immunode-
ficiency virus infection in parturients: Implications for human immunodeficien-
cy virus testing programs of pregnant women. JAMA 1987-258:2701-2703.

2 Quinn TC, Mann JM, Curran JW, et al: AIDS in Africa: An epidemuologic
paradigm. Science 1996:234:955.,

3. Scatt GB, Fischl MA, Klimas N, et ak Mothers of infants with the acquired
i ; ‘ieee vidence for both symptomatic and asympto-

on Children With HIV Infection
and Thew Famulies, publication DHHS HRS-D-MC 87-1. US Dept of Health
and Human Services, 1987.

6. Recommendations for assisting in the prevention of perinatal transmission of
human T-lymphotrope virus type (Mymphadenopathy-assomated virus and
scquired immunodeficiency syndrome. MMWR 1985;34:721-732.

Editonals 2737

 

 
Testimony of

James A. Wileys PhD

Survey Research Center and
San Francisco Men’s Health Study,
University of California, Berkeley

before the
Presidential Commission on the

Human Immunodeficiency Virus Epidemic

December li, 1987
Washington, D.C.

 

 

 
 

 

J.A. Wiley, 12-11-87

The introduction of the human immunodeficiency virus (HIV) into
heterosexual populations can result in three different patterns
of AIDS incidence among heterosexuals. The first type---a
“passive" epidemic---will result if all or most cases of HIV
infection are acquired from heterosexual contact with members of
known risk groups who were not infected heterosexually but by
other means, e@.g., through homosexual contact, sharing needles,
or tranfussion of contaminated blood products. The second type of
heterosexual epidemic is "mixed". It includes infections brought
about By contacts with established risk groups and a lesser
number of infections caused by heterosexual cantact with persons
who themselves acquired infection heterosexually. The third type
of heterosexual epidemic is "self-sustaining". It would maintain

itself more or less independently of HIV transmissions from

members of known risk groups.

In the United States, at present, the heterosexual epidemic
appears to be 2 passive one. Present evidence, taken at face
value, also appears to support the idea that a self-sustaining

epidemic among heterosexuals is unlikely:

1. The percentage of AIDS cases attributed to heterosexual
contact is very low, i.e.,; only @% of cases among American born
adults and adolescents. Virtually all are attributed to contacts
with members of known risk groups--persons who themselves became
infected from contacts with gay males, or contaminated needles or

blood products.

 

 
 

 

J.A. Wiley, 12-11-87

2. Both the African and U.S. gay epidemics are associated with
rates of sexual contact which exceed the rates for American
heterosexuals. Since epidemic spread is directly related to the
rate of sexual contact, American heterosexuals are regarded as

less vulnerable to this outcome.

3. It appears that the average risk of acquiring HIV infection
from a single heterosexual contact with an infected person is
quite low. For example, my Berkeley colleagues and I have
ascertained (based on interviews with and serological testing of
blood from the female partners of HIV-seropositive men) that the
average risk of male-to-female transmission is on the order of 1
per 1,000 contacts. The risk for female-to-male transmission is

probably lower.

4. Large-scale screening of military recruits, blood donors, and
applicants for marriage licenses indicates a fairly low
prevalence of HIV infection in these groups. From this it is
inferred that the rate of infection in the general population,

excluding high risk groups, is correspondingly low.

Does this evidence eliminate the possibility of a self-sustaining
epidemic or allow us to conclude that the probability of such an
occurrence is very low? In my apinion, because we do not yet
know enough about the disease or the factors that affect the rate

of its spread, the answer is no.

 

 
 

 

J.A. Wiley 12-11-87
1. While it is true that the current number of AIDS cases
attributed to heterosexual contact is low, we cannot conciude
that the number of persons infected through this route is
also law. Attempts to extrapolate from AIDS cases to infections
depend on making strong assumptions about two matters that are
presently unknown: the number of infections that will eventually
result in AIDS and the length of time from infection to
diagnosis. If the average time from infection to diagnosis is
quite long—--and we are beginning think that it is---then the
ratio of infections to cases can be very high, especially in the
early stages af an epidemic. This can mask the initial phases of

a self-sustaining epidemic.

e. Although research fas established that the African
heterosexual and U.S. gay male epidemics were fueled by high
rates of sexual contact, it does not follow that the lower rates
of sexual contact of American heterosexuals fall below the
threshold for a self-sustaining epidemic. Given differences in
rates of sexual contact and practices, it daes not appear likely
that American heterosexuals will experience an epidemic on the
scale of those experienced by African heterosexuals or American
Gay men. However, we do not know enough to rule out the
possibility of a smaller, but nevertheless devastating growth of
HIV infection (e.g. 2-5% prevalence among sexually active

heterosexuals) within a decade.

 

 
 

J.A. Wiley 12-11-87

3. The evidence suggests that the average risk of acquiring HIV
infection from a single heterosexual contact with an infected
person is quite low. However, a low average risk may still be
consistent with growth of HIV infection in the population. This
can happen in several ways. First, rates of sexual contact may
be high enough to produce high rates of new infections inspite of
low efficiency of transmission. Second, an average conceals wide
variability in risk. Some individuals are more infectious than
others; some are more susceptible than others. For example, as
much as ae ten-fold increase over the average risk can result
under any of the following circumstances: when the infected
partner is in the later stages of disease progression; when
either partner has genital lesions; or when sexual contact
includes anal intercourse. In making predictions about the
potential for epidemic spread, knowing about average risk may be
less important than knowledge about the number of highly
infectious individuals (the so-called super-spreaders) and their

sexual practices.

4. Although large-scale screening of special groups (e.g.;
military: recruits, blood donors, etc.) has resulted in fairly low
estimates of HIV prevalence, we cannot conclude that the rate of
infection in the general population is correspondingly low.
Because members of special groups are not representatives and may
even be self-selected in relation to their serostatus, their

prevalence rates do not reliably indicate the level of HIV

an

 

 
 

 

J.A. Wiley i2@-11-87

infection in the general population.

In my remarks before the committee, I attempted to convey the
uncertain state of knowledge about AIDS infection among
heterosexuals. It should be clear that the currently available
data do not permit us to predict the likely course of AIDS in
this population. At this time, we cannot even rule out the

terrible prospect of a self-sustaining heterosexual epidemic.

 

 
 

National Minority AIDS Council

Mencer D. Edwards
Executive Director

TESTIMONY BEFORE PRESIDENTIAL COMMISSION ON THE HIV EPIDEMIC

DECEMBER 11, 1987

HUBERT H. HUMPRHREY BUILDING

WASHINGTON, DC

MENCER D. EDWARDS, RN, MSN, MPH
EXECUTIVE DIRECTOR

NATIONAL MINORITY AIDS COUNCIL

 

Office Address: 1012 14th St. NW, Suite 601, Washington, DC 20005
Mailing Address: P.O. Box 28574, Washington, DC 20038

 

 
 

 

Good afternoon . Mr. Chairman and Commission members, my name is

Don Edwards. I am the Executive Director of the National Minority

AIDS Council.

The National Minority AIDS Council is an organization
dedicated to working with and through national minority
leadership organizations, minority businesses, local minority
community-based AIDS projects,and national and local health
agencies to assist in preventing the spread of hiv, and helping

communities to better care for minority persons already infected.

The council has four key objectives:

1. Advocacy
2. Education of national and local policy makers
3. Promotion of minority community-based AIDS projects

4. Application of minority economic resources to AIDS

I am personally pleased to offer testimony as part of these
hearings which have been looking at data collection and related
issues. Persons perceived as “advocates” are often excluded when
the focus is on the more “technical” aspects of scientific
phenomena. You know me for the advocacy I practice on behalf of
people of color regarding the absolute necessity for culturally
sensitive AIDS education, and the speciai kinds of direct
services needed in Black, Latino, Asian, and Native American

communities.

 

 
 

 

But I hope after today you will also know me aS a novice
epidemiologist, now lapsed, who was the. first person ‘in the U.S.
to use the CDC's Public Access Data Set to describe the |
epidemiology of AIDS among Black persons born in the US. Thus I
want to assure you that, like my distinguished co-panelists, I am

quite ready to discuss “the numbers” as regards the HIV epidemic.

In preparation for this testimony, I spoke with a number of
individuals in the Black community. I want you to know that in
the past sixty days, a quantum leap forward on the part of
leading minority clinical researchers and health practitioners
has taken place. Last week a National Institute on Drug Abuse
sponsored a technical review meeting hosted by your fellow
Commissioner Dr. Benny Prim. He and his colleagues at Addiction
Research and Treatment Corporation in Brooklyn brought together
minority scientists, health providers, and community activists
from across the nation to discuss issues relevant to the subjects
on which you are now taking testimony. In two prior meetings, one
sponsored by the Detroit chapter of the National Medical
Association, and again by the Provident Medical Society, the
Brooklyn NMA affiliate, critical attention was again devoted to

these issues among others.

My testimony is offerred as preface to what may become a rising
tide of interest -~ and with that interest, disturbing questions.
I want to share some of these key questions which keep being

raised in the Black community with you now.

 

 
 

 

i. Is it as bad as they say it is for Blacks?

No matter who says it, or what their motives are, yes it is as
bad as they say it is. As of November 23, 1987 47,022 cases of
AIDS have been reported to the Centers for Disease Control's
AIDS's Surveillance Program. Of this number 11,746 (25%) were
Black. This number represents 11,373 (25%) adults and adolescents
over the age of thirteen. The rest are 373 (55%) cases in

children under age 13.

Overall, the cummulative incidence or cases per million
population for all Black adults is 3.1 times that for white adult
cases. The cummulative incidence for Black women is 13.3 times
that for white women. Twenty-five percent of the total number of

Black cases are between the ages of 13-19 years.

As much as sixty percent of Black adult cases can be found in
three states: New York, New Jersey, and Florida. However, the
incidence of Black cases is climbing in each of the top 25 U.S.
cities with the highest prevalence of AIDS cases. In the Black
community, the previous East coast focus of the epidemic is
quickly disappearing as more cases among Blacks are reported in
Cities like St. Louis, Kansas City, and Denver. Increases have
also been noted in Los Angeles and San Francisco, where Blacks

make up small percentages of the overall metropolitan population.

Adult Black cases are expressed in significant percentages in

 

 
 

 

every category except hemoplilia/coagulation disorder which has
less than one percent of the cases. As of November 23, in adults
40% of cases are homosexual/bisexual; 35%, IV drug user; 7%
homosexual/bisexual male/IVDU; heterosexual cases, 11%; 1%,
transfusion related; and 5% in the incomplete information

category.

In the pediatric population, 2% of the cases are in the
hemoplilia/coagulation category; 87%, parent with/or at risk of

AIDS; 6%, transfusion related; and 5%, incomplete information.

The vast majority of cases in the adult and pediatric groups
report Pneumoncystis carinii pneumonia and/or other

opportunistic infections as the key diagnostic features.

The AIDS Surveillance Program of the Centers for Disease Control
is to be commended for the efforts they nave taxen to ensure that
the nation has a regular up-date of this type of data. Yet, I an
sure they know that they have not met the needs of everyone.

The question is should they be expected to? If not them, who?

I bring the following feedback regarding surveillance. First, in
addition to the national surveillance, it may be time to
institute systematic, standardized surveillance and reporting of
cases in the top 25 high prevalence cities. These cities,
coincidentally, have significant Black populations. The point is

to describe not one national epidemic, but to examine whether in

 

 
 

 

fact there are a number of regional or local epidemics occurring

which look very different from one another.

Additionally, considerable dissatisfaction has been expressed
regarding the lack of standardization and quality of surveillance
in states and localities. For instance 40% of the cases in
Alabama are Black, but hardly anyone in or out of Alabama knows

this. Whose responsibility is this -- the state’s or the CDC's?

Black activists are nearly unanimous in their desire to have
national and state geographic data by race, data on gender by
race, and especially AIDS cases by risk factor combinations by
race. They would also like to see homosexual cases separated from

bisexual cases for Blacks.

This aggregation of data might be done nonthly or bi-monthly, but
it is crucially needed to confirm or deny anectdotal community
information. If this responsibilty falls to the states, the CDC
should provide the standard for the collection and reporting of
this information. There are legitimate issues regarding breach of
confidentiality raised whenever data is cut too fine, but,

safeguards can be put in place.

Surveillance should reflect a greater sensitivity to

race/ethnicity as a highly significant variable.

 

 

 
 

 

2. Did AIDS become a health problem for Blacks only recently?
If yes, how did it happen so quickly? If no, why didn't we know

about it sooner?

As more information is diseminated, it will become clear that
scientific awareness of the disproportionate impact of the HIV on
the Blacks and other minorities was evident long before the
publication of the report titled "Acquired Immunodeficiency
Syndrome (AIDS) among Blacks and Hispanics-United States"
published in the Center for Disease Control's Morbidity and

Mortality Monthly Report of October 24, 1986.

There is a concern among many Black AIDS activists that as more
Black persons learn the about the early evidence of the
disproportionate impact dating back to at least 1983, there may
develop a “credibilty gap” as to whether any other data is being
held back. The suspicion and anger engendered may detract scarce
resources from education and direct services at a point where

progress is being made.

The answers to the questions above involve more than mere
curiosity. At their heart is the question of credibility--of the
current message about the impact of AIDS on blacks, and of the

messenger.

The fact is that there was evidence available which had it been

subjected to more informed interpretation could have hastened the

 

 
 

 

1986 report. In the MMWR dated August 28, 1981 (p. 409.)
demographic data provided for the first 102 cases by disease,
race, and sexual preference showed that 73% were white, 11.2%
Black, and 10.3% Hispanic. Had characteristics other than sexual
orientation been seriously considered, it might have been noted
that cases among Blacks already equaled their percentage in the
US population. If compared to the percentage of Blacks in each of
the two west coast cities from which most of the cases were drawn

an even more significant association might have been drawn.

I want to draw your attention to a statement which was made in
that report. It said, “the majority of the reported cases of
Kaposi's Sarcoma and/or Pneumocystis carinii pneumonia have
eccurred in white men." Examined in hindsight this is a chilling
statement because even after the numbers began to change nore
dramatically, the perception remained the same. It became a
confirmation of the classic warning given to all new researchers

to avoid looking for something lest that be all you find.

2. Why does it seem that all the bad news on AIDS is about Black
people? Are all these studies just another way to blame Blacks
for AIDS?

Numbers paint pictures. For Blacks, the picture has been
consistently suspect, and in the case of the Haitian community,
particularly ugly. First, in the minds of Black persons, the

Haitians were “blamed” for bringing AIDS to the US. Second,

 

 

 
 

 

initial reports on AIDS in Black Africa seemed more interested in
pointing out how the governments of these countries were standing
in the way of scientific progress, of help for their citizens,
and of an explanation of where AIDS started. Lastly, in the U.S.
data was released which fueled perceptions that it was black drug
addicts who were spreading AIDS across the country. Because of
them, AIDS is going to infect heterosexuals, and kill hundreds of

unsuspecting women and innocent babies.

To the Black person on the street, the perceptions listed above
confirmed what many of them already believed as rote: In America,
if there is no explanation to be found, blame it on Blacks. AIDS
education is now hampered by ali the poison that has been
injected into the community atmosphere by what was at its most
essential a stereotypical, insensitve, and uninformed processing
of experience and of data. I hope that the fundamental importance
of sensitivity to culture and tradition is now accepted and will
no longer contribute to the erection of tedious and divisive

barriers.

Researchers ‘need to keep a number of points in mind during both
the conceptualization and the conduct of any research in the US
Black community. How many times do these point have to be made?
First, the U.S. Black Community is not a monolith. In the U.S.,
Black identity includes early and recent immigrants from |
Carribean nations who may speak native languages, as well as the -

languages of their former colonizers -- English, French, Spanish,

 

 
 

 

Dutch -- and recent immigrants from African nations who despite
being from the same continent, are tremendously separated by
language, geography, and national experience. The U.S. Black
community consists of separate ethnic communities joined into a
multi-ethnic community by African ancestry, common cultural
values, cross-cultural marriage, and a common experince of racism

and racial pride.

Second, there is a psychology at work in the Black community that
serves many purposes. A key function of this psychology is
self-preservation. It appears that it may be contributing to the
denial already aggravated by stigmatization that has produced a
"SEE NO EVIL-HEAR NO EVIL-SAY NO EVIL" process which resists
education related to AIDS. Disrespect of this psychological
process only serves to reaffirm its need to exist. It is
particularly ironic that in the case of AIDS, this psychology
will not lead to self-preservation, but to death. But it will not
be a self-imposed death sentence. A large part of the
responsibilty will rest with those who ignore the social context

and consequences of research and data collectiion.

3. Is the impact on Blacks really as bad as they say? Is it

really so disproportionate?

There is disagreement on the answer to this question. From a
scientific standpoint, a number of Black researchers have pointed

out that using general population figures for denominator data

 
 

 

drives the disproportion up. Epidemiologically, it would make
more sense to use either the percentage of Blacks in the urban
center, or in the metropolitan SMSA as the denominator. On the
East Coast in paticular, given the high percentage of Blacks in
these cities, the ratio would probably begin to approach 1:1.
Others hold to the 2:1 ratio as correct considering the projected
spread of the epidemic into the heartland of the U.s., and the

higher degree of undercount that is suspected for Black cases.

The interesting point is that there seemed to be less concern
about the accuracy of this message than what would be the
benefits accruing from either one. If the higher ratio engages
the attention span of Blacks, and drives increased funding, it
should be maintained. However, past experience with other high
prevalence health problems affecting Black Americans has not led
to increased allocation of resources as the classic formula
dictates. Therefore, some activists suggest that the emphasis on
the 2:1 incidence rate unneccessarily increase stress and stigma.
They argue that Blacks will “open up" to AIDS if these two

elements are decreased or eliminated.

4. Is there a genetic link which explains the high incidence of

HIV infection among Blacks?

It is particularly important that this line of investigation be
recognized for the red flag it waves in the Black community.

Rightly or wrongly, anytime "genetics" is mentioned, it raises

 

 
 

 

the spectre of ever-recurring racist attempts to explain the
lives of black people with an up-date on archaic pseudo-
scientific social-biological theory. The research on the high
incidence of HTLV-1 in some parts of Africa which are contiguous
with the presence of HIV, as well as in a number af carribean
nations should be given appropriate attention. But the Black
community will reject anything remotely suggesting that something

inate to being a Black person increases one's risk for

infection.

To the Black person on the street, much of the talk about DNA and
RNA, as well as that about mothers passing the virus to their
children translates to "genetics". Researchers should also
understand that considerable concern exists regarding whether
AIDS is a "genocidal plot of the racist U.S. government to

destroy Black peoples world-wide".

The existence of Black people is not accepted by many Blacks
themselves as a given. Thus is should not come as a surprise that
the high number of AIDS cases among Black women and babies raises
the suspicion quotient of Blacks rather easily. They are also
quite mindful that the “Government” was responsible for the
“Tuskegee Experiment." All these fears can be subsumed under the
category -- Genes, Genetics, and Genocide. Resaarchers should

be mindful of the power these perceptions hold.

 

 
 

 

5. With whites, it seems only gay men get AIDS. With Blacks,

men, women, and children get AIDS. Why?

Though the epidemic is six years old, it was only thirteen
months ago that the Centers for Disease Control published a
report in the Morbidity and Mortality Weeekly Report titled
"Acquired Immunodeficiency Syndrome (AIDS) Among Blacks and
Hispanics-United States". Thus it has been difficult to accept
the data though it delivered a clear and unmistakeable message:

AIDS is not a gay white man's disease.

But for the most part, where it appears that so much has been
determined about the sexual practices, the degree of behavioral
change, percentage by population, etc. for gay men, Blacks have
been mostly left with a lot of interesting, but unanswered

questions.

Where are the longitudinal studies exclusively of Blacks or at
least oversampled for Blacks? Where are the case-control studies
linking cases and controls to ascertain independent or multipie
variables? For the category homosexual/bisexual male, where do
most black men fall? For homosexual/IV drug user, which
contributes the greatest to risk? If there is one area in which
there is almost complete unanymity, it is that the degree of
scientific investigation absolutely contradicts and renders

impotent the declaration that a crisis exists.

 

 
 

 

In addition to the potential "genetic" factor, at least two other
factors have been suggested as possible avenues of exploration.
One is that the racial distribution of AIDS cases reflects the
racial distribution of populations in the high prevalence areas.
A second factor suggested is economic and cultural factors that

may also account with observed differences in incidence.c

Both of these lines of investigation lie dormant. Perhaps in one
of them lies key information which will help save countless lives
in the Black community. The quesion being asked in the community
however, is "Do they want to know?" After they know, will they
spend the soney? If they start with AIDS, will they really use
what they already know about teenage pregnancy, drug use, and
sexually transmitted diseases in Blacks to do something to stop

that?

Commissioners, these are a mere sampling of the questions which I
was asked by key individuals in the Black community to raise with
you this morning. I would like to acknowledge Dr. Lawrence
Sanders, MD, MPH of the Philadelphia Health Department, Dr.
Helene Gayles, MD, MPH of the AIDS Surveillance Program at the
Centers for Disease Control, and Dr. Wayne Greaves, MD, MPH,
Chief of Infectious Diseases at Howard University Hospital

who was to my knowledge the first Black physician and
epidemiologist to ask the hard questions and demand answers about

the impact of AIDS in the Black community.

 

 
 

 

10 Dattas Gay News / Nov. 25, 1983

AIDS Update: No Longer Gay White Mans Disease

By Dion B. Sanders
Via Gay Prese Association Wire Service
SAN FRANCISCO—Shattering a
widely-held belief that AIDS is a “white
gay man's disease, edd a
previoualy-unpublicized data compiled b
the national Center for Diseage Contral in
Atlanta show that more than a third of ull
AIDS patients in the Uniled States gre
om racial and ethnic minorities.

~ Moreover, a doctor's publiched in
a leading medical journal says that while
a significant percentage of black and
Hispanic AIDS patienta—not counting
Haitiane—are intravenous drug users, 8n
equally-significant percentage of them are
‘upfront gays.

In a telephone interview from Atlanta,
Dr. Richard Selik, director of AIDS
information at the CDC, separted that as
of Oct. 19, out of a total of 2513 AIDS cases
nationwide, only 57.9 percent are white.

Blacks (including Haitians), Hispanica,

ians and Native Americans make up @

ombined 40.9 percent, Selik said, with the

emaining 2.2 percent of undetermined
ethnic origin. = - ~

For months, news medie reports have
tepeatedly stated that AIDS patients are

rimarily gay men, IV drug ueers,

aitians and hemophiliaca receiving
blood tranefusions. Gay men account for
71 percent of all AIDS cases nationwide.
_ The belief that AIDS is a “while gay
men's disease” atems from a long-held
perception of the gay community by the
general public—eapecially by minority
communitics—ee being exclusively white,
Aeapite the emergence of gay and lesbian

people of color into the publiceyeinrecent |

months. ’ ‘

(io San Francisco, thie perception has
led to charges by some minority
community leaders thet public. funds

being appropriated for AIDS are hein
taken away from existing health
programe forthe poor-—a diaproportionate
percentage of whom are binck and

ispanic, sccording to lncal presse
reports.) . .

As of Oct. 19, there have been 1048
deaths, for a mortality sate of 41.7 percent

abs Hes Keak vant

AIDS CASES IN THE U.S. .
BY RACIAL/ETHNIC ORIGIN

AS OF OCTOBER 19, 1983

ethala number peig
group of cases = of cases
White ° 1456 . 57.9%
Black (Non-Haillan) 648 -21.1%
Haitian 117 4.7%
Latinos 354 14.1%
Asians a. .%

Native American k

Other Ethnic Origin 47 °°2.2%
TOTAL 2513 100.0%

“Less than 1 percent.
Sousce: Center ke Disesse Contsol, Atlanta. Ge.

seported that blacks make up 21.1 percent,
“with the figife teing to 25.8 percent

when Haitiané are included.

*" “Hispanics make up 14.1 percent, with

Asiane and Native Americana combined
comprising approximately 1. percent,”
Selik said. - .

“The remaining 3.2 percent of AIDS
patients sre of undetermined ethnic
origin,” he reported.

. The proportion of blacka and Hie anica
among people with AIDS Is atrik ng in
{hat it is neatly double the proportion of

blacke and Hispanics In the U.S.

population as a whole.

Even more striking ie the fact that aso
Oct. 19, there have been only eight cise of
AIDS reported nationally among Asians
and only three among Native Americans.

No nationwide city-by-city - ethnic

Ba BANG AIO

Greater Bay Area as of Oct. 18, 90.3
percent are white, 4.9 percent are black, 3.8
percent are Iliapanic, and a fraction of 1
percent are Asian or Native American.
In sharp contrast, 50 percent of AIDS
cages reposted ip Philadelphia acearly as
lact July are black, according to that city’s
health department, prompting the
creation of an AIDS education program
simed specifically at black gays in that

city.

Philadelphia, Atlanta and Washington,
D.C. all have sizeable black populations,
with blacke an oyerwhelming majority in
the nation’s capital. —

Aaked what percentage of the non-white
AIDS cases are gay and what percentage
are IV drug users, Selik quoted figures
from an article by Dr. Harold Jaffe
published in the Aug. ‘83 issue of the

Journal of Infectious Diseases (Vol. 148, p.°

339). . .
The article, based on CDC reports of
2000 AIDS cases nationally as of last July,
reported that among non-Haitian blacks
who had AIDS, 17 percent were upfront
gaye, 46 percent were IV drug usere, end
37 percent were of undetermined risk
factors. . ;
(A similar breakdown of Haitian AIDS
patients by the Jaffe article has been
tendered unreliable because of recent
news reports disclosing that many such

Haitians were gay, but were unwilling tc
admit it, because of severe taboos againat
ain exuality in Haitian society, Selik
aid.
~Among Hispanic AIDS patients, the
Jaffe article reported that 11 percent were

upfront gays, eed ved
DENA TA SHR PS eR
Native Americans and Asians with AIN®
because there were no repert.
among them at the time the article wa.
published, Selik ssid. °

Nor did the article contain Inform..-’
on the number of AIDS pus.
women. ae

In the San Francisco area, only (1:
cases of AIDS ainong [V drug « - .
been reported so far, m..

Norman, coordinats
health services for the city.:

Two are white; the third is black.

. Norman noted that whereas 71 percent
ef AIDS es nationally are gay, the
figure rises to 90 percent in:the San
Francieco area, as there are “only a
handful of Haitians,” living here, .

’ Locally, there have been 96 deaths, fora
mortality -rate of epproximately 95
percent.

Norman attributed the disparity of the
national and local mortality rates to the
near-absence of IV and Haitian AID
casea here. .

:‘Selik reported that the percentage
ethnic breakdown ‘has been fairly
constant for more than a year.”-Asked

why the ethnic data had not previously

been published by either the gay or the

mainstream -media up to now, Soelik-
responded,’ *Pethaps they (editors) for

some reason thought t it wasn’t

‘newsworthy.” -

 

 
 

More Than 3,300 Affected

7

A chose look at the victims of AIDS reveals
a disproportionate number of blacks struck
by the deadly disease, while more than half of
the nation’s children with AIDS age black.
< Citing recent data from the Centers for
~~ Disease Control, Dr. Wayne Greaves, chief of
=— the division of infectious diseases at Howard
University Hospital in Washington, D.C.,
notes that he has found “a lot more cases
ge mons blacks than we previously thought.”
Nationally, blacks comprise 25 percent of
"4 all victims of Acquired Immune Deficiency
a Syndrome — more than double the 12 percent
black population in the nation. And some 56
SD percent of the nation’s children hit by AIDS
a=: are black, Greaves emphasizes in an interview.
S Since the AIDS virus was first seported in
this countsy in 1981, more than 13,400
Americans have been afflicted by the disease,
jy» which destroys the body's immune system
M and is ultimately fatal. Blacks account for

3/85

§ more than 3.300 of those victims, says Greaves,

E formerly with the CDC in Atlanta, indicating

© that half of all Americans contracting the
disease have died.

 

i+

Of the known “pediatric cases” - compris-
© ing childsen under age 13 ~ 107 of 191 child-
ten with AIDS have been bieck, he adds,
noting that children make up about ! percent
of all AIDS cases reported in the United
States.

The epidemiologist points to the state of
Maryland as a good example of the dispro-
portionately high incidence among blacks. Jn
that state, 116 of the 222 known cases have
been black victims, accounting for 52 percent
- more than douhle the 22 percent black
populstion in Maryland, he stresses.

=

a

~

Like the disease itself, mystery shrouds the
reasons why there is a high AIDS tncktence
among blacks.

“There's still belief (in the black commun-
ity) that AIDS is a white man’s disease,” says
Greaves, who is also an assistant professor of
medicine at Howard. “AIDS is color blind.”

Another problem is tha dlacks often tend
to seek health care when a medical problem
has reached an advanced stage of develop-
ment. At Howard University Hospital, where
most AIDS victims treated thus far have been
black, AIDS sufferers have generally lived
about six to eight months after diagnosis of
the disease. The median survival nationally is
18 months from diagnosis to death.

While Greaves js on what he calls “the
front line™ of treating AIDS patients, his col-
league, Dr. Winston Frederick, assistant pro-
fessor of medicine at Howard, is trying to
unlock the secrets of AIDS. The former
National fastitutes of Health sesearcher is
conducting research to find certain perame-
ters or “seliable markers” that may give clues
to the disease’s development. ~.

So far, AIDS has primarily struck homo-
sexual and bisexual men, intravenous (IV)
drug abusers and recipients of contaminated
blood. But in these high-risk groups, a stark
difference emerges between white and black
AIDS victims.

According to Greaves, about 85 percent of
the nation’s whites with the disease are gay
and 8 percent ase intravenous drug abusers.
Among black AIDS victims, 49 percent are
gay and 38 percent are [V drug abusers.

He believes that more altention needs to be
given to drug abusers,

 

AIDS Strikes Blacks At High Rate

 

   
  
  
 

tard geveat --tewes me Oy 8th - 6 0) fy

Teens in: oo Race LES are = Soest os
ee ya: at tea s ain ?* . .

bai 2 hand a 4 hee bares piieiae . “ . ‘i.
ie = iy Sine ae Free si “inthe: i ee "

4 "eg a-€ . Be wrt vey Oey

gee ARP | ae
4e gets

  

 

”

stg er, ae
. fen at coli ke

ae
“y

 

rigeilh 1 Sl

MRT: ot aM

  

  

 

 

Dr. Wayne Greaves, chief of "nfectioes diseases at Howard University Hospital, ‘consults
‘with Dr. Soon-Young Park for the university's College of Pharmacy and Pharmacal
Sciences on drug treatment of AIDS patients, most of whom are black at the Washington,
D.C, hospital. Greaves emphasizes that nationally blacks comprise 25 percent ofall AIDS
‘ victims—more than double the 12 percent black population in the nation.

 

 

 
 

 

 

 

 

 

THE IMPACT ON
HISPANICS IN
SELECTED STATES

A COSSMHO BACKGROUND PAPER

DRAFT--NOVEMBER '87

COSSMHO
The National Coalition of Hispanic
Health and Human Services Organizations

1030 15th Street, NW
Suite 1053
Washington, DC 20005
(202) 371-2100

 

 

 

 

 

 

 
 

 

 

 

 

 

 

 

 

 

Since the identification of AIDS, the disease
has had a disproportionate effect in the
Hispanic community, claiming the lives of
infants, denying families their care providers
and prematurely ending the lives of
thousands of Hispanic men and women in
the United States.

The data surrounding AIDS in the Hispanic
community is a sobering litany. (Data
according to November 16, 1987 Centers for
Disease Control reports)’

* While Hispanics represent 7.9 percent of
the population, according to 1987 Census
data, they represent 14 percent of all AIDS
cases—a proportion that has remained fairiy
constant in recent years.

+ Hispanics account for 20 percent of AIDS
cases among women=a rate that is almost
three times that which would be

figures translate into a cumulative incidence
of AIDS among Hispanics that is almost
three times that of non-Hispanic whites.
(Note: see Figure 1--doted area represents
cumulative incidence of AIDS for each 1
million population segment of the relevant
racial or ethnic group.)

Given the stability of Hispanic representation
among AIDS cases and the projection by the
Public Health Service that new AIDS cases
will continue to come from the currently
recognized transmission categories, it is
Teasonabie to project that Hispanics will
continue to represent a_ disproportionate
share of AIDS cases in the United States.

‘Furthermore, Hispanics are overrepresented

in transmission categories projected by the
Public Health Service to have the most
significant growth over the next four years.

 

expected based on the
proportion of Hispanic women
in the U.S. female population.

* Hispanics account for 23
percent of pediatric AIDS
cases—a rate that is more than
twice that which would be
expected based on the
Proportion of Hispanics in the
U.S. population of children
under the age of thirteen.

* Data presented by CDC]

researchers, —§ Selleck and
Rogers, at the 1987
International Conference on
AIDS, indicated that the

cumulative incidence of AIDS
‘for Hispanics is 290, compared
to 283 for Blacks and 98 for non-
Hispanic whites (per 1 million
population segment). Such

 

Ce Se Sree cn he ae ke i ae il hen

ne eed
ee er eee)

    

HISPANIC

 

  

WHITE (98)

Cumulative Incidence of AIDS by racial/ethnic
group per 1 million population segment

FIGURE 1

 

1,000,000

 

 

HISPANIC-AIDS--STATE DATA

 

 

 
 

 

 

 

 

 

 

 

 

 

 

The most significant transmission categories
(see Figure 2) for Hispanic AIDS cases are:

° Homosexual/Bisexual male which
accounts for 49 percent of cases,

¢ intravenous (IV) Drug Abuser which
accounts for 35 percent of cases,

¢ Homosexual Male and IV Drug Abuser
which accounts for 7 percent of cases, and

* Heterosexual Cases which accounts for 4
percent of AIDS cases among Hispanics in
the United States.

The representation of Hispanics in the
heterosexual transmission categories of "IV
Drug Abuser” and “Heterosexual Cases” is
Significant as Hispanics are 3.5 times as
likely to be represented in these categories
than are non-Hispanic whites. Hispanics are
seven times as likely as non-Hispanic whites
to fall under the transmission category of “IV

Drug Abuser":The Public Health Service has
projected that new cases acquired through
heterosexual contact will increase almost
seven-fold by 1991.

concern is the
Hispanics among

Aiso of great

overrepresentation of
pediatric AIDS cases. The majority of
pediatric AIDS cases fall under the
transmission category of "Parent with/at risk
of AIDS” with other transmission categories
being far less ‘significant. Hispanic pediatric
AIDS cases under this transmission category
are 81 percent compared to 42 percent for
non-Hispanic whites. Only 11 percent of
‘Hispanic pediatric AIDS cases fall under the
transmission category of "Transfusion,
Blood/Components” compared to 32 percent
of non-Hispanic white pediatric AIDS cases.
Given the development of blood screening

 

 

FIGURE 2

 

Percentage of Incidence of various Transmission Categories in Hispanic AIDS cases

 

WB HomosexualBisexual
Male

E32 intravenous (IV) Drug
Abuser

 

Es Homosexual Male and IV
Drug Abuser

a Hempohilia/Coaguiation
Disorder

49%

CO Heterosexual Cases

SJ Transfusion,
Bilood/Components

HON Undetermined

 

 

 

 

 

PIS) ob Voce Pai eee

 

 

 

 

 
 

 

 

 

 

 

 

 

 

 

procedures, ‘this category is not expected to
grow.. Hispanic children currently account
_for approximately one out of every four
pediatric AIDS cases. New pediatric AIDS
Cases are projected by the PHS to increase
almost ten-fold by 1991.

There is significant variation by state,
however, in the percentage of AIDS cases
represented by Hispanics and in the
transmission categories of those cases (see
Figures 3 and 4). The following sampling of
States include: Florida, Illinois, New Jersey,
New York, and Texas as well as Puerto
Rico. Other states were Surveyed by
COSSMHO when it became difficult to get
appropnate information from the Centers for
Disease Control, especially surrounding
State breakdowns of Hispanic AIDS by
transmission categories. There was also
difficulty, however, in getting this information
from the states. Many states cited issues of
confidentiality, but in states, such as
California, where this is not an issue

 

|
- |
because of a large enough sample of
Hispanic AIDS cases there was still not a
response. In no state was there an ongoing
effort to collect data on Hispanic AIDS in
terms of transmission categories and case-
fatality rates. All data received was the result
of special studies by COSSMHO contacts in
the various state AIDS epidemiology
programs. It is disturbing that this
information is not part of an ongoing effort by
COC or any states, but is a result of the
dedication of a handful of state government
employees. Studies of Hispanic AIDS cases
were done by states on an individual basis
and not as part of surveillance, therfore, the
following information is for different dates.

‘The Puerto Rico surveillance report is only

updated every two years and reflects the
latest available information.

 

80 r 77
70 +
PERCENT ¢0 «
Homosexual/
Bisexual 50 +
Male
transmission 40 +
category
comprises of 30 +
Hispanic

AlOS cases 20 +

10 §

 

0 «

4 Florida lilinors

 

FIGURE 3

     

New Jersey

Percentage of incidence of “Homosexual/Bisexual Male” as a Transmission Category in
Hispanic AIDS cases for selected states

New York
STATE

Texas Puerto Rico

 

 

HISPANIC AIDS--STATE DATA

 

 

 

 
 

 

 

 

 

 

 

 

 

 

 

FLORIDA

As of September 4, 1987, Hispanics
represented 13 percent of AIDS cases in
Florida. Of these 354 cases:

¢ 77% were Homosexual/Bisexuai Males,

¢ 8% were IV Drug Abusers,

* 6% were Homosexual Males and IV Drug
Abusers, and

¢ 1% were Heterosexual Cases.

As of the same date, there were 2 Hispanic
pediatric AIDS cases forbiding any further
breakdown of transmission category for

reasons of confidentiality.
ILLINOIS
As of August 31, 1987, Hispanics

represented 8 percent of AIDS cases in
Illinois. Of these 82 cases:

¢ 57.0% were Homosexual/Bisexual Males,

* 6.1% were IV Drug Abusers,

* 3.7% were Homosexual Males and IV
Drug Abusers, and

¢ 2.0% were Heterosexual Cases.

As of the same date, there were 2 Hispanic
pediatric AIDS cases forbiding any further
breakdown of transmission category for

reasons of confidentiality.
NEW JERSEY
As of November 1, 1987, Hispanics

represented 13 percent of AIDS cases in
New Jersey. Of these 349 cases:

° 35% were Homosexual/Bisexual Males,

« 46% were IV Drug Abusers,

° 5% were Homosexual Males and IV Drug
Abusers, and

¢ 8% were Heterosexual Cases.

As of the same date, there were 18 Hispanic
pediatric AIDS cases, 89% were in the
transmission category of Parent with/at risk
of AIDS.

 

60 ¢

PERCENT
IV Drug
Abuser

transmission
category
comprises of

Hispanic

AIDS cases

 

Flonda IIinors

 

FIGURE 4

 

New Jersey

Percentage of incidence of "IV Drug Abuser” as a Transmission Category in Hispanic
AIDS cases for selected states

51.9

Puerto Rico

New York
STATE

Texas

 

 

     

HISPANIC AIDS--STATE DATA

 

 
 

 

 

 

 

 

 

 

 

 

NEW YORK

The largest number of Hispanic AIDS cases
are in New York State. As of October 1,
1987, Hispanics represented 24.4 percent of
AIDS cases in New York State. Of these
2,937 cases:

¢ 33.6% were Homosexual/Bisexual Males,

* 51.9% were IV Drug Abusers,

¢ 5.7% were Homosexual Males and IV
Drug Abusers, and

« 4.4% were Heterosexual Cases.

As of the same date, Hispanics represented

34 percent of pediatric AIDS cases in New

York State. Of these 72 cases, 91.6% were

in the transmission category of Parent with/at
risk of AIDS.

TEXAS

As of October 1, 1987, Hispanics
represented 10 percent of AIDS cases in
Texas. Of these 294 cases:

° 77% were Homosexual/Bisexual Males,

* 4% were IV Drug Abusers,

* 10% were Homosexual Males and IV Drug
Abusers, and

°* 1% were Heterosexual Cases. -

As of the same date, Hispanics represented
26 percent of all pediatric AIDS cases in the
State. Of these 5 cases, 40% were in. the
transmission category of Parent with/at risk
of AIDS, 40% were in the transmission
category of Hemophilia/Coagulation
Disorder, and 20% were in the transmission
category of Transfusion Blood/Components.

_ PUERTO RICO

As of June 15, 1987, there were 456 cases
of AIDS in Puerto Rico. Of these cases:

~ 37% were Homosexual/Bisexual Males,

* 40% were IV Drug Abusers,

* 13% were Homosexual Males and IV Drug
Abusers, and

Tay e8 eee VE OYE.

      

‘first

_ commitment,

5% were Heterosexual Cases. — |
As of the same date, there were 22 pediatric
AIDS cases in Puerto Rico. Of these cases,
82% were in the transmission category of
Parent wittvat risk of AIDS.

The variation among geographic areas in
the significance of IV drug use as a
transmission category may be due to the
time the HIV virus was introduced in the IV
drug user community. Sera prevalence
Studies conducted in New York City indicate
that the HIV virus was introduced in the IV
drug user community in the late 1970's. All
sera collected between 1960 and 1976 in
this study wera HIV seronegative and the
seropositive serum sample was
collected in 1978. After this first positive
sample, seroprevalence rose rapidly to over
40% of the 1980 samples.3 tt is possible
that the HIV virus was introduced in some
States at a later date, e.g., Texas. Thus, at
this time [V drug use is not a significant
transmission category in some states.

Disease and death resulting from HIV
infection are likely to increase not only
through the next four years but into the next
century.4 Despite the tragic and devastating
effects of AIDS in the Hispanic community,
the prospect for turning back the tide is not
good. While Black and Hispanics account
for 40 percent of all AIDS cases, 71 percent
of all women with AIDS, and 77 percent of all
pediatric AIDS cases; there has been little
money directed to national and local
community-based Black and _ Hispanic
groups for the mounting of community and
Ccuturally-appropriate AIDS education,
prevention, and treatment programs. Such a
along with a _— greater
concentration on collecting Hispanic AIDS
epidemiological information, must be made
at the national level before the Hispanic

community can effectively deal with AIDS in
its communities.

 

 
 

 

 

 

 

 

 

 

 

 

 

1) All data derived from the Centers for Disease Control (CDC),
AIDS Program, Center for Infectious Diseases Aids Weekly
Surveillance Report --United States for November 16, 1987
unless otherwise stated.

2) State data derived from a COSSMHO survey of state AIDS
epidemiology programs.

3) Novic DM, MJ Kreek, DC Des Jarias, et al. Abstract of Clinical
Research Findings: Therapeutic and Historical Aspects.
Problems of Drug Dependence 1985: Proceedings of the 47th
Annual Scientific Meeting, The Committee on Problems of Drug
Dependence, Inc. NIDA Research Monograph 67, ed LJ Harris.
(Washington, DC: USGPO).

4) Institute of Medicine--National Academy of Sciences,
Confronting AIDS: Directions for Public Health, Health Care, and
Research, pg. 9. (Washington, DC: National Academy Press).

HISPANIC AIDS--STATE DATA

 

 

 
 

       

THE USE OF DOUBLING TIME

IN ASSESSING THE COURSE OF THE AIDS EPIDEMIC

Victor De Gruttola and Stephen W. Lagakos

Harvard School of Public Health

INTRODUCTION

The doubling time for the AIDS epidemic in the U.S. has been
observed to increase from 5 to 13 months over the past six years. What
can we infer from this increase? Does it mean, as some have suggested,
that the epidemic may have begun to ‘run its course’, or that behavioral
changes have had a major impact in reducing the incidence rate?

The purpose of this note is to discuss how interpretation of
doubling time depends on the dynamics of the AIDS epidemic. Apart from
changes in behavior, doubling time is influenced by three phenomena: 1)
as the prevalence of HIV infection among individuals at highest risk
increases, the rate of new infections in that population decreases; 2)
the populations at risk for AIDS are highly heterogeneous: some, such
as homosexually-active men practicing high risk behavior with many
partners can be almost entirely infected while others have lower

prevalences and rates of spread of HIV infection; and 3) the risk of

1 Condensed from ‘The Value of Doubling Time in Assessing the Course of the
AIDS Epidemic’.

 

 
 

 

AIDS after infection is not constant but appears to increase with time
since infection.

The first two of these phenomena are characteristic of most
epidemics resulting from introduction of a new infectious agent into a
population, and tends to cause the doubling time in cumulative infection
rate to increase over chronologic time. The third phenomenon also tends
to increase doubling time for most epidemics in which there can be long
delays between infection and disease. Thus, the observed increase in
doubling time for the AIDS epidemic is, at least in part, an inevitable
consequence of the dynamics of epidemics. We will argue from examples
that it is not possible to determine the degree to which behavioral
changes may have contributed to the increase in doubling time, and that
changes in doubling time are not particularly informative about future
spread of the epidemic to lower-risk populations such as heterosexual ly-

active individuals.

INFECTION, LATENCY ANDINCIDENCE DISTRIBUTIONS

Doubling time in incidence is difficult to interpret because it
reflects two features of the epidemic: 1) the rate over time of new HIV
infections; and 2) the risk of contracting AIDS once infected. The
doubling time in AIDS incidence will be constant if the doubling time in
the cumulative number of infections is constant and the risk of
contracting AIDS once ‘infected is constant. Even in simple epidemics in
a homogeneous population, however, constant doubling time in cumulative
number of infections will occur only early in the epidemic when the
number of infectives is negligible relative to the number of

susceptibles. As the epidemic spreads, the reduction in the proportion

 

 
 

 

of the population that remains susceptible causes the doubling time in
the total number of people infected to increase with chronologic time.
The proportion infected does not have to be very large for the doubling
time in AIDS incidence to increase substantially. For example, with
plausible models for rates of infection and risk of AIDS after infection
(estimated from a cohort in San Francisco described below), the doubling
time in cumulative incidence of AIDS increases by more than 50% by the
time the prevalence of infection is 25%.

An additional factor that causes doubling time in cumulative number
of infection to increase is that the populations at risk are
heterogeneous. Heterogeneity refers not only to different ‘risk groups’
such as homosexually-active men and IV drug users, but also to
heterogeneity in behavior within groups. For example, variations in the
rate of new partner acquisition, type and frequency of sexual acts per
partner, and duration of relationship all affect the rate of new
infection. This heterogeneity is believed to be the reason that the
observed rise in infection in cohorts of homosexually-active men is more
nearly linear than exponential.

Even if rate of new infection were exponentially increasing, it
would also be necessary that the risk of contracting AIDS does not
change with time since infection. However, neither of these assumptions
about risk of AIDS after infection are consistent with the available

data.

INCREASING DOUBLING TIME
What type of changes in doubling time for cumulative AIDS incidence

should be expected based on what is known about rate of new infection

 

 
 

 

and of onset of AIDS after infection? As indicated above the rise in
the rate of HIV infection does not appear to be exponential, but more
nearly linear or quadratic. In addition, the risk of AIDS appears to
increase with time after infection: recent data from a cohort of men at
high risk for AIDS in San Francisco suggest that the risk of AIDS is
very low for about 2 years after infection, constant or slightly
increasing for the next 4 years, and more rapidly increasing in years 6
and 7. With any of these combinations for the latency distribution and
a rate of new infection, the doubling time in incidence of AIDS
increases with chronologic time. Thus, from what we know about the
dynamics of the AIDS epidemic and the risk of AIDs after infection, one
would expect the doubling time in cumulative AIDS incidence to increase
over chronologic time, even if there were no changes in sexual practices

and/or I.V. drug use.

ESTIMATING HIV INFECTION RATES

If the risk of AIDS after infection and AIDS incidence were known
exactly, then the rates of HIV infection over time could, in principle
be estimated accurately. In practice, however, very different models of
rate of HIV infection in combination with given estimates of risk of
AIDS after infection yeild estimates of incidence that match the
observed data equally well. This means that there is still considerable
uncertainty in the distribution of times of HIV infection. To
illustrate this point, we estimated the incidence of AIDS for three
different models for rates of infection assuming that the risk of AIDS
after infection is known. The three models used for rise in cumulative

number of infections are the linear, quadratic and epidemic--the latter

 

 
 

 

model results from an assumption that the rate of new infection is
proportional to the number of susceptibel individuals in a population
times the number who are infectious (Figure 1). Figure 2 gives the
corresponding estimates of cumulative AIDS incidence. Although the
three models lead to an estimated cumulative AIDS incidence that mathces
the observed data almost equally well, the three estimates of rise in
number of infected individuals are substantially different and, as we
show below, can lead to very different interpretacions of the magnitude
of the epidemic.

To carry the example further, consider the use of cumulative AIDS
incidence data to estimate the current number of infected individuals.
Such estimates are sometimes obtained by multiplying the total number of
AIDS cases by the ratio of infecteds to cases in a defined cohort of
subjects. In our example, the infected:AIDS ratio in March 1987 is
67:1, 21:1 and 13:1 for the epidemic, quadratic and linear models for
rise in infection respectively. Thus, the three models give vastly
different estimates of the current prevalence of HIV infection. Note
also that incidence data does not provide information about changes in
behavior that have occurred in the past two years because the risk of
AIDS appears to be very small for at least two years after HIV
infection. Thus, had transmission of AIDS among homosexual men ceased
in 1985, the incidence curve for AIDS would be essentially the same as
it is now, yet the 1987 infected:AIDS ratio would then be closer to 20

to 1 for the epidemic model.

INTERPRETING HIV INFECTION DISTRIBUTION

 

 

 
 

 

Finally, consider how the epidemic curve of infection might be

interpreted if an accurate estimate of it were available. The shape of
this curve is influenced by a variety of factors--none well
characterized-- including heterogeneity of the population at risk,
efficiencies of transmission by various routes, and variability in
infectivity of HIV. Population heterogeneity refers both to variability
in behavior among individuals and to behavioral change over time. Even
in populations with no behavioral change over time, the variability
among individuals alone could produce rise in rate of HIV infection
arbitrarily close to any given epidemic curve; although as changes in
the slope of the curve become more abrupt, increasingly complex
population dynamics dre needed to produce them. Therefore, virtually
nothing about changes in behavior can be inferred from this curve. A
decrease in the rate ‘of rise of number of infected ccurs when the number
of individuals at risk drops substantially, yet this drop can result
from a change in behavior or near saturation of the pools most at risk.
Similarly, the curve of HIV infection gives no information about the
degree of spread to lower risk populations (homosexual men practicing
safe sex, promiscuous heterosexuals, etc.) even though such groups may
account for an increasingly large proportion of cases in the future.
Without additional information, changes in doubling times, taken
alone, are of very limited use in explaining or predicting the future

course of the AIDS epidemic.

 

 
 

 

FIGURE 1

Cumulative HIV Infection
(millions)

2.5

 

Ls /

 

 

Pitted Values of Cumulative HIV Infection: 1981-1987

anno ------W, (Tt) = ky (tt, ) for t > % (=June 1979)
————— W(t) - ko (tt)? for t > tT. (January 1979) |
—— — W(t) « of Ik, S(u)I(u) 1/{Z(u)+S(u) Ingdu for t 2 t; (mnuily 1975)

 

 
 

 

FIGURE 2

Incidence of AIDS

 

30000

20000

10000

 

 

 

4 ,

1981 1984 1987

0

Observed arn] Pitted Values of Cumulative AIDS Incidence: 1981-1987
Observed

o+-e----~-- Pitted, based on: W, (t) = k, (tt) for t > % (June 1979)
-—— —— Fitted, based on: W,(t) = ho (tt)? for t > To (—January 1979)
—— ——Fitted, based on: W3(t) = 9/*[kS(u)I(u) J/(1(u)+S(u) Ingdu

for t > tT; (—July 1975)

 

 
 

 

Anthony Robertson:

Testimony before the Presidential Commission on HIV
December 11, 1987 |

i. General.

To control the spread of an infection completely it
is axiomatic that one must be able to tell who is infected
while they are infectious. This information is needed not
only for the purposes of infection containment, but also for
the purposes of planning and of determining whether control
measures have been effective. It is even more important than
usual when, as is the case for AIDS, there is neither cure
nor vaccine nor are any likely.

The AIDS epidemic is unusual in that for various
reasons we do not know who is infected even though we could,
in principle, detect the vast majority. Furthermore, the
syndrome on which reporting, in the United States, is based
occurs at a late, pre-terminal stage in the disease. It
therefore does not give an accurate representation of the
current pattern of infection, even though in broad terms
current infection incidence is likely to-mimic the
geographical pattern of past AIDS cases. Experience from
other countries shows the potential danger of relying on
reported AIDS cases as the best measure of the epidemic.

2. What do the current data show?

The epidemic is still at an early, exponential,
phase of growth except perhaps in some homosexual
populations. We have too little information to make
predictions yet, but must rather model trends from the
current AIDS case data. These empirical models, as used by
the Centers for Disease Control, are useful for short-term
projection, but they are dependent on present trends - for
example of reporting - remaining constant. We have
essentially no current information about reporting
efficiency, except that there is some under-reporting which,
anecdotally, is believed to be high amongst drug addicts.

The limited data on the prevalence of HIV infection
in the United States do not provide grounds for optimisn.
They show that in isolated groups, for example the San
Francisco City Clinic Cohort, infection spread rapidly then
saturated with most people infected. We have almost no
information about the spread of infection in well-observed
groups in the early stages of the epidemic, so we have little
knowledge on which to base any predictions of the spread of
disease amongst the "general" population, where typical
prevalences of infection are well below 1%. It should be
sobering that, even though the surveys involved small numbers

 

 
 

 

of people, quite high prevalences have already been found in
dinner, city: pregnant-women and in men attending .an: emergency
room. In the very limited data available from the sentinel
hospital survey patients from medium risk areas and not in
infectious disease or emergency units have shown prevalences
of up to 0.9%. This too is a disturbing number since it is in
a group at relatively low risk.

The AIDS cases reported to the CDC show several
phases of growth, being roughly exponential up to late 1983,
with a doubling time of about 9 months, then with increasing
doubling times until September 1987 when the overall
epidemic had a doubling time of about 16 months. The case
definition was changed and reporting under the new definition
has been in force since October 1, 1987. The rate ofincrease
in new cases has gone up by about 40%; it will take some time
before we know whether this increase includes only those who
are, in effect, being diagnosed earlier or those who would
not have been reported ever under the old definition. In
either case the growth rate has accelerated with a current
doubling time, if the increase in reporting rate persists, of
12 months. I do not mean to say that the epidemic has
increased in severity, though it may have, but that even the
reported number of AIDS cases is subject to quite extreme
fluctuations and is therefore not a reliable basis for
projection, even to determine medical costs.

The data, insofar as they are available, show no
great differences amongst the various risk groups in rate of
progression to AIDS if infected, nor in death rate. Apart
from a possible effect of age the major differences amongst
the groups relate to the times at which the infection entered
them, as can be seen by looking at the growth rates of AIDS
cases in the groups (attached tables and paper). There is
nothing in these data to suggest that we have had an effect
on the spread of disease (as can be seen by comparing the
"transfusion" data with, say the "heterosexual" data) nor
that there is any great difference between the efficacies of
heterosexual and homosexual transmission.

The data from small cohorts do show two trends that
make modeling and prediction even more difficult: both
infectiousness and AIDS attack rate increase with time.
Indeed, it now seems that the AIDS attack rate is close to
zero for the first two years after infection, at least in
adults. This makes any estimate of the numbers in the general
population infected now from the mubers of AIDS cases
reported now even more hazardous.

3. Data needed.
Por all the reasons I have mentioned we need much

better data of the current spread of the virus. Ideally, a
large random survey of the population should be done and

2

 

 
 

 

repeated at intervals. The next most useful data would come
from expansions of--the. sentinel hospital survey and, perhaps
the easiest, in testing the blood of newborns. Both these
kinds of testing should be started on a wide scale
immediately. Even if they were to show little or no spread of
the virus they would be useful (and encouraging), but this is
unlkikely. As it is we have no way of even comparing our
situation with that of other countries which are ahead of us
in the epidemic. If we delay we shall never get the data on
the early part of the epidemic, and modeling endeavors will
be made much more difficult. Furthermore it seems to me that
we shall have to decide what rates of infection and ultimate
disease are acceptable to our society. A few years ago the
idea of even a million Americans infected with a lethal virus
would have been shocking; we have adapted to it quite readily
- and we are probably being optimistic. Again, it is useful
to look at the actual numbers of reported AIDS cases, which
have just exceeded 1,000 in the CDC's original "heterosexual"
contact risk group. These have doubled three times in the
past two years and stand where the cases reported for
homosexual men were about four years ago. It is sometimes
erroneously assumed that a comparable growth rate does not
equate with a comparable spread of disease; that is, that a
"breaking-out" into the heterosexual population would be
marked by an “explosive” increase in new AIDS cases. The
data we have show that the growth rates, at comparable phases
in the epidemic, were quite similar in the heterosexual and
homosexual groups. We lack the most important data, the
incidence of new infections in the population at large.

 

 

 
 

 

ChC Total Adult AIDS Cases

Coc OATE ‘Bisen.Men IV Drug Abuser Hewophniliacs Hetorosenuals Tf ssions Ushnown/Other Fetal Cases
30 Nov 1987 33,994 7,849 471 1,047 1,046 2,200 46,607
23 Nov 1967 33,860 7,746 471 1,030 1,939 2,194 16,430
16 Nov 198? 33,207 7,528 464 966 4,001 2,197 45,245
9 Nov 1987 32,867 7,366 459 919 286 2,196 14,795
2 Nov 196? 32,332 7,311 452 913 ‘350 2,163 44,129
26 Oct 1987 32,080 ?,258 443 966 936 2,090 43,775
19 Oct 1967 31,548 7,109 430 928. ‘206 2,013 412,934
12 Oct 1987 31,162 6,976 423 930 566 1,973 12,730
8 Oct 1987 NG NEW DATA NEM CASE OFINITN
28 Sep 1967 30,717 6,885 363 916 369 = 1,980 41,770
21 Sep 1987 30,612 6,653 379 902 862. 1,974 41,602
14 Sep 1987 30,360 6,776 3?7 66S ar2 1,960 41,250
? Sep 1997 30,320 6,758 373 ar? 869 1,956 41,161
31 Aug 1967 30,046 | 6,708 370 872 362 1,93? 410,795
24 Aug 1967 29,683 6,61? 369 651 850 1,932 10,282
17? Aug 1987 29,477 6,546 365 836 a44 1,902 39,970
10 Aug 1987 29,0098 6,506 364 626 839 1,690 39,403
3 Aug 1987 26,780 6,396 356 809 829 1,866 39,036
2? Jul 198? 26,516 6,337 354 601 816 1,649 38,712
20 Jul 1967 28,205 6,268 348 792 809 1,833 38,275
A3*Jul 198? 27,892 , 6, 185, 343 e773 79? 1,795 37,705
6 Jul 1987 27,776 6,164 342 769 793 1,791 37,635
29 Jun 198? 27,548 6,120 341 758 vet 1,791 37,347
22 Jun 1987 27,145 6,091 336 ?42 764 1,792 36,870
13 Jun 1987 26,648 6,061 335 733 rai 1,789 36,507
8 Jun 1967 26,421 6,037 328 ?29 729 1,756 36,003
1 Jun 1987 26,067 5,919 328 716 716 1,608 35,544
25 May 1967 26,013 5,902 32? 714 716 3,605 35,47?
10 May 1967 25,847 5,870 . 326 704 7a 1,605 393,266
11 Mey 19987? 25,667 5,843 313 693 701i 1,801 35,020
4 May 1987 25,456 5,800 -309 60? 691 1,782 34,725
2? Apr 1987 25,339 5,793 302 602 606 1,770 34,575
20 Apr 1967 25,016 5,646 300 £63 675 1,?2? 34,029
13 Apr 1987 24,601 6,604 293 654 655 1,729 33,518
6 Apr 1387 24,424 5,565 269 637 644 1,690 33,245
30 Mar 1987 24,242 6,540 284 630 640 1,675 33,011
23 Mar 1987 24,015 6,492 261 608 635 1,676 32,696
16 Mer 1987 23,601 6,421 2r4 399 626 - 1,664 32,263
9 Mar 1987 23,141 6,344 " 266 570 602 1,635 31,526
2 Maer 1967 23,019 5,312 263 Sé? S98 1,614 33,387
23 Feb 1967? 22,460 5,155 258 556 S77? 1,635 30,593
16 Feb 19867 22,409 5,136 25? $50 574 1,569 30,395
9 Feb 3987 22,127 S,128 254 54? 56? 1,565 30, 168
2 Feb 1987 21,9382 6,099 254 340 564 1,950 29,959
26 Jan 198? 21,3728 41,955 254 S22 54? 1,510 29,159
19 Joan 1987 21,339 4,961 252 32% 544 1,507 29,114
12 Jan 1987 21,296 4,921 249 SiS 537 1,508 29,019
3 Jan 1987 21,0486 4,903 247 510 S25 1,494 20,727
29 Dec 1986 20,960 4,646 246 308 524 1,489 26,593
22 Dec 1986 20,898 4,840 245 S06 S22 1,465 28 , 496
15 Dee 1966 20,521 4,705 242 498 515 1,463 28, 124
@ Dec 1986 20,327 4,723 240 463 505 1,424 27,704
1 Dec 1986 20,417 4,760 243 466 5038 1,428 27,843
24 Now 1986 20,357 4,756 241 483 S03 1,42? 27,76?
1? Nov 1986 20, 134 4,643 240 464 419? 1,412 27,390
10 Nov 1966 19,963 4,594 239 459 469 1,396 27,140
3 Now 1966 19,774 9,547 235 454 4165 1,363 26,878
2? Oct 1966 19,489 4,502 229 450 469 1,364 26,503
20 Oct 1966 19,200 4,452 220 446 4158 1,355 26,199
43 Oct 1986 16,950 4,421 226 441 443 1,333 25,814
6 Oct 1986 16,839 4,378 249 433 444 1,329 25,642
29 Sep 1986 16,587 4,322 235 425 436 1,393 23,276
22 Sep 1966 16,506 4,205 214 421 433 1,304 25,163
15 Sep 1906 17,946 4,222 208 a1? 23 1,294 24,510
@ Sep 1986 17,739 4,165 205 413 ats 1,297 24.22A

 

 
 

 

p 1986

196
198
1986

g 1986

1986
1986
1986
19806
19386
1966
1966
1936
1936
1906
19866
1986

y 1986

1986
1966
1986
4966
1986
1986
1986
1986
1986
1986
1986
1986
19386
1936
1986
1986
1986
1985
1985
1985
1985
1985
1985
1985
1983
19083
1995
19385
19385
1985
1983
1985
1984
1964
1964
1984
1984
1984
19384
19904
1984
1984
1984
19384
1983
1983
1983
1983
1983
1983
1983
1983
1993
963
19863

°* 633
‘?2

49

, 166
16,691
16,762
16,523
16,392
16, 196
16,044
15,876
15,707
15,540
18,403
15,22?
14,998
14,842
14,670
14,520
14,351
14,141
13,809
13,646
13,449
13,322
13.097
12,935
12,839
12,689
12,570
12,307
12, 156
11,998
11,910
11,608
11,513
11,118
10,958
10,491
10,308
10,003
9,365
6,861
8,222
?,°32
7,294
6,7??7
6,377
5,921
5,601
4,901
4,566
4,217
3,931
3,620
3,372
3,143
2,819
2,232
2,257
2,052
1,909
1,734
1,é12
1,400
1,224
1,077
971
933
812
745

C732)
Ces
C7323
CP3e
Cran
CP? ano
C73
C?P3HD
CP3HD
C73
C?3zD
C?3%9
C732
C7382
Cesena
CP3R
Cesena
C?Snd
CP3na
Ce3e
C73na
e?3ao
C732)
Ce32d
CP 42>
CP 329
C?32>
CPSR>
C?73e
C?3n>
Cr3ea
C?3H
e7323
C?32>
C2323
C?3H>
CP? 3n>
C?73n>
CP3z>
CPD
Ce3e3
CP3HD
crgn3
C732:
Ces
«?4n9
Ce3e3
CPS
CP322
CP3e>
CP IRD
C?3n>
CF2aa)
Cada
C7223
C?2e3
C72n>
CP2]RD
C?22)
CV 28>
CP2d
CP2aS
CP2HS
C7229
C?P2no
CP ino
CP ida
C72a3
C7 2k
C72Ho
CP?2n>
CPSio

4,142
4,068
4,058
4,006
3,933
3,889
3,850
3,624
3,770
3,755
3,693
3,671
3,641
3,612
3,571
3,524
3,400
3,437
3,332
3,297
3,238
3,164
3,113
3,098
3,036
3,007
2,984
2,940
2,919
2,800
2,804
2,778
2,766
2,698
2,684
2,597
2,557
2,448
2,438
2,342
2,178
2,062
1,914
1,822
1,690
1,505
1,466
1,324
1,209
1,154
1,076
1,010

966

ers

816

759

697

635

594

546

490

460

404

364

338

297

254

231

217?

10?

C177
€172>
C17
€1?%>
€172>
C22
C1?H>
C17HD
C172
Cire
CiPz>
Cis
C17zs
C1Pzs
C1722
Clzea
CAPHs
C172)
C1i?Z2>
C172
C1i7H>
C1779
CLPH>
C172
C172)
C172
C177
C1?2>
C1733
€1?2>
C172
C1P2>
€1?2>
C1?7>
C1729
C172
C1772
C1723
C172)
ct?72>
€17%>
Ci7ea
Cirda
Chee
chen
'¢ &t>)
C1i7e>
C177
C1? HD
C1P HD
C172)
C1723
©1867
C1?7>
C122
ci?
€107
€1087>
18723
©1822
©1872
©1872
©1772
C17
C1675
©1687!
CI7PHz>
Cipro
CLPH>
C172

Cin>
cin>
C123
Clad
Cid
(325
12>
Cle
C1a>
12>
Cia>
e125
Cie
Clad
Cleo
cin
Cd
Cito
Cise>
Cand
Cien>
«12>
Cin>
C1i>
C12
12>
C123
C1i>
Clad
€12>
Ciao
C1ln>
€ia>
Cand
Cia
12>
Cina
Cio
Clas
Ctn5
qin
€1n#>
18>
©1293
C129
€ 12>
C123
C13
C1n>
Cin>
C ino
C43
C1e>
Cla
e123
Cle>
Cln>
C1z>
Cleg3
C423
Cia>
C4>
€1a3
Chea
Cand
C4z>
C122
Clez>
€1n>
© he>

407
406
404
36?
3804
379
31
303
360
329
346
326
322
312
308
304
301
294

wWittrrrarerbt ptt tetraes

to
w

22>
27>
27>
29>
€22>
622)
€2%>
(22>
27>
27>
(22>
C122
€12)
€1%>
c12Z>
c12>
€17>
€172>
12>
€1%>
12>
c12%>
c17>
€12>
€17#>
c172>
¢1z>
47>
37>
€17>
¢€12>
€12>
€17%>
17>
€12>
17>
€12>
€1%>
c1z>
c17%>
€1z>
cin>
Cix>
edn
ci23
€44>
ciz>
ci2>
€12>
12>
€12>
c1iz>

C2323

76%

3194
390
3864
375
307
361
353
351
351
342
339
332
326
321
318
310
308
299
296
268
264
278
27?
275
2°77
279
266
206
203
261
252
252
239

(ob fb eh ote one ot ptr BD A AD AD
OtwLt@ Ou AW
GUwUaUaWwsiaQneagas

162

PONV
QnNnann

 evud ttetbiat_taar~eterea

(22)
C27)
C22)
€272>
22>
«27>
«27>
C22>
(27>
<27>
<27

c27>
C22>
€22>
«22>
€22>
23>
«27>
C273
€3%2>
€22>
€22>
€27%>
€27%>
€2%>
€272>
€2%>
22>
C223
€272>
c22>
C222>
22>
22>
22>
22>
C273
22)
27>
27>
C22>
cin>
cin
€42>
cin
17>
cin
C1iz>
€12>
c12>
€12>
c12>

1,286
1,271
1,255
1,210
1,193
1,190
1,16?
1,169
1,177
1,109
1,163
1,197
1,107
1,177
1,104
1,155
1,134
1,137
1, 126
1,116
1,111
1,069
1,07?
1,004
1,045
1,016
1,015
1,004
998
1,010
993
995
987
984
968
968
949
933
892
683
851
814
785
737
667
663
612
590
575
506
519
491
458
499
541
S03
466
437
410
366
3e1
334
307
292
202
24?
219
202
165
153
139
119
1?

cSr>
C5
cSz>
C5529
S223
«52>
€62>
54>
€$2>
C62)
C62)
CoZz>
C63
C62>
Cé>
C629
C6029
C6243
C6722
¢ FP
«62>
C629
K6>
C6229
€62>
cé2>
«62>
C6223
€6¢>
67>
£62)
«62>
C629
«623
£62)
€62n3
©6293
€67#>
€62>
«62>
62>
e772)
C?e2
Cen?
Crh
Cede
C72)
creda
cra
Cen?
C6e>
€en>
©1023
©1025
©1023
€1025
Chae
Cau
Cries
Cans
alan
€138>
Clits
©1992
C128
€127>
Q1 20
C122>
Cans
Cling
$23

24,085
23,844
23,66?
23,363
22,963
22,792
22,494
22,319
22,044
21,863
21,60?
21,420
21,213
21,010
20,771
20,473
20,240
20.016
19,607
19,540
19,201
16,90?
16,610
16,310
16,143
17,814
17,617
17,469
17,266
17,115
16,759
16,543
16,343
16,227
15,909
15,719
15,176
14,955
14,313
1+, 169
13,443
12,76?
412, 107
11,219
10,553
9,93
9,23
8,693
8,120
7,694
7,064
6,720
6,261
5,627
5,479
5,037
4,690
4,370
3,954
3,640
3,409
3,157
2,868
2,678
2,416
2,259
1,972
1,737
1,508
1, 566
1,300
1,228
1,025»

 

 
 

 

 

Jan 1983
Dec 1982
Now 1962
Oct 1962
3962
Aug 1962
Jul 19682
dun §962

os
QaeeauvwGaeew
wv
e
7

boy
590
634
476
435
361
339
315

C7429
<7S2>
C?Sn>
C762)
C762)
€?62)
c?32)
C??2>

844 €16%3 ~ 86 €20%> OP in
420 ¢€162> - 7@ ¢€102> 760
810 C1645 = 72 €38u> 7386
93 ¢15%> - 63 ¢1023 634
@3 ¢€352> ~ 60 ¢€112> 570
70 €342%> - 64 ¢102%> S03
79 ¢162> - 34 ¢62z)> 4520
76 1923 - 22 (623 413

= denotes a tine-pericd in which COC redefined risk groups.

CD Total Adult ALOS Cases
COC DATE = Hono/Bisen.ften IV Drug Abuser Henophiliace

Heteresewels Vransfusiens Unknown/Other otal Cases