AARP Public Policy Institute INSIGHT on the Issues How Consumer Choice Affects Health Coverage Plan Design This paper outlines some of the challenges of designing a sustainable health coverage program to serve people with differing health needs. Coverage choices can help attract broad enrollment but can have costs. Policymakers should include program design elements in their reform programs to moderate the costs of choice. Introduction health reform, policymakers will also have to grapple with the issue of what Health care reform was a top issue in the health coverage should look like, how presidential campaign and is a high much choice to offer, and how to priority for the Obama administration. balance that choice with the One of the many aspects of reform is the accompanying costs. challenge of developing and shaping new health coverage programs for To help think through some of the many people who do not currently have factors involved in addressing this coverage. While there are many paths to problem, the AARP Public Policy expanding access to coverage, all of Institute convened a small group of them present the challenge of designing policy researchers, health care industry a program to meet the differing needs participants, state policymakers, and within the targeted population. consumer advocates to discuss a number of topics related to designing health care Policymakers in state reform efforts benefit packages. Drawing on its have had to grapple with these issues. actuarial analysis and experience For instance, Vermont policymakers had advising large employer health plans, to decide whether to standardize Mercer provided the information benefits, what benefits to offer under presented here to illustrate some key Vermont’s Catamount Health program, issues related to choice in health care and how to design the coverage so it programs. would be attractive to potential enrollees and affordable to both enrollees and the  the uneven distribution of health state, which subsidizes the cost of the costs across the population program. Massachusetts policymakers  how individuals’ choices about their had to decide what minimum benefits coverage options play out over time health coverage must have to satisfy the  The costs stemming from mandate that all state residents have individuals’ choices coverage. They also decided on the These issues present challenges to design of four different levels of employers, and in the context of a health coverage to be available through the reform debate, to policymakers Connector. In the context of national designing reforms. Choice of coverage How Consumer Choice Affects Health Coverage Plan Design can help attract participation of a cross because participants often (and section of the healthy and sick. But it naturally) choose their health coverage also potentially reduces the pooling and in their own best financial interest. For spreading of risk. Unless policymakers instance, if offered a choice of plans, recognize the impact of adverse individuals in good health may select the selection and include program design less expensive option or may decide not approaches to mitigate its impact, there to get coverage. can be unanticipated negative effects on costs for individuals and program To provide insight into how adverse sponsors. selection can affect the plans being considered in the health care reform Understanding the Distribution of debate, this analysis presents a Health Claims Costs distribution of claims in a large population of beneficiaries with Employers consider many factors as they employer-provided coverage. The try to control the ever-increasing costs of analysis is based on data obtained from health care. The primary areas of focus the MarketScan® databases from for employer health plans are as follows: Thomson Reuters (see appendix). The MarketScan databases contain fully  Managing the care of those who are integrated inpatient and outpatient ill or injured through programs such medical claims and encounters, as catastrophic care management, prescription drug experience, and centers of excellence, and disease enrollment and eligibility data. The management programs analysis focuses only on active members  Maintaining the health of those who who had medical and pharmacy are not yet ill through health coverage for the full 12 months of 2006. promotion, disease prevention, and This subset of the database represents wellness programs about 7.4 million members. While there  Engaging consumers to increase are likely differences between behaviors cooperative efforts in managing and observed in a population of beneficiaries maintaining health with employer-provided coverage and a  Aggressively managing their vendors population covered under a public and programs to keep costs within program, these data should be useful in budgets illustrating the distribution of costs across a covered population and the As employers and their advisors project impact of adverse selection costs driven the cost of health programs for the by various plan design decisions. purpose of budgeting and setting premium levels, they consider utilization Figure 1 shows the portion of total cost of services (the amount of health care attributable to various components of the consumed), the price of these services, total population. The column on the left and other factors. When employers offer shows a percentage of members and the a choice of options—either multiple plan column on the right shows the options or the choice of participating in a percentage of total claims, or costs, single plan vs. opting out of coverage associated with that percentage of entirely—one of the other factors they members. The chart shows that the most must consider is adverse selection. costly 5 percent of plan members are Adverse selection occurs when a responsible for almost 50 percent of total disproportionately high number of claims cost. Adding the next 15 percent people in poorer than average health of members shows that the most costly enroll in a health plan. This occurs 20 percent of members drive about 80 2 How Consumer Choice Affects Health Coverage Plan Design Figure 1 Distribution of Program Members and Their Associated Health Claims Costs 0% % of Members % of Cost Less0% Expensive 0% 5% 0% 16% 0% 50% 0% 0% 31% 0% 0% 30% 0% 48% 0% More0% 15% Expensive 0% 5% percent of total claims cost. Note also health needs to decide which option to that the least costly 50 percent of choose. If healthier people choose lower members drive only 5 percent of total levels of coverage, and less healthy cost. people choose plans they perceive to have more generous coverage, risks and In addition to the “80/20 group,” it is costs are not distributed evenly across useful to focus on two “50/5” groups— available health plan options. Employers the most expensive 5 percent of and policymakers designing health participants, who account for programs need to weigh this fact when approximately 50 percent of cost, and determining how much choice to offer the least expensive 50 percent of and when budgeting for their benefits participants, who account for program and setting premiums. approximately 5 percent of cost. To control current-year costs, employers The Challenge of Adverse must focus on the 5 percent of Selection participants who are responsible for 50 percent of total cost as well as the What follows is an analysis of claims remainder of the most expensive 20 data to show the importance of the percent. To control future-year costs, distribution of claims. The enrollment employers must also help the least decisions of a few people with large expensive 50 percent to 80 percent claims costs will affect the average maintain their health. spending across everyone in the plan. A plan sponsor’s understanding of the To make coverage attractive to a wider claims distribution is also important for range of employees and to encourage informing decisions about what plan(s) it them to purchase coverage, employers offers and how it sets premiums. often offer a choice of plans. In doing so, however, employers must consider the To allow a deeper understanding of the financial impact of their employees’ impact of adverse selection, the analysis choices. Adverse selection occurs when looks at additional detail from the claims individuals use knowledge of their own distribution. Figure 2 illustrates a portion 3 How Consumer Choice Affects Health Coverage Plan Design Figure 2 Cumulative Claims Distribution for Large Employer Health Plans 100.0% Members with claims less than $X 90.0% 68% below $2,000 80.0% Mean = $3,337; 70.0% 76% below mean 60.0% 50.0% Median = $884; 73% below $2,600 40.0% 50% below median 30.0% 20.0% 10.0% 0.0% 0 0 0 0 0 0 0 0 0 0 0 00 $0 ,00 ,50 ,00 ,50 ,00 ,50 ,00 ,50 ,00 ,50 ,00 $5 $1 $1 $2 $2 $3 $3 $4 $4 $5 $5 $6 Claim s per Mem ber Percentage of members w ith claims less than $X of the cumulative distribution of claims raises the average claim significantly in a typical large employer plan. It above the median. shows the percentage of enrollees with annual claims of $6,000 or less. The two charts in Figure 4 show the difference between the distribution of Based on the claims distribution, the medical claims in an employer group median annual claim for a member is and a normal curve distribution (note $884; half the members would have that these charts show the percentage of annual medical claims below that people at a specific value, not the amount. However, the mean, or average, cumulative percentage below a value). annual claims cost for this group is $3,337; 76 percent of the members of The difference between the median and this group have costs below the mean. mean has significant implications for the The fact that more than three-quarters of design of a program offering choice— members have claims lower than the particularly when one of the choices is to average surprises many people, who are opt out of coverage entirely. In light of more used to working with a normal the underlying claims distribution and (bell-shaped) distribution where the the varying generosity of plans, mean and median are the same. employers or other program sponsors face a number of challenges in Figure 3 expands the illustration to show determining both what plan(s) to offer the cumulative distribution of annual and how to set the participants’ claims for amounts up to $50,000 to help premium. explain why this occurs. The relatively small number of extremely high annual  If a program charged every member claims, commonly associated with the full average cost for coverage events such as catastrophic accidents, (i.e., no employer premium premature births, and serious disease, contribution) and required all 4 How Consumer Choice Affects Health Coverage Plan Design Figure 3 Cumulative Claims Distribution for Large Employer Health Plans Members with claims less than 100.0% 90.0% 80.0% Mean = $3,337; 76% below mean 70.0% $X 60.0% 50.0% Median = $884; 50% below median 40.0% 30.0% 20.0% 10.0% 0.0% $5 0 0 0 00 $1 00 $1 00 00 $2 00 $2 00 $2 00 $3 00 00 $3 00 $3 00 $4 00 $4 00 00 $4 00 $5 00 00 $2 0 ,50 ,00 ,50 $ 0,0 2,5 5,0 7,5 0,0 2,5 5,0 7,5 0,0 2,5 5,0 7,5 0,0 2,5 5,0 7,5 0,0 $7 $1 $1 $2 $3 $4 Claim s per Mem ber Percentage of members w ith claims less than $X members to participate, about one-  If the entire population enrolled in a quarter of the members would high-deductible plan with a $2,600 benefit financially (i.e., would have deductible, premiums would be claims higher than their premium) lower, but 73 percent would have to and the remaining three-quarters pay all their claims out of pocket. would contribute more than they  If the entire population enrolled in a received in benefits in any single limited benefit “mini-med” plan with year. This could lead to discontent a $100 deductible and a maximum among a large portion of the annual benefit of $2,000, population, although the insurance approximately 70 percent would would spread risk over the total have all their claims paid by the plan. population. However, about 30 percent would Figure 4 Claims Distribution Examples Claims Distribution Normal Distribution (% of Enrollees with Claims Below...) 14% 16% Median = Mean 12% 14% Median = $884 12% 10% Mean = $3,337 10% 8% 8% 6% 6% 4% 4% 2% 2% 0% 0% 200 400 600 800 1,000 1,500 2,000 2,500 3,500 4,500 6,000 8,000 10,000 15,000 20,000 30,000 40,000 50,000 200 400 600 800 1,000 1,500 2,000 2,500 3,337 4,000 5,000 7,000 9,000 12,500 17,500 25,000 35,000 45,000 60,000 - - 5 How Consumer Choice Affects Health Coverage Plan Design have to pay out of pocket for claims not distributed across plans in direct in excess of the annual maximum. proportion to enrollment, and adverse The expensive 80/20 and 5/50 selection occurs. Mercer actuaries have groups could suffer catastrophic observed the impact of adverse selection losses. in multi-option employer health Participants clearly do not know programs since the mid-1980s. Based on precisely what claims they will have in an analysis of empirical data, observed the following year, so they cannot experience has been categorized into “perfectly” select a plan to their best levels of intensity of selection impact. financial advantage, and factors Table 1 compares light, moderate, and unrelated to anticipated claims intense selection to neutral selection (no experience, such as the following, impact of selection; all plan choices have clearly affect coverage choices: the same proportion of high- and low- cost participants as the total population).  Risk aversion vs. risk acceptance— Mercer also shows “perfect” selection, in some will buy a “rich” plan just to which the highest cost participants reduce the possibility of significant choose the most generous plan and the unanticipated cost, while others are lowest cost choose the leanest plan. willing to accept the risk of higher Although perfect selection is not out-of-pocket claims cost in return observed in actual experience, it for a lower premium. provides an additional point of  Lack of financial resources to comparison. purchase coverage. Generally, there is an inverse  Inertia—reluctance to change from the current plan or unwillingness to relationship between the impact of adverse selection and the proportion of spend the time to research options. the population in the high option plan— However, when offered a choice of the greater the percentage selecting the health plans, many consumers will use high option, the less the impact of their knowledge of likely medical costs adverse selection. Table 1 compares the to get the best deal. When consumers impact of adverse selection if 20 percent with above-average health care needs of the total population selects the high prefer “richer” plans, risks and costs are Table 1 Effect of Selection and Enrollment Levels on Plan Cost 20% Enrolled in High Option 80% Enrolled in High Option Level of % of claims in Selection % of claims in High Ratio to Neutral High Ratio to Neutral Perfect 77.3% 3.87 99.9% 1.25 Intense 50.1% 2.51 91.4% 1.14 Moderate 38.6% 1.93 88.2% 1.10 Light 27.0% 1.35 83.9% 1.05 Neutral 20.0% 1.00 80.0% 1.00 No selection when 20 percent of the No selection when 80 percent of the population selects the high option: population selects the high option: 20 percent of the population generates 80 percent of the population generates 20 percent of the claims. 80 percent of the claims. 6 How Consumer Choice Affects Health Coverage Plan Design option vs. 80 percent of the population (greater participation = less severe selecting the high option. adverse selection)  The financial effectiveness of the  With no selection at all (“neutral” choice of plan options: How well selection), 20 percent of the people match their coverage to population would generate 20 claims they incur (greater percent of claims, and 80 percent effectiveness = more severe adverse would drive 80 percent of claims. selection) Under a “light” selection scenario, At a more concrete level, the following 20 percent in the high option factors affect the severity of adverse accounts for 27 percent of claims; a selection: “moderate” selection scenario has 38 percent, and an “intense” scenario  Level of subsidy from plan sponsor: results in 50 percent of claims driven The higher the subsidy, the greater by the 20 percent enrolled in the high the participation and the less severe option. None of these scenarios is the adverse selection nearly as bad as the 79 percent of  The length of time until a choice can claims cost attributable to the sickest be changed: The shorter the time a 20 percent of enrollees, but a participant must wait to make a significant selection issue is present change, the lower the participation when a fairly small percentage of and the more severe the adverse people are in the high option plan. selection (e.g., if participants could Under these theoretical scenarios, the sign up at the beginning of any ratio to neutral is anywhere from 35 month, many would not pay for percent to 150 percent above neutral. coverage until after they become ill  Looking at the high option data in or injured) the table, with 80 percent of  The difference in perceived value of enrollees in the high option plan, the plan choices: The greater the total cost experience would vary difference in perceived value, the from about 84 percent to 91 percent more effective the financial choice instead of 80 percent with neutral and the more severe the adverse selection. This is between a 5 percent selection (e.g., as the difference in and 14 percent excess above neutral the plans’ deductibles increases, the experience. As the percentage of severity of adverse selection enrollees participating in the high typically increases as well) option plan increases, the potential for adverse selection decreases.  The ability to project future-year costs: The greater an enrollee’s Limiting the Impact of Selection ability to predict future cost, the greater the adverse selection (e.g., Mercer’s observations of actual selection chronically ill people with significant experience over the past 20 years maintenance drug costs can do a indicate a number of factors that affect much better job of projecting future the level of adverse selection. At a cost and picking financially optimal conceptual level, the severity of adverse plans than people who are currently selection is affected by the following: healthy) Relative to design implications, wider  The participation percentage: The choice is beneficial because it satisfies a percentage of those who are eligible broader range of desires and attracts who actually buy coverage more enrollees, but problematic because (participate) and do not opt out it increases the intensity of adverse 7 How Consumer Choice Affects Health Coverage Plan Design selection. For example, programs would financially because the cost is so high need to cover catastrophic claims to that only those who are reasonably protect the sickest 20 percent of the certain of high future claims costs will population (who account for 80 percent take the plan, creating a “death spiral” of the costs). They might also need to for the program. To reduce the chance of offer some benefits to the healthiest this, program managers should carefully participants—the 50 percent with choose design elements such as 5 percent of claims—to encourage their eligibility rules, waiting periods, and participation. If the difference between premium levels. plans that protect the high-cost participants and those that appeal to the Among the design components that healthy is too great, they could fail could help reduce this concern are the financially due to unmanageable adverse following: selection cost. Program managers should  For someone who had opted out and weigh this fact when deciding on the wants to opt in later, adding an extra range of choices to offer. fee to the premium a participant must Relative to design implications, as pay as time elapses participants’ level of knowledge  Putting limits on coverage related to increases through health risk conditions that emerged in the time assessments, screening programs, and between opting out and opting in communication programs, participants  Increasing the time that someone will likely become more engaged with must wait between a change in health programs that help to control cost, such and the purchase of “richer” as health promotion, disease prevention, coverage and disease management. At the same  Structuring the participants’ time, an increase in participants’ premium contributions so people are knowledge of their health risk and likely more indifferent between plan future expenditures, as well as greater options, so the premium for a knowledge of health benefits, may “richer” plan may be set at a smaller increase the financial cost associated share of average cost to make it with adverse selection. Program attractive to healthy people managers must recognize both these  Offering wellness benefits or other factors when determining design and benefits to attract healthy prices. participants to the higher cost plans Some design choices can lead to Conclusion increases in the intensity of adverse selection so severe that they could render Much more than adverse selection needs an otherwise quality program financially to be considered when policymakers unmanageable. As an example, allowing develop reform proposals and plan participants to opt out when they do not sponsors look at health care plans. think they need coverage and opt in Health care plans have to manage the without penalty when circumstances sick, maintain the health of those who change would render a program are healthy, and engage consumers. financially unsound, as it violates the There are opportunities for policymakers fundamental principles of insurance. If and plan sponsors to proactively the program ends up with a small incorporate program provisions that percentage of eligible members mitigate the impact of adverse selection. participating, it becomes extremely After critical decisions have been made difficult for the plan to survive 8 How Consumer Choice Affects Health Coverage Plan Design about the elements to be included in a program, it is important to consider the implications of adverse selection in program design and pricing. Without understanding the impact of adverse selection and incorporating design INSIGHT on the Issues components to mitigate its effect, an otherwise well-designed program runs a greater risk of failure. Insight on the Issues I23, February, 2009 Written by George Wagoner, Mercer AARP Public Policy Institute, 601 E Street, NW, Washington, DC 20049 www.aarp.org/ppi 202-434-3849, ppi@aarp.org © 2009, AARP. Reprinting with permission only. 9 How Consumer Choice Affects Health Coverage Plan Design Appendix The MarketScan databases from Thomson Reuters represent several data sets containing fully integrated inpatient and outpatient medical claims and encounters, prescription drug experience, enrollment and eligibility information, and productivity data. MarketScan databases contain information on 73 million unique people (44.5 million commercial lives alone), from which Thomson Reuters creates population-specific extracts to meet a wide range of analytic needs for researchers, academicians, business professionals, and government agencies. Information from the large employer claims distribution (from the subset of the MarketScan data set) used in this Insight on the Issues follows. Large Employer Group Claims Distribution: Annual Claims Annual Claims Amount Percent Below Amount Percent Below $100 18.0%* $4,500 82.9% $200 24.5% $5,000 84.5% $300 30.0% $6,000 87.1% $400 34.6% $7,000 89.1% $500 38.5% $8,000 90.7% $600 42.0% $9,000 91.9% $700 45.1% $10,000 93.0% $800 47.9% $12,500 94.8% $900 50.4% $15,000 96.0% $1,000 52.7% $17,500 96.8% $1,250 57.6% $20,000 97.4% $1,500 61.6% $25,000 98.2% $1,750 64.9% $30,000 98.6% $2,000 67.7% $35,000 98.9% $2,250 70.2% $40,000 99.1% $2,500 72.3% $45,000 99.3% $3,000 75.8% $50,000 99.4% $3,500 78.6% Over $50,000 100.0% $4,000 81.0% * 12.5% have $0 claims 10