Medicare reform has been the focus of federal health policy since

Medicare Medicare Versus Private Insurance: Rhetoric And Reality Medicare provides a level of security that is not typically found in employer or individual coverage markets. by Karen Davis, Cathy Schoen, Michelle Doty, and Katie Tenney ABSTRACT: Many policymakers have called for the remodeling of Medicare to more closely resemble private insurance, which is often assumed to work better than public programs do. However, evidence from this 2001 survey demonstrates that Medicare beneficiaries are generally more satisfied with their health care than are persons under age sixty-five who are covered by private insurance. Medicare beneficiaries report fewer problems getting access to care, greater confidence about their access, and fewer instances of financial hardship as a result of medical bills. Making the program more like private insurance runs the risk of undermining a program that is working well from the perspective of beneficiaries. Medicare reform has been the focus of federal health policy since 97. 1 The National Bipartisan Commission on the Future of Medicare was established by the Balanced Budget Act (BBA) of 97, primarily because Medicare has been consuming an increasing share of the federal budget and the nation s economic resources, and those shares will rise markedly as the baby-boom generation reaches retirement. 2 Converting Medicare to a definedcontribution program, through Medicare+Choice (M+C), premium support, or vouchers, is viewed by some policymakers as a mechanism for controlling federal budget outlays and potentially achieving savings through financial incentives for beneficiaries and competition among private plans. 3 However, Medicare beneficiaries out-of-pocket health expenses will also rise markedly between 2000 and 202. 4 For those concerned with beneficiaries financial burdens and their access to needed care, Medicare reform is primarily about modernizing its benefit package, most notably adding a prescription drug benefit. The discussion over a Medicare prescription drug benefit has been stalled, in part, by a debate over the role of private insurance. A common theme running through the debate is that Medicare ought to be more like private insurance, which is depicted as being effective in controlling costs and providing beneficiaries with the kinds of health care choices they might Karen Davis is president of the Commonwealth Fund in New York City. Cathy Schoen is its vice-president for health policy, research, and evaluation; Michelle Doty, a senior analyst; and Katie Tenney, special assistant to the president. HEALTH AFFAIRS ~ Web Exclusive W311 2002 Project HOPE The People-to-People Health Foundation, Inc.

DataWatch desire. In the mid-90s, when private health insurance spending was increasing at a slower rate than that of Medicare spending, the short-term trends were cited as evidence that Medicare could achieve savings by adopting private managed care techniques. In the late 90s, however, Medicare spending grew at a slower rate than did private insurance spending. Long-term trend analysis reveals that Medicare and private outlays tend to grow at similar rates over time. 6 In the debate over Medicare costs and reform, few have considered how well Medicare works for its beneficiaries compared with persons under age sixty-five who are covered by private health insurance. Surveys have found a higher degree of satisfaction with coverage among Medicare beneficiaries compared with privately insured working families, better care experiences for persons newly covered by Medicare (ages 6 70) than for persons not yet covered (ages 0 64), and better ratings of managed care plans by Medicare enrollees than by those covered under employer-sponsored managed care plans. 7 In this paper we present new evidence on how well Medicare works for beneficiaries compared with how private insurance works for persons under age sixtyfive. Based on a 2001 survey of adults age nineteen and older, the analysis explores how the experiences of Medicare beneficiaries compare with the experiences of persons under age sixty-five who are insured through private employer-sponsored insurance in achieving two key goals of insurance: assuring that those covered are able to obtain health care services when needed, and assuring that they are protected against financial hardship from medical bills. We also compare how Medicare beneficiaries and those with private employer-sponsored coverage rate their coverage and their confidence about being able to obtain care in the future. Data And Methods The analysis is based on data from the Commonwealth Fund s 2001 Survey of Health Insurance, which was conducted by Princeton Survey Research Associates from April through 29 July 2001. The survey consisted of twenty-five-minute telephone interviews with a random national sample of 3,0 adults, age nineteen and older, living in households with telephones in the continental United States. The interviews included 2,29 adults ages 64 and 62 adults age 6 and older. The analysis drops from the sample fifty-one respondents who did not provide an age, yielding a study sample of 3,47 persons age nineteen and older. The study oversampled adults from low-income communities to yield a final sample with disproportionately large numbers of low- and moderate-income and minority households. The final sample is weighted to the adult population by sex, age, race/ethnicity, education, region, telephone service interruption, and household size, using the U.S. Census Bureau s March 2000 Current Population Survey (CPS). The resulting sample is representative of the 6 million adults living in the continental United States. Of those contacted for interviews, 69 percent agreed to participate. Counting eligible adults who were not reached by phone, despite nu- W3 9 October 2002

Medicare merous attempts, the overall survey response rate was 4 percent. The response rate is typical of most recent random-digit-dialed telephone surveys. Our exhibits compare experiences among Medicare beneficiaries age sixty-five and older; disabled Medicare beneficiaries; Medicaid beneficiaries; and adults under age sixty-five who are covered by private insurance (91 percent are employer-based plans, 9 percent individual), with subanalysis in logit regression results that differentiate between employer-based coverage, individual coverage, and the uninsured. In the analyses, persons with more than one source of coverage were assigned hierarchically to the Medicare, Medicaid, employer, and individual insurance categories, so that Medicare beneficiaries with supplemental coverage such as Medicaid, retiree coverage, or Medigap coverage are categorized as being Medicare enrollees. The discussion focuses primarily on comparisons between Medicare elderly and nonelderly adults who are insured through employer-group plans. Logit regression results for the Medicare disabled and individually insured are also presented but represent small samples for these groups (n = 7 and 7, respectively); caution must be exercised in interpreting these results, given the small sample sizes. Results for Medicaid beneficiaries and the uninsured are presented as well, where applicable. We initially present responses to questions about access, cost, and care experiences by insurance group, regardless of differences in income or health status. However, since different health care experiences are likely to be a function of varying needs for health care and ability to pay for care, we also present summary results of regression analyses (odds ratios from logit analyses) that take into consideration the effects of health and income. These exhibits compare the relative odds of reporting a problem or experience by insurance group (with adults under age sixty-five who are insured through employer plans as the reference group), taking into account the effects of health status, poverty, and other factors, including presence or absence of a prescription drug benefit. Exhibits indicate where findings and odds ratios are significant at the p <.0, p <.01, or p <.001 levels. Study Results Demographics. Medicare beneficiaries are more likely than the privately insured are to be in poor health and have low incomes. In the survey, two-thirds of persons under age sixty-five with private health insurance rated their health status as excellent or very good, compared with two-fifths of elderly Medicare beneficiaries (Exhibit 1). The proportion of elderly Medicare beneficiaries rating their health as fair or poor was three times higher than that of privately insured adults. Four of five Medicare beneficiaries had a chronic condition, compared with just over one-third of the privately insured. Medicare beneficiaries were four times as likely as the privately insured were to report having two or more chronic conditions. Medicare beneficiaries were more than twice as likely as the privately insured HEALTH AFFAIRS ~ Web Exclusive W313

DataWatch EXHIBIT 1 Characteristics Of The Study Population, By Insurance Status, 2001 Total Private insurance (ages 64) Medicare elderly (age 6 and older) Medicare disabled (under age 6) Medicaid Uninsured Estimated millions of adults Unweighted sample size 3.7 3,47 1.4 1,7 30.2 4 7. 7 9. 21.1 466 Sex Male Female 47% 3 49% 1 41% 9 42% 2% 72 4% 46 Race/ethnicity White Black Hispanic 73 77 11 3 9 6 1 23 6 11 26 Self-rated health status Excellent or very good Good Fair or poor 7 2 6 43 33 29 4 39 29 32 42 3 23 Chronic conditions None One Two or more 4 23 64 14 62 43 2 29 63 Annual income <$20,000 $20,000 $34,999 $3,000 $9,999 $60,000+ 26 2 34 2 31 0 17 4 74 2 6 2 Poverty status < 200% 200% or more 64 20 0 4 2 63 37 73 Adult work status Full time Part time Not currently employed 4 72 11 17 4 91 3 7 13 60 3 11 3 SOURCE: Commonwealth Fund 2001 Health Insurance Survey. were to have incomes below 200 percent of the federal poverty level. Four-fifths of privately insured persons were employed either full or part time, compared with only one in ten Medicare beneficiaries. Satisfaction, access, and cost experiences. Levels of satisfaction with and access to health care also vary by source of coverage, as do experiences with health care costs. Elderly Medicare beneficiaries were more likely to rate their health insurance as excellent and less likely to have negative experiences with their insurance than were the privately insured (Exhibit 2). Sixty-two percent of elderly Medicare beneficiaries reported being very satisfied with their care, compared with 1 percent of those covered by private insurance. Elderly Medicare beneficiaries were also more confident in their future ability to get care than were the privately insured. W314 9 October 2002

Medicare EXHIBIT 2 Experiences With Insurance And With Care, By Insurance Status, 2001 Experiences with insurance Total Private insurance (ages 64) Medicare elderly (age 6 and older) Medicare disabled (under age 6) Medicaid Uninsured Rating of insurance Excellent Very good Good Only fair Poor 23% 32 26 13 6 20% 34 13 6 32% 34 11 3 21% 23 13 % 21 23 7 Negative insurance experiences Plan did not pay anything for care respondent thought was covered Plan covered only a part of service Reached limit on what plan paid for specific illness/injury 21 1 6 11 9 37 7 21 31 32 Any one of the above three experiences Paid a lot out of pocket for Rx or dental Had difficulty getting referral to specialist 7 9 61 37 9 43 31 2 64 2 1 33 Experiences with care Satisfaction with care Very satisfied Somewhat satisfied Somewhat dissatisfied Very dissatisfied 1% 7 6 1% 3 4 62% 29 4 46% 33 6 49% 3 33% 39 17 Rating of physician Excellent Very good Good Fair/poor 34 37 21 6 39 40 44 26 2 23 3 Confidence in future ability to get care Very confident Somewhat confident Not too confident Not at all confident 3 40 37 44 13 6 0 31 7 26 39 3 26 Went without needed care in past year due to costs Did not fill prescription Did not get needed specialist care Skipped recommended test or follow-up 17 13 14 7 9 14 4 7 32 20 2 20 31 2 34 Had a medical problem, did not visit doctor or clinic Any one of above access problems Did not go to the dentist 21 11 42 32 21 41 43 46 HEALTH AFFAIRS ~ Web Exclusive W3

DataWatch EXHIBIT 2 Experiences With Insurance And With Care, By Insurance Status, 2001 (cont.) Experiences with care Total Private insurance (ages 64) Medicare elderly (age 6 and older) Medicare disabled (under age 6) Medicaid Uninsured Medical bill problems in past year Not able to pay bills Contacted by a collection agency for bills % % % 39% 30% 29 49% 3 Had to change way of life to pay bills Any one of above bill problems 13 29 9 2 33 4 4 Total out-of-pocket costs >$00 Total out-of-pocket costs >% of income 3 39 47 29 43 40 26 SOURCE: Commonwealth Fund 2001 Health Insurance Survey. a Not applicable. In spite of having poorer health and lower incomes than the privately insured had, elderly Medicare beneficiaries were somewhat less likely to report access problems resulting from costs, such as not getting needed specialist care or having a medical problem but not visiting a doctor or clinic. They were also less likely to report having problems paying their medical bills, such as not being able to pay, being contacted by a collection agency, or having to change their way of life to pay medical bills. However, they were more likely than the privately insured were to pay more than $00 a year out of pocket for health care and to devote more than percent of their income to medical expenses. Interestingly, 4 percent of Medicaid beneficiaries reported a medical bill problem. This might reflect the fact that 32 percent of Medicaid beneficiaries were covered for less than a year (not shown), so that bills might have been incurred while they were uninsured, or Medicaid beneficiaries might have paid out of pocket when seeing nonparticipating providers or after reaching limits on covered services. The observed differences between the health care experiences of Medicare beneficiaries and the privately insured are likely to reflect underlying differences in health status and income as well as the presence or absence of insurance benefits such as prescription drug coverage. Exhibits 3, 4, and present the odds of reporting problems with insurance, experiences with quality of care, access to care, or financial burdens by Medicare beneficiaries relative to those with employer coverage, when income, health status, and drug coverage are taken into account. Satisfaction with insurance. After differences in income, health status, and drug coverage were accounted for, respondents insured through the two main public insurance programs elderly Medicare and Medicaid beneficiaries were found W3 9 October 2002

Medicare to be more satisfied with their insurance than were those with employer coverage (Exhibit 3). Elderly Medicare beneficiaries were 2.7 times more likely than those with employer coverage were to rate their health insurance plan as excellent, and Medicaid beneficiaries were 2.1 times more likely than those with employer coveragetodoso. Elderly Medicare beneficiaries were only one-third as likely as those with employer coverage were to report negative experiences with their coverage. Negative experiences included finding out that the plan did not pay for a given medical service, that it paid for only part of a medical bill, or that the limit of what the plan would pay for a specific illness or injury had been reached. Forty-three percent of elderly Medicare beneficiaries reported any of these negative experiences, compared with 61 percent of privately insured adults (Exhibit 2). Elderly Medicare beneficiaries were found to be less than half as likely as those with employer coverage were to report that they paid a lot out of pocket for pre- EXHIBIT 3 Effect Of Insurance Status On Experiences With Insurance Plans, Expressed As Odds Ratio, Based On Logit Models, 2001 Variable Excellent rating on insurance Negative plan experiences a Paid a lot out of pocket for drugs or dental services Insurance type (ref = employer-sponsored insurance, age 64) Medicare age 6+ Medicare disabled Medicaid Individual (ages 64) 2.66*** 1.9 2.09*** 0.76 0.32**** 0.67 0.4*** 0.3 0.43**** 0.99 0.9** 1. No prescription drug coverage (ref = prescription coverage) 0.7*** 1.2 2.**** Poverty status (ref = 200% of poverty or more) Less than 0% 0 9% 0.63** 0.73** 0.97 0.1 0.9 1.00 Health status (ref = excellent) Very good Good Fair Poor 0.7**** 0.4**** 0.40**** 0.42** 1.** 1.43** 1.44 1.91** 1.37** 1.71**** 1.91*** 2.32*** Chronic conditions (ref = no chronic) One Two or more 1.41** 1.17 1.47*** 2.**** 1.23 1.92**** SOURCE: Commonwealth Fund 2001 Health Insurance Survey. NOTE: Comparison group: Individual age 64 with employer-sponsored insurance, prescription coverage, income 200 percent of poverty or above, reporting excellent health status, with no chronic conditions. a Experienced one or more of the following: plan didn t pay anything or only paid part of bill, or reached limit of what plan would pay. **p <.0 ***p <.01 ****p <.001 HEALTH AFFAIRS ~ Web Exclusive W317

DataWatch scriptions or dental services. Medicaid, which typically includes drug and dental coverage with little or no patient cost sharing, was also more effective than employer coverage was in protecting beneficiaries from out-of-pocket costs. The analysis indicates that having a prescription drug benefit, across all age groups, is associated with more positive overall ratings of coverage as well as cost experiences. As illustrated in Exhibit 3, having insurance with drug benefits increases the likelihood of rating coverage as excellent and reduces the risk of large out-of-pocket expenses. These results may be a direct reflection of drug benefits or may reflect the fact that plans with such benefits tend to provide more comprehensive benefits in general. As reported in other studies, the survey finds that respondents income and health status affect their levels of satisfaction with coverage. 9 Across all insurance sources, the poor and near-poor were less likely than adults with incomes at or above 200 percent of poverty were to rate their coverage as excellent (in 2002 the federal poverty level for a single person in the contiguous United States was $,60). Respondents with self-rated health status that was less than excellent were also significantly (p <.0, p <.001) less likely to give their insurance an excellent rating. Those with two or more chronic conditions were about twice as likely as others were to report negative experiences with their health insurance plan and to report paying a lot out of pocket for prescription drugs and dental services. Quality of care. Elderly Medicare beneficiaries were more likely than those with employer coverage were to report being very satisfied with their care, more likely to rate their doctor as excellent, and more likely to be very confident in their ability to get care in the future (Exhibit 4). Having any source of insurance was better than being uninsured on all quality-of-care measures. The uninsured were systematically less likely to be very satisfied with the overall quality of care, to rate their physician as excellent, or to be very confident of their ability to get care in the future. Employer coverage clearly improves patient ratings of care, compared with having no coverage at all. On five of the six measures of insurance rating and quality of care, the ratings by adults with individual (nongroup) insurance coverage were similar (no significant differences) to those by adults with employer coverage. However, given the small number of individually insured adults in the survey sample, caution should be exercised in interpreting this result. Medicare beneficiaries generally have very stable coverage. Ninety-nine percent of them had been insured the entire year when surveyed, compared with 92 percent of persons with employer coverage (not shown). However, even for continuously insured employed persons, employers change the plans that are available during annual enrollment periods. Changing plans often means changing physicians, especially for the privately insured under age sixty-five, who are much more likely than Medicare beneficiaries are to be enrolled in managed care plans. This instability can lead to disruptions in continuity of care. Access to care and financial problems. Based on reports of access and bill W3 9 October 2002

Medicare EXHIBIT 4 Effect Of Insurance Status On Experiences With Quality Of Care, Expressed As Odds Ratio, Based On Logit Models, 2001 Variable Very satisfied with overall quality of care Excellent rating for quality of care from physician Very confident in future ability to get care Insurance type (ref = employer-sponsored insurance, age 64) Medicare age 6+ Medicare disabled Medicaid Individual (ages 64) Uninsured 1.79**** 1.0 1.0 0.9 0.49**** 1.37** 1.6** 1.37 1.6** 0.4**** 2.04**** 1.3 0.6 0.97 0.3**** No prescription drug coverage (ref = prescription coverage) 0.94 0.71** 0.0 Poverty status (ref = 200% of poverty or more) Less than 0% 0 9% 0.96 1.0 0.66** 1.01 0.9 0.9 Health status (ref = excellent) Very good Good Fair Poor 0.9**** 0.44**** 0.44**** 0.41*** 0.49**** 0.40**** 0.6*** 0.61 0.9**** 0.37**** 0.41**** 0.40*** Chronic conditions (ref = no chronic) One Two or more 1.26 1.03 1.72**** 1.26 1. 1. SOURCE: Commonwealth Fund 2001 Health Insurance Survey. NOTE: Comparison group: Individual age 64 with employer-sponsored insurance, prescription coverage, income 200 percent of poverty or above, reporting excellent health status, with no chronic conditions. **p <.0 ***p <.01 ****p <.001 problems, after health status and income are controlled for, Medicare successfully fulfills the two main purposes of health insurance: assuring access to needed health care services and preventing financial burdens from medical bills. Elderly Medicare beneficiaries were one-third as likely as those with employer coverage were to experience access problems because of cost, after income, health status, and drug coverage were adjusted for (Exhibit.) Access problems include avoiding filling a prescription, not getting needed specialist care, skipping recommended tests or follow-up care, or having medical problems but not visiting a doctor or clinic because of cost. Among the insurance groups, the uninsured were generally most at risk for access problems: They were 3.6 times as likely as those with employer coverage were to experience one or more access problem as a result of cost. These findings underscore how well employer coverage performs in improving access to care relative to having no coverage at all. Persons with low incomes, in poor health, or without prescription drug coverage reported more problems getting access to health care HEALTH AFFAIRS ~ Web Exclusive W3

DataWatch EXHIBIT Effect Of Insurance Status On Access Problems And Financial Burdens Due To Medical Bills, Expressed As Odds Ratios, Based On Logit Models, 2001 Variable Any access problems due to cost a Any bill problems b Spent more than $00 out of pocket Spent % or more of income out of pocket Insurance type (ref = employer-sponsored insurance, age 64) Medicare age 6+ Medicare disabled Medicaid Individual (ages 64) Uninsured 0.33**** 0.9 1.42 1.23 3.64**** 0.2**** 0.6 1.03 0.1 2.63**** 0.2 0.73 0.**** 2.0*** 1. 1. 1.47 0.1*** 2.6**** 1.93*** No prescription drug coverage (ref = prescription coverage) 1.7**** 1.30 1.33** 2.32**** Poverty status (ref = 200% of poverty or more) Less than 0% 0 9% 1. 1.3**** 2.06**** 2.47**** 0.**** 0.63*** 3.46**** 3.41**** Health status (ref = excellent) Very good Good Fair Poor 1.62*** 2.2**** 2.06***.11**** 1.6*** 2.**** 2.72**** 4.02**** 1.3** 1.77**** 1.62** 2.44*** 1.14 1.47 1.39 2.69*** Chronic conditions (ref = no chronic) One Two or more 1.7**** 2.20**** 2.00**** 2.73**** 2.40**** 3.7**** 1.91**** 4.0**** SOURCE: Commonwealth Fund 2001 Health Insurance Survey. NOTE: Comparison group: Individual age 64 with employer-sponsored insurance, prescription coverage, income 200 percent of poverty or above, reporting excellent health status, with no chronic conditions. a Access problems due to cost include the following: did not fill prescription, did not get needed specialist care, skipped recommended test or follow-up, had medical problem but did not visit doctor or clinic. b Bill problems include the following: not able to pay medical bills, contacted by a collection agency for medical bills, or had to change way of life to pay bills. **p <.0 ***p <.01 ****p <.001 than others reported. Medicare provides better coverage against financial hardship than employer coverage provides, after income and health status are controlled for. Elderly Medicare beneficiaries were one-fourth as likely as those with employer coverage were to report problems with medical bills, including not being able to pay, being contacted by a collection agency, or having to change their way of life to pay medical bills. The uninsured, by contrast, were 2.6 times as likely as those with employer coverage were to report a medical bill problem. Lower income and poorer health status also significantly (p <.01, p <.001) increased the likelihood of experiencing medical bill problems. Medicare beneficiaries and those with employer coverage reported similar ex- W320 9 October 2002

Medicare periences of high medical costs, defined in this analysis as out-of-pocket expenses greater than $00 a year or totaling percent or more of income. Medicaid beneficiaries, by contrast, were much less likely than those with employer coverage were to incur substantial out-of-pocket costs. Those without prescription drug coverage were more likely to experience high out-of-pocket costs than were those with such coverage. Lower income levels and chronic conditions were also associated with higher out-of-pocket costs. Income, health status, and insurance ratings. Medicare was systematically more likely than employer coverage was to be rated as excellent across all income and health status categories (Exhibit 6). For example, 43 percent of healthy elderly Medicare beneficiaries with incomes greater than twice the poverty level rated their insurance as excellent, compared with percent of persons with employer coverage at comparable income and health-status levels. While only 26 percent of elderly Medicare beneficiaries with low incomes and poor health rated their insurance as excellent, those with employer coverage at similar income and health-status levels were even less likely to rate their insurance as excellent ( percent). 11 Discussion And Policy Implications This survey s findings that Medicare beneficiaries are generally more satisfied with their health care than the privately insured are, are more confident about their access to care, and for key services experience fewer access problems challenge the accepted notion that Medicare is out of date and should catch up with the private insurance model. Medicare beneficiaries have less comprehensive EXHIBIT 6 Predicted Rating Of Health Insurance Coverage, By Health, Poverty, And Insurance Status, 2001 Percent rating coverage as excellent 0 Medicare, age 6+ Employer coverage, ages 64 40 30 20 0 Lower income, sick Lower income, healthy Higher income, sick Higher income, healthy SOURCE: Commonweath Fund 2001 Health Insurance Survey. NOTES: Sick denotes persons in good/fair/poor health status with an average number of chronic conditions for this group. Healthy denotes excellent/very good health status with average number of chronic conditions for this group. Lower-income denotes income of 200 percent of the federal poverty level or lower; higher-income, more than 200 percent. HEALTH AFFAIRS ~ Web Exclusive W321

DataWatch benefits and often pay higher out-of-pocket premiums than those covered by employer plans pay, and their Part B premiums exceed premiums paid directly by employees for employer coverage. Nevertheless, elderly Medicare beneficiaries are less likely than persons with employer coverage are to report negative insurance experiences or problems paying medical bills. This may reflect greater denial of claims under employer coverage or the lower continuity of coverage under employer plans that periodically leaves some persons without insurance. Medicare beneficiaries more positive access experiences and ratings of their care indicate that their coverage is working relatively well in providing choice of services and access to needed care. This may be due to the fact that most Medicare beneficiaries are covered under the traditional fee-for-service program, while adults with employer coverage are more likely to be enrolled in managed care plans. As a result, Medicare beneficiaries have a wider choice of physicians and fewer restrictions on care, such as the prior approval for specialist services required by many managed care plans. In markets with M+C options, Medicare beneficiaries might also have more plan choices than persons with employer coverage have. Only about 41 percent of employees have a choice of employer-sponsored insurance plans, and fee-forservice options are not common. 13 A number of studies have found that consumer choice leads to higher satisfaction with coverage. 14 The more positive access and insurance satisfaction findings from Medicare beneficiaries might also reflect the stability and relative simplicity of coverage under Medicare. Except for the minority in M+C plans, the core set of benefits and insurance rules as well as coverage have remained constant over time. Private coverage varies widely. It was not possible to contrast the experiences of persons covered under large employer plans with those in the small-business market. The financial burdens of those with individual insurance point to the limited benefits available in that market, which may also be true of some employer plans. It was also not possible, given the sample size, to contrast the experiences of those with Medicare managed care coverage with those in Medicare s traditional fee-for-service plan. Proposals to privatize Medicare might be able to avoid some of the disadvantages of private coverage by keeping central features of social insurance guaranteed coverage regardless of health status, defined benefits, and multiple choice of plans and by retaining Medicare s fee-for-service option. In spite of these positive findings for Medicare for seniors, the need for a Medicare prescription drug benefit remains clear. For those without such a benefit, ratings of insurance are lower, access problems are greater, and out-of-pocket costs for all medical and dental care and for prescription drugs are higher. There are too few disabled (nonelderly) Medicare beneficiaries in this survey to draw conclusions about this group, although findings indicate that this group is at risk. A recent study of the Medicare disabled also finds that this group is vulnerable because of the combined effects of low income, poor health, and coverage gaps, W3 9 October 2002

Medicare with consequent needs for comprehensive benefits. Further investigation of this group s experiences is warranted. In the policy debates over the future of Medicare, it is important to listen to the experiences of individuals, whether covered by Medicare or by private insurance. Apart from the lack of a prescription drug benefit, Medicare is reportedly working better for its beneficiaries than is the employer-group coverage available to most persons under age sixty-five. The greater confidence in getting care when needed and lower incidence of access problems reported by Medicare beneficiaries are notable, given that Medicare beneficiaries are disproportionately sicker and poorer than the privately insured are. The survey also raises questions about how well private coverage is protecting persons at greater risk because of lower incomes or poor health or both. Current trends to increase cost sharing or shift responsibility for health coverage to individuals in the form of defined-contribution plans or personal health accounts may make low-income and chronically ill adults more vulnerable. Increased cost sharing as part of Medicare reform may also be particularly problematic for chronically ill Medicare beneficiaries. Medicare provides a level of security not typically found in employer or individual coverage. Its beneficiaries are assured that they will not lose their coverage, while coverage for persons under age sixty-five can vary with employment status, employers decisions to change plans, or even the onset of a serious illness. Medicare beneficiaries also live in a world with more stable benefits and, for most, fewer complex insurance arrangements than the privately insured have. Thus, attempts to reform Medicare that would pattern coverage on private employer coverage run the risk of undermining the confidence of the people it is designed to serve. The authors thank the three anonymous reviewers for their helpful comments. The views expressed are those of the authors and should not be attributed to the Commonwealth Fund, its directors, or its officers. NOTES 1. Reform proposals include converting Medicare to a program like the Federal Employees Health Benefits Program (FEHBP); converting Medicare to a premium-support program; encouraging enrollment in managed care plans; and moving to competitive pricing or value-based purchasing under Medicare+Choice. Incremental changes to the fee-for-service program include improving benefits, tightening provider payments, and setting budget caps on Medicare. See S.M. Butler, The FEHBP as a Model for a New Medicare Program, Health Affairs (Winter 9): 47 61; H.P. Cain II, Moving Medicare to the FEHBP Model, or How to Make an Elephant Fly, Health Affairs (July/Aug 99): 2 39; H.J. Aaron and R.D. Reischauer, The Medicare Reform Debate: What Is the Next Step? Health Affairs (Winter 9): 30; R.A. Berenson, Medicare+Choice: Doubling or Disappearing? 2 November 2001, www.healthaffairs.org/web Exclusives/Berenson_Web_Excl_01.htm (6 September 2002); M. Moon and K. Davis, Preserving and Strengthening Medicare, Health Affairs (Winter 9): 31 46; J. Oberlander, Is Premium Support the Right Medicine for Medicare? Health Affairs (Sep/Oct 2000): 4 99; and K. Thorpe and A. Atherly, Reforming Medicare: Impacts on Federal Spending and Choice of Health Plans, October 2001, www. healthaffairs.org/webexclusives/thorpe_web_excl_01.htm (6 September 2002). 2. C. Kahn and H. Kuttner, Budget Bills and Medicare Policy: The Politics of the BBA, Health Affairs (Jan/Feb HEALTH AFFAIRS ~ Web Exclusive W323

DataWatch 99): 37 47. 3. National Bipartisan Commission on the Future of Medicare, Final Proposal: Breaux-Thomas Premium Support Model for Medicare (Washington: Bipartisan Commission, March 99). 4. M. Moon, Growth in Medicare Spending: What Will Beneficiaries Pay? (New York: Commonwealth Fund, May 99); M. Moon, Restructuring Medicare: Impacts on Beneficiaries (New York: Commonwealth Fund, May 99); and S. Maxwell, M. Moon, and M. Segal, Growth in Medicare and Out-of-Pocket Spending: Impact on Vulnerable Beneficiaries (New York: Commonwealth Fund, January 2001).. P. Neuman et al., Understanding the Diverse Needs of the Medicare Population: Implications for Medicare Reform, Journal of Aging and Social Policy, no. 4 (99): 2 0; M. Davis et al., Prescription Drug Coverage, Utilization, and Spending among Medicare Beneficiaries, HealthAffairs (Jan/Feb 99): 231 3; J. Poisal and L. Murray, Growing Differences between Medicare Beneficiaries With and Without Drug Coverage, Health Affairs (Mar/Apr 2001): 74 ; J. Blustein, Drug Coverage and Drug Purchases by Medicare Beneficiaries with Hypertension, Health Affairs (Mar/Apr 2000): 2 230; M. Moon, Will the Care Be There? Vulnerable Beneficiaries and Medicare Reform, Health Affairs (Jan/Feb 99): 7 117; and C. Cassel, R. Besdine, and L. Siegel, Restructuring Medicare for the Next Century: What Will Beneficiaries Really Need? Health Affairs (Jan/Feb 99): 1 131. 6. C. Boccuti and M. Moon, Comparing Medicare and Private Insurance: Growth Rates in Spending for Health Care over Thirty Years, Draft Report (New York: Commonwealth Fund, May 2002); and M. Moon, Beneath the Averages: An Analysis of Medicare and Private Expenditures (Washington: Henry J. Kaiser Family Foundation, September 99). 7. C. Hoffman et al., Gaps in Health Coverage among Working-Age Americans and the Consequences, Journal of Health Care for the Poor and Underserved, no. 3 (2001): 46 ; C. Schoen et al., Counting on Medicare: PerspectivesandConcernsofAmericansAgesFiftytoSeventy (New York: Commonwealth Fund, July 2000); and H. Taylor, Most People Continue to Think Well of Their Health Plans, Harris Interactive Health Care News 2, no. 3 (2002): 1 4.. Reflecting coverage patterns for the under-sixty-five population, all but 9 percent of persons with private coverage in the survey were covered through the employer-group market. 9. B. Druss et al., Chronic Illness and Plan Satisfaction under Managed Care, Health Affairs (Jan/Feb 2000): 203 209; and K. Davis et al., Choice Matters: Enrollees Views of Their Health Plans, Health Affairs (Summer 9): 99 1.. Only percent of Medicare beneficiaries were enrolled in managed care plans in 2000. Among persons with employer-based coverage, 26 percent were enrolled in HMO plans and 2 percent in PPO plans in 2002. J. Gabel et al., Job-Based Health Benefits in 2002: Some Important Trends, Health Affairs (Sep/Oct 2002): 143 1; and M. Gold and L. Achman, Trends in Premiums, Cost-Sharing, and Benefits in Medicare+Choice Health Plans, 99 2001 (New York: Commonwealth Fund, April 2001). 11. Healthy is defined as being in excellent or very good health status, while higher income is defined as having income of 200 percent of the federal poverty level or greater. Predicted probabilities in each category are based on appropriate values for chronic conditions for each income and self-assessed health status group, while holding prescription drug coverage constant.. L. Levitt, E. Holve, and J. Wang, Employer Health Benefits, 2001 Annual Survey (Washington: Henry J. Kaiser Family Foundation and Health Research and Educational Trust, 2001); and K. Davis, Medicare s Cost Sharing: Implications for Beneficiaries, Testimony before the House Ways and Means Committee, 9 May 2001. 13. B.S. Schone and P.F. Cooper, Assessing the Impact of Health Plan Choice, Health Affairs (Jan/Feb 2001): 267. 14. See, for example, R. Miller and H. Luft, HMO Plan Performance Update: An Analysis of the Literature, 97 2001, Health Affairs (July/Aug 2002): 63 6; T. Rice et al., Workers and Their Health Plans: Free to Choose? Health Affairs (Jan/Feb 2002): 2 7; A. Gawande et al., Does Dissatisfaction with Health Plans Stem from Having No Choices? HealthAffairs (Sept/Oct 9): 4 4; and Davis et al., Choice Matters.. Neuman et al., Understanding the Diverse Needs of the Medicare Population ; Davis et al., Prescription Drug Coverage, Utilization, and Spending ; and Poisal and Murray, Growing Differences between Medicare Beneficiaries With and Without Drug Coverage.. B. Briesacher et al., Medicare s Disabled Beneficiaries: The Forgotten Population in the Debate over Drug Benefits (New York: Commonwealth Fund, September 2002). W3 9 October 2002