AARP Public Policy Institute INSIGHT on the Issues Excluding Older, Sicker Patients from Clinical Trials: Issues, Concerns, and Solutions Keith D. Lind, JD, MS AARP Public Policy Institute This Insight on the Issues discusses issues, concerns, and possible approaches to expanding clinical trials to include older people and those with chronic conditions to make research more relevant for clinicians treating these people. Background seek to isolate the treatment effect from other factors, such as age, sex, and People with multiple chronic conditions health status. In the most rigorous form (also referred to as comorbidities) of clinical trials—randomized controlled represent more than one-quarter of all trials (RCTs)—members of a single Americans and are responsible for group of volunteers are randomly 65 percent of all health care utilization. assigned to either a treatment or control As people get older, the prevalence of group. Differences in the clinical chronic conditions increases. Medically outcomes experienced by otherwise complex, older patients account for a similar patients are assumed to arise disproportionate share of health care solely from differences in treatment. services, use more prescription drugs and medical devices, and experience The Problem more complications and adverse drug events than younger, healthier patients. Most hospital and physician encounters By the time they reach Medicare involve older people with multiple eligibility, one-quarter of the population chronic conditions. Yet clinical trials has five or more chronic conditions and often exclude or under represent this is responsible for more than two-thirds group of people. This of Medicare spending. underrepresentation is not limited to prescription drug studies but Clinical trials are intended to test the encompasses the full range of medical efficacy and safety of medical interventions, including medical devices interventions, such as prescription drugs, and procedures. Underrepresentation of medical devices, and surgical older, sicker adults in clinical trials procedures. Medical researchers perform makes it difficult to understand the full clinical trials to obtain reliable, unbiased impact of medical interventions and may information about the effects of various lead to inappropriate use of interventions clinical interventions on patients with a in these populations. A report from the target illness or condition. By comparing U.S. Food and Drug Administration the effects of a discrete intervention on (FDA) finds that adults over 50 are more one group of patients, the treatment likely to experience a more intense and group, with the effect of standard or greater variety of side effects to any routine therapy on a second group of single medication than other adults.1 patients, the control group, researchers Excluding Older, Sicker Patients from Clinical Trials: Issues, Concerns, and Solutions about the risks and benefits of Clinical Trials Often Exclude Older, participation. Anecdotal evidence Sicker Patients suggests that many researchers Many clinical trials explicitly exclude incorrectly assume that older and sicker older, sicker patients. Even in trials adults do not want to participate. without explicit exclusion criteria, the However, this exclusion is often poorly percentage of older adults included is justified. Many older people want to low. A comprehensive study of the participate in clinical research for a extent to which RCTs published in major range of reasons, from altruism to self- medical journals have excluded certain interest. Further, there is little evidence patient populations found that, over to suggest that older patients cannot 12 years (1994–2006), 81 percent of tolerate or benefit from participation in trials excluded complex, older patients clinical trials. In any case, methods for with common medical conditions, while increasing the participation of older 72 percent excluded patients on the basis patients are available.5 of age.2 While the prevalence of age-based Arbitrary exclusions, particularly of a exclusions per se is declining, exclusions patient population at higher risk of the based on comorbid conditions are clinical condition, raise questions increasing, which also has the effect of regarding the generalizability of clinical excluding older adults. A study of RCTs trial results and their relevance for published in five major medical journals routine clinical practice. For instance, a in 2007 found that 46 percent of studies study of 20,388 Medicare beneficiaries excluded patients based on criteria such with heart failure who were discharged as physical or functional limitations, from acute care hospitals in the United cognitive impairment, or shortened life States found that only 13 to 25percent expectancy, and 20 percent excluded met the enrollment criteria of three patients based on an upper age limit. landmark RCTs that have influenced the These criteria disproportionately exclude treatment of all patients with congestive older adults.3 heart failure.6 RCTs conducted internationally also Research based on clinical trials that exclude or under represent older adults have excluded complex, frail, older and people with co- morbidities. A study patients can produce findings that are of trials under way in the United States incomplete or of dubious relevance, and Europe in 2008 of patients with leading to poorly founded practice heart failure found that 80 percent guidelines and quality of care measures. excluded patients with comorbidities and In this case, clinicians may be faced with 25 percent excluded patients on the basis a “risk-treatment paradox” in which of an arbitrary upper age limit.4 high-risk patients are denied safe interventions capable of materially improving survival and/or quality of life While many studies have successfully or, on the other hand, receiving recruited medically complex, older inappropriate interventions in which patients, the majority of trials are not risks are likely to outweigh benefits. recruiting them proactively. Physicians raise a number of concerns about Physicians may also face a dilemma recommending that their patients when seeking to apply multiple practice participate in clinical trials, such as guidelines that conflict or are doubts about their patients’ ability to incompatible because the evidence on adhere to complex trial regimens and which they were based did not take 2 Excluding Older, Sicker Patients from Clinical Trials: Issues, Concerns, and Solutions adequate account of patients with Yet the exclusion of older people or multiple chronic conditions. those with multiple chronic conditions also raises safety concerns. FDA rules Although the FDA has published allow off-label use of prescription drugs guidelines for the study of drugs likely for patients with characteristics that were to be used in the older population, these not represented in clinical trials in which guidelines do not include specific the drugs were tested. Off-label use of standards governing the inclusion of many prescription drugs is often higher older people or those with complex in older populations (e.g., antipsychotics medical conditions in clinical trials for for agitation in older people). Because drugs or devices.7 older, sicker patients are often underrepresented in trials, understanding Another obstacle to expanding clinical of their response to off-label use of a trials is the cost of the intervention. drug (such as side effects, adverse While Medicare will pay the cost of reactions, and interactions) may be routine patient care for beneficiaries incomplete. The ongoing practice of off- enrolled in FDA-approved clinical trials, label use suggests the need to include Medicare and most private insurers will populations with these characteristics in not cover the cost of items or services clinical trials. that are considered experimental or investigational. In many cases, these Some experts feel it is unethical NOT to costs are borne by the trial sponsor. In include these populations in trials, since some cases, patients may be required to they represent a large percentage of the bear these costs. people who receive drugs and other medical interventions. These experts There is general agreement in the contend that it is better to include these scientific and clinical communities that people so as to identify issues in a underrepresentation of older adults and controlled setting where adverse effects people with multiple chronic conditions can be monitored and addressed, rather is problematic. However, there is less than leaving practitioners and patients to agreement about the best approaches to fend for themselves in unsupervised address this problem. environments. However, participating in Proposals to expand participation in clinical trials may place older, sicker clinical trials have raised a number of patients at unnecessary or issues related to the following: disproportionately higher risk without the likelihood of commensurate benefit, ■ Safety and ethical considerations raising ethical concerns. ■ Efficacy and effectiveness These competing concerns raise ■ Costs questions about whether clinical trial participation must be in the best interest Safety and Ethical Considerations of the patient or can be conducted for the benefit of future patients and Safety concerns around expanding practitioners, even though the patient clinical trials include potentially higher might be placed at increased risk. rates of morbidity and mortality among Concerns about the safety and ethics of populations that may be frail and at expanding trials raise questions about heightened risk of complications, and whether safety concerns can be thus more prone to suffer serious adverse separated from ethical concerns. Such outcomes and side effects in trials. questions suggest that issues related to expanded trial participation may deserve 3 Excluding Older, Sicker Patients from Clinical Trials: Issues, Concerns, and Solutions further analysis from an ethical thus, to estimate the trial size necessary perspective. to obtain reliable results. Efficacy/Effectiveness When selecting outcome measures for Considerations trials, in addition to measuring the impact of an intervention on morbidity When trying to design trials that produce and mortality, it is important to include more relevant and generalizable results patient-centered measures that reflect the for clinicians caring for older, sicker values of the target population. At the patients, the goal is to include trial same time, the response burden of participants who are similar to patients additional measures on trial participants whom clinicians typically face in routine and sponsors’ resources must be practice. Such trials, known as pragmatic considered. In order to give advice based clinical trials, are likely to include on individual patient priorities, clinicians patients who are older and more need relevant information on patient- medically complex (i.e., more centered measures. For instance, older, heterogeneous) than most current trial sicker patients may value an participants. However, there are intervention’s effect on quality of life legitimate scientific concerns about (i.e., function and comfort) as much as making trials more heterogeneous. For or more than its effect on morbidity and instance, due to the variety of factors mortality. Collecting data on that may affect outcomes, such trials are standardized quality-of-life measures likely to make reliable measurement of across all trials could increase the the intervention effect more difficult. relevance and comparability of trial results, but it could also add significant Whether more relevant and reliable cost to trials. results can be obtained by expanding trial participants to include older, sicker Cost and Business Considerations patients (or by conducting separate trials dedicated to such populations) is likely Increasing the proportion of older, sicker to depend on the specific intervention patients could increase costs associated and the characteristics of the populations with conducting clinical trials. A larger for whom relevant results are desired— number of patients could be required to for example, whether the target ensure the numerical power necessary to population is middle aged and relatively achieve statistical significance among healthy (homogeneous) except for the subgroups in a more diverse population. target illness, or older and sicker Increasing sample size, particularly with (heterogeneous). older, sicker patients, could produce more adverse events, which could In addition, in order to ensure a increase monitoring costs. The cost of comparable level of confidence in the follow-up efforts could also increase due results of a heterogeneous trial and draw to changes in seasonal residence and conclusions about specific more deaths during the trial. subpopulations, the number of trial participants must be increased, On the other hand, expanding the pool of sometimes substantially, compared to a potential participants could make more homogeneous population. In any recruiting patients and trial sites easier case, when enrolling more and faster. Accelerating the recruitment heterogeneous populations, it may be process could shorten the duration of difficult to predict the level of potential trials and lower the screening-to- differences in treatment effects and, enrollment ratio of participants. These 4 Excluding Older, Sicker Patients from Clinical Trials: Issues, Concerns, and Solutions measures could reduce the number of The following are lower cost alternatives trial sites and administrative personnel. to clinical trials dedicated to older, Relying on centralized follow-up and sicker patients: monitoring facilities might also reduce costs. Shortening the duration of clinical ■ Informal trials in which patients trials could accelerate the time to get serve as their own controls during commercial products to market and randomly sequenced periods of use generate revenue that would offset the vs. nonuse cost of research and development. ■ Post market surveillance during which data are collected and Whether the net effect of these changes analyzed following FDA approval to would increase or decrease costs would market a drug or device depend on the trial. Performing case studies and computer simulation models ■ Observational data analysis using for several types of trial cases and their data collected for non-research costs could improve our understanding purposes, such as insurance claims of the interaction of various factors that data might affect total cost. ■ Relying on expert opinion through As a competitive matter, industry consensus guidelines, rather than funders may be concerned that performing trials expanding inclusion criteria could However, each of the alternatives has increase the cost of trials and produce a limitations, including less validity and less favorable profile of results, reliability than RCTs. particularly adverse effects, when compared to competing interventions Discussion that were previously tested under narrower inclusion criteria. A number of policy options are available to address inappropriate exclusions and Rather than expanding clinical trials to expand clinical trials to include older include older, sick patients, resorting to people and those with multiple chronic dedicated clinical trials that include conditions. However, many of these mostly older, sicker patients could options could be controversial and produce more relevant and reliable potentially costly. Some options would clinical data. This approach could avoid require further development and testing the added complexity associated with before implementation. The following conducting more trials that enroll more a list is offered merely to identify possible homogeneous, low-risk population. policy options and is not intended to However, undertaking dedicated trials represent an endorsement or even a specifically focused on medically ranking of their relative strengths and complex, older patients would add the weaknesses. cost of these trials to the total cost of obtaining information about these ■ Federal agencies could take the populations without offsetting savings. initiative to ask whether older adults Some experts feel it is unrealistic to or people with multiple chronic expect separate trials to be undertaken conditions will be included in specifically for older adults and/or proposed trials. multiple chronic conditions. — The Department of Health and Unfortunately, funding remains Human Services has established inadequate to sufficiently expand a “Strategic Framework for geriatric research.8 Multiple Chronic Conditions” 5 Excluding Older, Sicker Patients from Clinical Trials: Issues, Concerns, and Solutions that calls for ensuring that ■ The FDA could require label individuals with multiple chronic warnings regarding use of drugs and conditions are not unnecessarily devices for older, sicker patients. excluded from clinical trials.9 — Such warnings might be designed — Broader inclusion criteria could to reduce the risk of avoidable be made part of federal research complications in vulnerable grant-scoring criteria, much as populations for which the the National Institutes of Health products have not been tested. (NIH) established policies in the — However, the FDA could not mid-1980s for inclusion of extend such protections to use of women and minorities in clinical procedures that it does not research.10 regulate. ■ Post market surveillance studies ■ Congress could create financial could be undertaken to assess incentives to encourage adverse drug reactions arising from manufacturers to expand clinical prescribing patterns for older adults trials to include older, sicker patients and people with multiple chronic by offering longer exclusivity for conditions compared to younger, drugs, devices, and procedures that healthier populations. have been tested on such patients ■ Institutional review boards, which through extended patent protection, must give prior approval for clinical exclusive FDA approval, and trials, could encourage investigators favorable Medicare reimbursement to adopt broader inclusion criteria for policies. proposed trials. — For example, “orphan drugs” ■ Medical journals could encourage currently receive extended broader inclusion criteria through exclusivity through the FDA their publication guidelines and approval process. manuscript selection process. Congress could mandate expanded ■ ■ Public payors, such as Medicare and participation in federally funded the Department of Veterans Affairs, clinical trials. and private payors, such as health — For example, Congress has insurers and employer health plans, previously mandated inclusion of could offer “conditional coverage” women and minorities in clinical for experimental and investigational trials when it codified similar items and services that are not NIH policies in 1993.11 currently covered, as long as they are furnished as part of an approved — Until effective approaches and clinical trial, to encourage research potential pitfall have been on older, sicker patients. adequately explored, mandating expanded clinical trials could have unintended consequences 6 Excluding Older, Sicker Patients from Clinical Trials: Issues, Concerns, and Solutions Endnotes 1 M. Meadows, “Medication Use and Older Adults,” FDA Consumer Magazine, July–Aug. 2006. 2 H. G. C. Van Spall, A. Toren, A. Kiss, and R. A. Fowler, “Eligibility Criteria of Randomized Controlled Trials Published in High-Impact General Medical Journals,” JAMA 297(11):1233–40. INSIGHT on the Issues 3 D. M. Zulman, J. B. Sussman, X. Chen, et al., “Examining the Evidence: A Systematic Review of the Inclusion and Analysis of Older Adults in Randomized Controlled Trials,” Journal of General Internal Medicine 26 (2011):783–90. 4 A. Cherubini, J. Oristrell, X. Pla, et al., “The Persistent Exclusion of Older Patients From Ongoing Clinical Trials Regarding Heart Failure,” Archives of Internal Medicine 171(6) (2011):550–56. 5 M. D. Witham and M. E. T. McMurdo, “How to get older people included in clinical studies,” Drugs & Aging 24(3) (2007):187–96. 6 F. A. Masoudi, E. P. Havranek, P. Wolfe, et al., “Most hospitalized older persons do not meet enrollment criteria for clinical trials in heart failure,” American Heart Journal 146 (2003):250–57. 7 U.S. Food and Drug Administration, Guidelines for the Study of Drugs Likely to be Used in the Elderly (Rockville, MD: U.S. Food and Drug Administration, Center for Drug Evaluation and Research, 1989). 8 R. Besdine, C. Boult, S. Brangman, et al., “Caring for older Americans: the future of geriatric medicine,” Journal of the American Geriatric Society 53(6 Supp) (2005):S245–56. 9 U.S. Department of Health and Human Services, Multiple Chronic Conditions—A Strategic Framework: Optimum Health and Quality of Life for Individuals with Multiple Chronic Conditions (Washington, DC: U.S. Department of Health and Human Services, Insight on the Issues 57, December, 2011 December 2010) http://www.hhs.gov/ash/initiatives/ mcc/mcc_framework.pdf. AARP Public Policy Institute 10 National Institutes of Health, Office of 601 E Street NW, Washington, DC 20049 Research on Women’s Health, “Inclusion of www.aarp.org/ppi Women in Research,” 202-434-3845, ppi@aarp.org http://orwh.od.nih.gov/inclusion.html. © 2011, AARP. 11 NIH Revitalization Act of 1993 (PL 103-43). Reprinting by permission only. 7