By Gina Livermore, David C. Stapleton, and Meghan O'Toole

Transcription

1 doi: /hlthaff HEALTH AFFAIRS 30, NO. 9 (2011): Project HOPE The People-to-People Health Foundation, Inc. By Gina Livermore, David C. Stapleton, and Meghan O'Toole Health Care Costs Are A Key Driver Of Growth In Federal And State Assistance To Working-Age People With Disabilities Gina Livermore (glivermore@ mathematica-mpr.com) is a senior researcher at Mathematica Policy Research, in Washington, D.C. David C. Stapleton is a senior fellow at Mathematica Policy Research and director of Mathematica s Centerfor Studying Disability Policy. Meghan O Toole is a strategy analyst at City Year, in Boston, Massachusetts. ABSTRACT A large and rapidly growing share of US government pays for assistance to working-age people with disabilities. In 2008 federal spending for disability assistance totaled $357 billion, representing 12 percent of all federal outlays. The states share of joint federal-state disability programs, more than 90 percent of it for Medicaid, was $71 billion. The increased cost of health care which represented 55 percent of combined state and federal outlays for this population in 2008 is one of the two main causes of spending growth for people with disabilities. Health care is already likely to be a target of further efforts by states and the federal government to contain or reduce spending, and it is therefore probable that spending restraints will affect the working-age population with disabilities. In fact, unless ways can be identified to make delivery of health care to this population more efficient, policy makers may be unable to avoid funding cuts that will further compromise its well-being. Government spending for workingage adults ages with disabilities is of interest to the policy community for several reasons. The first reason is the sheer size of this population. The Census Bureau estimated that 6.4 percent of the working-age population residing in the community in 2008 had a disability about nineteen million people. 1,2 The second reason is that federal and state spending to support this population represents a large and growing share of all federal and state, which is likely to continue to grow at a very fast rate. That expectation reflects anticipated growth both in the cost of health care and in the number of working-age Americans with disabilities as a result of the aging baby boom generation. In fiscal year 2008, more than half of workingage people with disabilities were covered by public health insurance: Medicare, Medicaid, and Department of Veterans Affairs programs. 3 In fiscal year 2008 working-age Medicare beneficiaries with disabilities accounted for 17 percent of all Medicare enrollees, and working-age Medicaid beneficiaries with disabilities accounted for 15 percent of all Medicaid enrollees. 4,5 Medical advances are also contributing to growth in the size of this population, as they improve the chances of survival following injury and the onset of chronic diseases. 6 What s more, the implementation of the Affordable Care Act of 2010 will have important implications for working-age people with disabilities because of their reliance on publicly funded medical and disability services. A large share of working-age people with disabilities have low incomes and so will become eligible for Medicaid (if not already enrolled) and health insurance subsidies. Given the size and rapid growth of spending under current law, it will be very difficult for policy makers to avoid making the population of working-age people with disabilities the target of efforts to reduce federal deficits. Similarly, 1664 Health Affairs September :9

2 states are finding it increasingly difficult to avoid reducing their spending for this population. Without the current strong social commitment of support, however, the high poverty rates and material hardships experienced by people in this population 7,8 would be far worse than they are. A clear picture of current federal and state funding for health care services within the context of all federal and state funding for this population will help inform the choices that policy makers and administrators face as deficit-cutting pressures mount and the Affordable Care Act is implemented. In particular, policy makers will need to balance this population s health care needs and the need to slow the growth in federal and state health care spending. This will no doubt be a very difficult political and social choice to make. Hence, policy makers will probably seek opportunities to reduce health care spending growth while maintaining, or even improving, the delivery of publicly funded services. Such opportunities might lie within broader opportunities to improve the efficiency of support for workingage people with disabilities. In this article we present estimates of federal and state spending for the support of workingage people with disabilities in fiscal year We itemize government spending for this population that spans sixty-three federal or federalstate programs a subset of an even larger number of programs that provide some level of support for these adults. (Examples of these supports include income benefits as well as housing and food assistance.) We describe the methodology we used to make these estimates. We conclude with a discussion of the implications. Study Data And Methods We used published statistics, primarily derived from federal program administrative data, to develop the estimates of federal and state spending for working-age people with disabilities in fiscal year The online Appendix 9 contains specific details on the methodology we used to develop the estimates for each of the sixty-three federal and state programs considered. In developing the estimates, we followed the methods used in a recent study that provided analogous figures for fiscal year By following that study s methodology, we were able to produce comparable spending figures for fiscal year 2008 and to assess the growth since We included programs targeted to workingage people with disabilities, as well as programs commonly used by this group but not specifically designed for their benefit. For all programs, we estimated and included only the that represent direct services and cash benefits to working-age people with disabilities, such as health care, education and training services, and cash and in-kind assistance. Other, such as for research and administration, were excluded. In some cases, we could not replicate the methods used to develop the 2002 estimates; in others, we identified ways to improve on them. 11 The differences are noted in the Appendix. 9 For two reasons, we believe our estimates to be lower than the actual federal and state for working-age people with disabilities in 2008 and for programs for which published data were not available. First, we excluded certain types of, such as for administrative costs, for programs not providing direct services, and for programs for which published data were not available. Second, for health care associated with programs not targeted to people with disabilities (for example, Medicaid spending for participants with disabilities in the Temporary Assistance for Needy Families program and spending for veterans health care), we allocated the proportion of attributed to people with disabilities based solely on the proportion of program participants they represented, without making adjustments to reflect their generally higher per person health care costs. The information needed to make such adjustments was not available. It should be noted that the definition of disability varies from program to program. We used the program s own definition when eligibility was contingent upon applicants providing evidence of a disability. Such definitions generally are much more stringent than those found in national surveys, such as the National Health Interview Survey and surveys of the incarcerated population. For other programs, the definition of disability was typically based on the findings of national surveys. The surveys and underlying definitions differed across the specific studies that were used to develop the spending estimates. Study Results Federal Spending Our estimates indicate that federal spending for working-age adults with disabilities equaled $357 billion in fiscal year We grouped spending into five categories: income maintenance; health care; housing and food assistance; education, training, and employment; and other services (Exhibit 1). Health care and income maintenance each accounted for more than 47 percent of total. Health care included $62.9 billion for Medicare; $88.8 billion for Medicaid enroll- September :9 Health Affairs 1665

3 Exhibit 1 Federal Expenditures For Working-Age People With Disabilities, By Type Of Expenditure, Billions Of Dollars, Fiscal Year 2008 SOURCE Authors analysis. See the Appendix for details; see Note 9 in text. ees (including $78.4 billion for those in the disabled Medicaid category, and $10.3 billion for participants with disabilities in the Temporary Assistance to Needy Families program); $13.8 billion for veterans medical care and prosthetics; and miscellaneous smaller program. Income maintenance included $101.7 billion in Social Security Disability Insurance benefits for disabled workers and their dependents, $26.2 billion in federal Supplemental Security Income payments, $20.1 billion in Veterans Compensation and Disability Pensions, and miscellaneous smaller program. From fiscal year 2002 to fiscal year 2008, federal spending for working-age people with disabilities increased by 56 percent, or $129 billion (Exhibit 2), which reflected growth in the number of people served and in per person served. During the same period, both the gross domestic product and total federal outlays increased by approximately 35 percent. Federal spending for this population grew from 2.1 percent to 2.5 percent of gross domestic product, and from 11.4 percent to 12.0 percent of total federal outlays. Inflation-adjusted increased in all categories except education, training, and employment. Health care grew much more rapidly than any other component by 34 percent after adjusting for inflation.within health care, expenditure growth rates for Medicare, Medicaid, and veterans health benefits were all very high. Inflation-adjusted Medicare spending grew by 70.1 percent, including 16.6 percentage points that can be attributed to the introduction of Medicare Part D. 12 Part of the growth in Medicare spending can also be attributed to a 33.4 percent increase in the number of disabled Medicare beneficiaries. 4 The remainder of the inflation-adjusted increase (20.0 percentage points) must be attributed to an increase in cost per beneficiary. Growth in Medicare spending per disabled beneficiary is only slightly higher than growth in Medicare spending per elderly beneficiary; the latter increased by 18.9 percent after adjusting for inflation over the same period. 4 Growth per beneficiary in both Medicare populations exceeds growth in national health care spending per capita over approximately the same period; the latter grew by an estimated 15.3 percent from calendar year 2002 through calendar year Medicare spending for disabled enrollees increased substantially as a share of all Medicare from 15.4 percent in fiscal year 2002 to 17.9 percent in fiscal year 2008 Exhibit 2 Federal Expenditures For Working-Age People With Disabilities, By Major Expenditure Category, Fiscal Years 2002 And 2008 Category FY 2002 Percent change, FY 2002 Income maintenance 109, , Health care 105, , Housing and food assistance 8,252 11, Education, training, and employment 3,706 4, Other services 2,035 2, Total 228, , Percent change, FY 2002, adjusted for inflation SOURCE Authors analysis. See the Appendix for details; see Note 9 in text. NOTE Fiscal year (FY) 2002 dollar values were adjusted for inflation using the annual Consumer Price Index for all urban consumers Health Affairs September :9

4 primarily because of a 33.4 percent increase in the number of disabled beneficiaries. In contrast, the number of elderly beneficiaries grew by only 8.9 percent. Inflation-adjusted Medicaid spending for working-age disabled enrollees grew by 17.5 percent during this period. 14 Growth in the number of enrollees accounts for 15.0 percentage points of that amount, while growth in spending per enrollee accounts for the remaining 2.5 percentage points much slower than the inflationadjusted growth in cost per disabled Medicare enrollee or national health per capita. 5,15,16 Inflation-adjusted veterans health increased by 37.0 percent, reflecting rapid growth in the number of working-age veterans with disabilities. Inflation-adjusted growth in income support was also high (29.5 percent) with especially high growth rates for Social Security Disability Insurance benefits and veterans programs. This rate is lower than the corresponding figure for health care spending (34.4 percent); exclusive of Medicare Part D, inflationadjusted health also increased by 29.5 percent. Expenditures for housing and food assistance grew substantially after adjusting for inflation (18.0 percent). Nonetheless, this category remained a small percentage of total spending. State Spending For Federal-State Programs State governments also expend substantial sums under some federal-state programs most notably Medicaid to support working-age people with disabilities. Also provided are state supplements to Supplemental Security Income; vocational rehabilitation; and various education, training, and employment programs. We estimate that state for working-age people with disabilities under federal-state programs totaled $71 billion in fiscal year 2008 (Exhibit 3). More than 90 percent of these were for Medicaid. Inflation-adjusted state spending for the four programs increased by 16.2 percent from fiscal year 2002 to fiscal year This increase was predominantly driven by a 19.7 percent growth in inflation-adjusted state Medicaid spending, a rate comparable to the federal spending growth rate. State spending in every other category actually fell during the same period, after adjustment for inflation. Thus, during this period state support shifted substantially toward health care and away from income support and vocational rehabilitation. State disability accounted for 5.7 percent of all direct state outlays in fiscal year 2008 an increase from 5.6 percent in fiscal year Total Federal and State Spending Total combined federal and selected state spending for working-age people with disabilities in fiscal year 2008 were $429 billion (Exhibit 4). After adjustment for inflation, that figure represents a 28.2 percent increase from fiscal year Health care spending was the largest single component, accounting for 54.9 percent of all spending in fiscal year Health spending was also the fastest-growing component, increasing by 29.9 percent from fiscal year 2002 after adjustment for inflation. The second-largest category, income maintenance, accounted for 40.6 percent of all spending, and it too grew rapidly during this period by 27.7 percent after adjusting for inflation. Growth in both figures was driven by rapid Exhibit 3 State Expenditures For Working-Age People With Disabilities Under Selected Federal-State Programs, Fiscal Years 2002 And 2008 Program FY 2002 Percent change, FY 2002 Medicaid 46,330 66, Temporary Assistance for Needy Families 1,940 1, State supplements to Supplemental Security Income 2,248 2, Vocational rehabilitation Total 51,198 71, Percent change, FY 2002, adjusted for inflation SOURCE Authors analysis. See the Appendix for details; see Note 9 in text. NOTES Fiscal year (FY) 2002 dollar values were adjusted for inflation using the annual Consumer Price Index for all urban consumers. Our estimates do not include state and local for state-only programs that support this population. September :9 Health Affairs 1667

5 Exhibit 4 Combined Federal And Selected State Expenditures By Major Expenditure Category, Fiscal Years 2002 And 2008 Category FY 2002 Percent of FY 2002 Percent of Income maintenance 113, , Health care 151, , Housing and food assistance 8, , Education, training, and employment 4, , Other services 2, , Total 279, , Percent change, FY 2002, adjusted for inflation SOURCE Authors analysis.seetheappendixfordetails(seenote9intext).note Fiscal year (FY) 2002 dollar values were adjusted for inflation using the annual Consumer Price Index for all urban consumers. growth in the number of people supported. Growth in health care spending per capita was also driven by rapid growth in the cost of health care, and by the introduction of Medicare Part D. Exclusive of Part D, inflation-adjusted health spending increased by 26.5 percent. Inflation-adjusted spending for housing and food assistance and miscellaneous other services increased at much lower rates, while spending for education, training, and employment actually fell. Combined, state and federal disability accounted for 10.1 percent of all state and federal outlays in fiscal year 2008 an increase from 9.6 percent in fiscal year These accounted for 3.0 percent of gross domestic product in fiscal year 2008, up from 2.6 percent in fiscal year Per Capita Spending Although we focused on estimates of gross programmatic at the state and federal levels, we acknowledge that it would be highly desirable to be able to present two other types of estimates: per person and disaggregated into broad categories defined by the primary health conditions underlying disability. Per capita would provide another measure of the magnitude of the support for workingage people with disabilities, and by cause of disability might point to areas for future investments in prevention and other cost control measures. Unfortunately, it is not possible to develop either type of estimate with the data currently available on the number of working-age people with disabilities and the types of disabilities experienced by program participants. Administrative data on the number of people participating in each program are available for most programs, but there are few statistics on the numbers participating in two or more programs. And with the exception of Social Security disability programs, little information is available on the causes of disability among participants in public programs. Nevertheless, a sense of the level of per capita spending can be obtained by dividing total spending by the estimated nineteen million working-age people with disabilities in the United States. 1 This calculation shows that total federal and state spending was equivalent to $22,561 per person in fiscal year 2008; however, some unknown share of this population received no public program benefits. Approximate spending for income and health support per Social Security disability program participant can also be calculated, because we have an unduplicated count of Social Security Disability Insurance and Supplemental Security Income recipients, and we can identify almost all of their for income and health. There were 10.6 million Social Security Disability Insurance and Supplemental Security Income beneficiaries in December 2007 (the third month of fiscal year 2008), 18 and the combined income and health care for their Social Security Disability Insurance, Supplemental Security Income, and Medicare and Medicaid benefits was about $31,933 per person. 19 This excludes other public assistance received by many beneficiaries (for example, food and housing assistance and veterans benefits), but it includes some Medicaid spending for the small share of Medicaid enrollees with disabilities who are not receiving Social Security Disability Insurance or Supplemental Security Income. The comparable numbers in fiscal year 2002 were 8.8 million beneficiaries and $29,450 per person (adjusted for inflation). The increase in spending per beneficiary primarily reflects the 1668 Health Affairs September :9

6 growth in Medicare and Medicaid spending. Real wage growth, which is faster than price growth, also contributed to the growth in inflation-adjusted income benefits per capita. Discussion The findings have a number of implications, both in terms of the levels and growth of the themselves, and in terms of the potential effects of efforts to reduce them. There are also important factors likely to affect expenditure growth in the future. Effects On Other Disability Programs The fact that health care were the largest and fastest growing component of all for working-age people with disabilities and that for some types of services experienced no growth or even declined might suggest that the growth in health care is crowding out other supports for people with disabilities. We do not know if this is the case. Numerous factors affect the budgetary decisions of state and federal policy makers. But high health care costs and their rapid growth can certainly make it more difficult for states and the federal government to invest in other programs that support this population. Effects On Nondisability Programs Efforts to reduce spending for programs that serve broader groups of the working-age population can have consequences for those with disabilities. This is illustrated in the declines in spending for the Temporary Assistance for Needy Families program. State spending under this program for parents with disabilities fell by 32 percent during the period after adjustment for inflation. Inflation-adjusted federal for the same group fell by 20 percent (see Table 1A in the Appendix). 9 It is likely that this decline reflects intensified state efforts to reduce caseloads, including efforts to move participants with disabilities into the Social Security disability programs. The program s 2006 reauthorization pushed states in this direction, by intensifying the program s work requirements. Findings from other studies suggest considerable overlap and strong interactions between the Supplemental Security Income and Temporary Assistance for Needy Families programs. 20,21 Thus, reductions to nondisability programs, such as Temporary Assistance for Needy Families, Unemployment Insurance, and the Supplemental Nutrition Assistance program, are likely to affect many disabled, low-income participants. Prospects For Future Spending Growth Under current law, several factors will probably continue to fuel the growth of federal for working-age people with disabilities in the near future. First, the severe recession appears to have spurred a large increase in participation in the Social Security Disability Insurance program, a trend that has continued beyond fiscal year Disability awards increased by 10.2 percent in fiscal year 2009 and by another 7.9 percent in fiscal year Some people quickly became eligible for Medicaid, and those surviving for twenty-four months after becoming eligible for Social Security Disability Insurance became eligible for Medicare.We can also expect the aging of the baby-boom generation to contribute to growth in Social Security Disability Insurance awards for at least another decade. Second, for veterans programs are expected to continue to grow rapidly because of the rising number of disabled veterans from the wars in Iraq and Afghanistan, combined with aggressive government efforts to meet their needs. It seems likely that implementation of the Affordable Care Act will increase growth in government health for workingage adults with disabilities. However, there are many factors to consider, including whether provisions of the law will even be implemented and, if so, how. If implemented as currently envisioned, the Affordable Care Act is likely to increase Medicaid enrollment among those with disabilities who are not already enrolled but whose household incomes are below 133 percent of the federal poverty level. Many with higher incomes are likely to receive subsidies to purchase insurance from a health insurance exchange. It is also possible that the Affordable Care Act will affect workers entry into Social Security Disability Insurance, and therefore into Medicare, but the hypothesized direction of the effect is ambiguous. Better access to health insurance without Medicare might help workers with serious medical conditions return to work instead of seeking Social Security benefits. However, availability of public support for health insurance during the twenty-four-month Medicare waiting period will reduce the cost of applying for those benefits. Greater availability of Medicaid to low-income people not enrolled in Supplemental Security Income or Temporary Assistance for Needy Families could reduce use of these income support programs by those with disabilities. The Community Living Assistance Services and Supports (CLASS) Act is an Affordable Care Act program that, if implemented, would provide payments for personal assistance services September :9 Health Affairs 1669

7 and other disability-related services for those who have paid voluntary premiums. If the program is self-financing, as planned, it could reduce federal and state to support working-age adults with disabilities. Some beneficiaries might rely less on Medicaid or other programs to pay for such services, and better access to such services might help some stay in the labor force longer than otherwise. Whether the program can be self-financing will depend on how willing workers are to pay the premiums instead of opting out. Policy Implications The sheer size of federal for working-age people with disabilities and the large number of programs that serve them reflect a strong social commitment to providing support for this population. Yet these represent such a large and growing share of all federal outlays that any serious effort to control federal spending must consider limiting the growth in spending for this population. Federal health care, which are growing rapidly for all covered groups, are particularly likely to be targeted, as evidenced in the report of the National Commission on Fiscal Responsibility and Reform. 23 A specific example is the proposal to require people dually eligible for Medicare and Medicaid most of whom have disabilities to enroll in Medicaid managed care plans. Policy makers will be reticent to reduce spending to support this population. But such reductions may be hidden in reductions in spending for other programs, just as the 1996 welfare reform, targeted at low-income families, affected low-income parents with disabilities. Efforts to reduce rapid growth in Medicare and Medicaid are not likely to selectively target working-age people with disabilities. Yet it is hard to imagine that successful efforts to rein in Medicare and Medicaid spending will leave spending for this population untouched. Our findings pose major challenges for policy makers who are attempting to slow spending growth but who do not wish to make workingage people with disabilities even more vulnerable than they already are. Given the nature of current programs, it is difficult to see how they can meet the social goal of providing adequate support to the millions of working-age people with disabilities who rely heavily on federal and state support, while reducing spending growth for other programs. In fact, there is compelling evidence that the average financial well-being of the working-age population with disabilities is already declining. Relative to others, the employment rate of this population has declined since the late 1980s, as has their household income. 24 Reflecting the decline in employment and economic well-being, there is interest in developing or expanding programs that help some in this population increase earnings and reduce their dependence on public benefits. Examples of such expansions include a variety of provisions under the Ticket to Work and Work Incentives Improvement Act of 1999; efforts to reduce or eliminate the two-year waiting period before recipients of Social Security Disability Insurance qualify for Medicare; and demonstrations being conducted by the Social Security Administration, such as the Benefit Offset National Demonstration. Policy changes that increase employment and earnings are the most prominent example of changes that might conceivably pay for themselves, but it is not known whether such programs can be successful. 25 The success of efforts to integrate and coordinate the health care of people with disabilities, along with other efforts to reform care delivery and provider payments, could deliver substantial health care savings and improve the lives of those served. There are, however, numerous practical barriers to establishing care coordination mechanisms, including lack of experience managing the long-term care of people with disabilities, the costs of initial investments to establish programs, separate Medicare and Medicaid payment rules, and stakeholder resistance. 26 To date, the experience in this area has been mixed, especially with respect to costs. 26,27 But given the size and growth of health care spending for people with disabilities, pursuit of these efforts is worthwhile. Conclusions The large and rapidly growing share of government to support workingage people with disabilities, combined with political pressure to reduce the federal deficit, increases the urgency of efforts to make the support system for this population more efficient. Efforts to reduce spending growth by increasing the employment of this population and reducing their reliance on public programs have yet to bear fruit. Because a majority of the are for health care, it is critical to find more efficient ways to provide health care support. In the absence of efficiency gains, policy makers might be unable to avoid funding cuts that will further compromise the well-being of this vulnerable population Health Affairs September :9

8 This study was supported by the National Institute on Disability and Rehabilitation Research, Department of Education, through its Rehabilitation Research and Training Center on Disability Statistics and Demographics grant to Hunter College, City University of New York (No. H133B A). Thecontentsofthisarticledonot necessarily represent the policy of the Department of Education or any other federal agency (Education Department General Administrative Regulations, (b)). The authors are solely responsible for all views expressed. NOTES 1 Census Bureau. Selected social characteristics in the United States: 2008 [Internet]. Washington (DC): Census Bureau; 2010 [cited 2010 Oct 11]. Available from: factfinder.census.gov/servlet/ ADPTable?_bm=y&-geo_id= 01000US&-qr_name=ACS_2008_ 1YR_G00_DP2&-ds_name=ACS_ 2008_1YR_G00_&-_lang=en&-_ caller=geoselect&-redolog= false&-format 2 These figures would be somewhat higher if we included people residing in institutions, but we could not identify published statistics for that population in Unduplicated counts of working-age people with disabilities covered by these programs in 2008 are not available. This estimate is based on the fact that the number enrolled in these programs is approximately equal to the number receiving Social Security Disability Insurance or Supplemental Security Income benefits because of a disability. 4 Centers for Medicare and Medicaid Services. Table 2.4: Medicare enrollment: Hospital Insurance and/or Supplementary Medical iinsurance enrollees, by age, as of July 1, 2009: selected calendar years In: Medicare and Medicaid statistical supplement 2010 edition [Internet]. Baltimore (MD): CMS; 2010 [cited 2011 Jul 20]. Available for download from: MedicareMedicaidStatSupp/ 09_2010.asp#TopOfPage 5 Centers for Medicare and Medicaid Services. Table 13.4: Number of Medicaid persons served (beneficiaries), by eligibility group: fiscal years In: Medicare and Medicaid statistical supplement 2010 edition [Internet]. Baltimore (MD): CMS; 2010 [cited 2011 Jul 20]. Available for download from: MedicareMedicaidStatSupp/ 09_2010.asp#TopOfPage 6 Lakdawalla D, Bhattacharya J, Goldman D. Are the young becoming more disabled? Health Aff (Millwood). 2004;23(1): She P, Livermore G. Long-term poverty and disability among working-age adults. Journal of Disability Policy Studies. 2009;19(4): She P, Livermore G. Material hardship, disability, and poverty among working-age adults. Soc Sci Q. 2007; 88(4): To access the Appendix, click on the Appendix link in the box to the right of the article online. 10 Goodman N, Stapleton D. Federal for working-age people with disabilities. Journal of Disability Policy Studies. 2007;18(2): Changes to the 2002 estimates affected few programs and were done because updated or more accurate information became available subsequent to the publication of the earlier study. Our revised 2002 estimate is greater by $2.4 billion than the estimate of the earlier study, primarily reflecting our inclusion of from five additional sources. See the online Appendix for details (see Note 9). 12 Because Part D are not reported separately for those under age sixty-five, we assumed that in this category were proportional to Parts A and B for those under age sixty-five. The Part D estimate is $6.8 billion. See the online Appendix for details (see Note 9). 13 Centers for Medicare and Medicaid Services. Table 1: National health aggregate, per capita amounts, percent distribution, and average annual percent growth: selected calendar years In: National health data [Internet]. Baltimore (MD): CMS; 2010 [cited 2011 Aug 6]. Available from: NationalHealthExpendData/ downloads/tables.pdf 14 Estimated growth in inflationadjusted Medicaid for working-age adults with disabilities other than those in the disabled Medicaid category was also high (25.0 percent), but represents only 13.1 percent of growth in Medicaid. Because of data limitations (see the Appendix; see Note 9), this estimate is also probably less accurate than the estimate for growth in Medicaid for working-age adults in the disability category. 15 Centers for Medicare and Medicaid Services. Table III.2: Program /trends. In: 2010 CMS statistics [Internet]. Baltimore (MD): CMS; 2010 [cited 2010 Aug 24]. Available from: _CMSStatistics.asp#TopOfPage 16 The corresponding figures for other working-age Medicaid enrollees (relevant to our estimates for Temporary Assistance for Needy Families recipients with disabilities) are 29.9 percent for inflation-adjusted 17.4 percentage points for growth in the number of enrollees, and 12.5 percentage points for growth in per enrollee. 17 Census Bureau. State and local government finance [Internet]. Washington (DC): Census Bureau; 2011 [cited 2011 Jul 21]. Available from: estimate/historical_data_2002.html 18 Social Security Administration. Table 65: Disabled beneficiaries receiving Social Security, SSI, or both: number aged 18 64, by program, December In: Annual statistical report on the Social Security Disability Program, 2009 [Internet]. Baltimore (MD): SSA; 2010 [cited 2011 Jul 21]. Available from: /index.html 19 This estimate includes all Social Security, Supplemental Security Income, and Medicare presented in the Appendix (see Note 9), plus the federal and state Medicaid for enrollees in the disabled Medicaid category. 20 Wamhoff S, Wiseman M. The TANF- SSI connection. Soc Secur Bull. 2006;66(4): Nadel M, Wamhoff S, Wiseman M. Disability, welfare reform, and supplemental security income. Soc Secur Bull. 2004;65(3): Social Security Administration. Actuarial publications: selected data from Social Security s disability program [Internet]. Baltimore (MD): Social Security Administration; [cited 2010 Dec 9]. Available from: oact/stats/dibstat.html 23 National Commission on Fiscal Responsibility and Reform. The moment of truth [Internet]. Washington (DC): National Commission on Fiscal Responsibility and Reform; 2010 [cited 2010 Dec 9]. Available from: files/documents/themomentof Truth12_1_2010.pdf 24 Burkhauser R, Daly M. The declining work and welfare of people with September :9 Health Affairs 1671

9 disabilities: what went wrong and a strategy for change. Washington (DC): AEI Press; See, for example, Stapleton D, Burkhauser R, She P, Weathers R, Livermore G. Income security for workers: a stressed support system in need of innovation. Journal of Disability Policy Studies. 2009; 19(4): Medicare Payment Advisory Commission. Coordinating the care of dual-eligible beneficiaries. In: Report to the Congress: aligning the incentives in Medicare [Internet]. Washington (DC): MedPAC; 2010 [cited 2011 Aug 5]. Available from: documents/jun10_entirereport.pdf 27 Esposito D, Brown R, Chen A, Schore J, Shapiro R. Impacts of a disease management program for dually eligible beneficiaries. Health Care Financ Rev. 2008;30(1): ABOUT THE AUTHORS: GINA LIVERMORE, DAVID C. STAPLETON & MEGHAN O TOOLE Gina Livermore is a senior researcher at Mathematica Policy Research. In this month s Health Affairs, Gina Livermore and current and former colleagues from Mathematica Policy Research report that escalating health costs have driven big increases in federal spending on the working-age people with disabilities now 12 percent of all federal. The resulting pressures will leave policy makers with just three main choices, Livermore says: changing eligibility criteria for programs that serve people with disabilities; reducing benefits; or finding ways to keep these people in the labor market, which will be no easy task in an economy with high joblessness Livermore is a senior researcher at Mathematica, the consulting firm that does research and evaluation mainly for state and federal clients. Her work focuses on issues related to improving the economic well-being and selfsufficiency of working-age people with disabilities and has included research on the prevalence of longterm poverty and material hardship among people with disabilities. Livermore was formerly assistant director of Cornell University s Institute for Policy Research and a vice president at the Lewin Group, a health policy research and management firm. Livermore has a master of public health degree in epidemiology from Tulane University and a doctorate in economics from the University of Wisconsin Madison. David C. Stapleton is a senior fellow at Mathematica Policy Research. David Stapleton is a senior fellow at Mathematica and director of Mathematica s CenterforStudying Disability Policy. He has conducted economic research on the effects of income support and health policy on the employment of people with disabilities for the past twenty years. Stapleton was formerly director of Cornell s Institute for Policy Research, a senior vice president at the Lewin Group, and an associate professor at Dartmouth College and the University of Maryland, College Park. He has a doctorate in economics from the University of Wisconsin-Madison. Meghan O Toole is astrategyanalyst at City Year. Meghan O Toole is a strategy analyst at City Year, a national nonprofit focused on education. At the time of the study featured in this article, she was a research assistant/programmer at Mathematica, focusing on studies of disability, health, and education policy. She has a bachelor s degree from Duke University in public policy studies, French, and European studies Health Affairs September :9