At the Intersection of Health, Health Care and Policy Cite this article as: Teresa A. Coughlin, Timothy A. Waidmann and Lokendra Phadera Among Dual Eligibles, Identifying The Highest-Cost Individuals Could Help In Crafting More Targeted And Effective Responses Health Affairs, , no. (2012): doi: 10.1377/hlthaff.2011.0729
The online version of this article, along with updated information and services, is available at: http://content.healthaffairs.org/content/early/2012/04/02/hlthaff.2011.0729.full.html
For Reprints, Links & Permissions: http://healthaffairs.org/1340_reprints.php E-mail Alerts : http://content.healthaffairs.org/subscriptions/etoc.dtl To Subscribe: http://content.healthaffairs.org/subscriptions/online.shtml
Health Affairs is published monthly by Project HOPE at 7500 Old Georgetown Road, Suite 600, Bethesda, MD 20814-6133. Copyright © 2012 by Project HOPE - The People-to-People Health Foundation. As provided by United States copyright law (Title 17, U.S. Code), no part of Health Affairs may be reproduced, displayed, or transmitted in any form or by any means, electronic or mechanical, including photocopying or by information storage or retrieval systems, without prior written permission from the Publisher. All rights reserved.
Advance online articles have been peer reviewed and accepted for publication but have not yet appeared in the paper journal (edited, typeset versions may be posted when available prior to final publication). Advance online articles are citable and establish publication priority; they are indexed by PubMed from initial publication. Citations to Advance online articles must include the digital object identifier (DOIs) and date of initial publication. Not for commercial use or unauthorized distribution Downloaded from content.healthaffairs.org by Health Affairs on April 20, 2012 at LAW OFFICES OF LUCIEN WULSIN
Web First By Teresa A. Coughlin, Timothy A. Waidmann, and Lokendra Phadera
TH E CA RE SPAN
Among Dual Eligibles, Identifying The Highest-Cost Individuals Could Help In Crafting More Targeted And Effective Responses The nearly nine million people who receive Medicare and Medicaid benefits, known as dual eligibles, constitute one of the nation’s most vulnerable and costly populations. Several initiatives authorized by the Affordable Care Act are intended to improve the health care delivered to dual eligibles and, at the same time, to achieve greater control of spending growth for the two government programs. We examined the 2007 costs and service use associated with dual eligibles. Although the population is indeed costly, we found nearly 40 percent of dual eligibles had lower average per capita spending than non-dual-eligible Medicare beneficiaries. In addition, we found that about 20 percent of dual eligibles accounted for more than 60 percent of combined Medicaid and Medicare spending on the dual-eligible population. But even among these high-cost dual eligibles, we found subgroups. For example, fewer than 1 percent of dual eligibles were in high-cost categories for both Medicare and Medicaid. These findings suggest that decision makers should tailor reform initiatives to account for subpopulations of dual eligibles, their costs, and their service use. ABSTRACT
N
ationwide, there are nine million people who receive both Medicaid and Medicare benefits.1 Generally referred to as dual eligibles, this group has come under renewed scrutiny and has taken center stage in the nation’s health policy debate.2,3 Dual eligibles are among the most vulnerable Americans. They live with substantial health needs that can be costly to care for. Recent cost estimates show that dual eligibles account for a disproportionate share of spending in both Medicare and Medicaid.4,5 Strategies aimed at finding ways to better coordinate and integrate care and to change financing for dual eligibles are currently being explored both by states and the federal
doi: 10.1377/hlthaff.2011.0729 HEALTH AFFAIRS 31, NO. 5 (2012): – ©2012 Project HOPE— The People-to-People Health Foundation, Inc.
Teresa A. Coughlin (
[email protected]) is a senior fellow at the Urban Institute Health Policy Center, in Washington, D.C. Timothy A. Waidmann is a senior fellow at the Urban Institute Health Policy Center. Lokendra Phadera is a research associate at the Urban Institute Health Policy Center.
government. At the federal level, for example, the Affordable Care Act of 2010 established the Federal Coordinated Health Care Office, which, among other things, is attempting to align and coordinate benefits between Medicare and Medicaid for dual-eligible users.6 The Center for Medicare and Medicaid Innovation is similarly testing innovative payment and delivery system models that improve quality and value of care in Medicare and Medicaid, as well as other health programs.7 Developing realistic, effective approaches to serve dual eligibles, however, is challenging for many reasons. One of the most salient is that Medicaid and Medicare are two different, very large public health programs that were never intended to work together and, as a result, someM ay 2 0 1 2
Downloaded from content.healthaffairs.org by Health Affairs on April 20, 2012 at LAW OFFICES OF LUCIEN WULSIN
3 1 :5
H e a lt h A f fai r s
1
Web First times work at cross-purposes.8,9 Some policy analysts argue that the lack of coordination between Medicare and Medicaid can lead to poor quality of care for dual eligibles, inefficient care delivery, and increased spending. There has been only limited examination of combined Medicaid and Medicare service use and spending patterns for dual eligibles, despite the major policy importance of this population. In this article we present findings based on an analysis of linked individual-level Medicare and Medicaid data for the national Medicare population in 2007.
Background
16
◀ %
Dual eligibles
More than 16 percent of dual eligibles are in institutions, compared to just 2 percent of other Medicare beneficiaries.
2
Health A ffairs
Past research has demonstrated that there is considerable diversity within the dual-eligible population. This diversity has important implications for these beneficiaries’ health spending and use patterns.10 There are, for example, different types of dual eligibles, depending upon how a person qualifies for Medicaid. About threequarters of dual eligibles are “full duals” and entitled to all Medicaid benefits, including long-term care services.5 In addition, Medicaid pays Medicare cost sharing for these full duals. The balance of dual eligibles are so-called partial duals, who do not receive Medicaid benefits except for help with Medicare cost sharing.11 The dual-eligible population also varies in other dimensions.12 About one-third are disabled and under age sixty-five; two-thirds are lowincome elderly; and many have substantial health care needs, although others have very limited needs. Previous studies have also shown that dual eligibles differ from other Medicare beneficiaries in several ways.4,13 For example, the dually eligible are more likely to be younger and poorer and to report lower health status than other Medicare beneficiaries. They are also more likely never to have been married and to be a member of a racial or ethnic minority. Finally, and most importantly, dual eligibles are much more likely than other Medicare beneficiaries to live in an institution. More than 16 percent of the dualeligible population is institutionalized, compared to just 2 percent of other Medicare beneficiaries.
Study Data And Methods Data For this study, we relied on two primary data sources: the 2007 Medicare Current Beneficiary Survey and the 2007 Medicaid Statistical Information System Summary File. The latter has been adjusted to reflect Medicaid spending as reported on form CMS-64, which is a quarterly May 2 012
3 1:5
statement of expenditures for which states are entitled to federal reimbursement. Below we provide a summary of the characteristics of these data sources. For additional detail on these sources and their key limitations, see the online Appendix,14 which also describes our analytical approach to problems we encountered developing our estimates. Medicare Current Beneficiary Survey The Medicare Current Beneficiary Survey provides detailed information on spending by Medicare and Medicaid as well as other payment sources, such as private insurance or beneficiary out-ofpocket spending. In our study, however, we focused on spending by public programs. As such, our analysis examined only that spending by Medicare beneficiaries that was paid by Medicare, and only that spending by dual eligibles that was paid for by Medicare and Medicaid. Limiting spending to what was paid for by Medicare and Medicaid explains why we found that other Medicare beneficiaries (non–dual eligibles) had no long-term care spending (Exhibit 1). If we had included spending by other sources, such as out-of-pocket or private longterm care insurance, other Medicare beneficiaries would have shown spending for long-term care services as well as higher spending on other services. For this analysis, we used the 2007 Medicare Current Beneficiary Survey Cost and Use file, which provided information on Medicare beneficiaries who were ever enrolled in the program during that calendar year. This category includes those who died before the end of the year as well as those who enrolled during the year. The Cost and Use file, which is a combination of the Medicare Current Beneficiary Survey and Medicare administrative data, contained information on several health care services used and their associated costs as well as a variety of information on health status and social and demographic characteristics. It also included information on prescription drugs and long-term care services. Although the Medicare Current Beneficiary Survey collects information on Medicaid spending, it does not provide accurate spending information for home and community-based waiver programs—an important expenditure item for the dual-eligible population. For this reason, we relied exclusively on the Medicaid Statistical Information System for Medicaid spending information for our study. Medicaid Statistical Information System The 2007 Medicaid Statistical Information System Summary File covered the 2007 federal fiscal year—that is, October 1, 2006, to September 30, 2007. For each federal fiscal year, the summary file contains individual-level data for all fifty
Downloaded from content.healthaffairs.org by Health Affairs on April 20, 2012 at LAW OFFICES OF LUCIEN WULSIN
Exhibit 1 Medicare And Medicaid Spending Patterns By Dual Eligibles, 2007 Per capita spending ($)
Total spending Acute care Inpatient care Ambulatory care Rx drugs Other acute care (includes capitation) Subacute care Skilled nursing facility care Home health care Long-term care Institutional care Communitybased care
Per user spending ($)
Total spending ($ millions)
Combined
Share Medicaid (%)
Combined
Medicare
Medicaid
Combined
Medicare
Medicaid
Medicare
Medicaid
29,868 17,227 8,312 3,928 2,961
15,850 13,783 7,864 2,629 2,878
14,018 3,444 448 1,299 83
30,347 17,504 26,839 4,155 3,390
16,474 14,329 27,815 2,863 3,738
17,323 4,368 4,017 2,156 224
265,702 153,252 73,945 34,942 26,342
141,000 122,615 69,957 23,384 25,601
124,703 30,637 3,988 11,558 741
46.9 20.0 5.4 33.1 2.8
2,026 2,531
413 2,067
1,613 464
2,622 11,763
1,038 11,625
2,398 7,924
18,024 22,514
3,673 18,385
14,350 4,129
79.6 18.3
1,139 1,392 10,110 6,789
1,139 928 —a —a
—a 464 10,110 6,789
12,686 8,561 33,689 42,410
12,686 7,584 —a —a
—a 7,924 33,689 42,410
10,133 12,381 89,936 60,393
10,133 8,252 —a —a
—a 4,129 89,936 60,393
0.0 33.4 100.0 100.0
3,321
—a
3,321
21,242
—a
21,242
29,543
—a
29,543
100.0
SOURCE Medicaid Statistical Information System and Medicare Current Beneficiary Survey 2007 linked file. aNot applicable.
states and the District of Columbia, with aggregate measures of Medicaid spending for thirty service categories, including ambulatory, acute, and institutional services, for all who received Medicaid services nationwide. For complete Medicaid service spending, we matched the Medicaid Statistical Information System service expenditures to those reported on the form CMS-64, which is considered a more accurate financial data source for Medicaid service spending. By benchmarking to the CMS-64 report, our spending estimates included special payments made by Medicaid, such as disproportionateshare hospital payments, which provide support to hospitals that serve a large number of lowincome patients, such as people with Medicaid and the uninsured. So, although Medicaid spending estimates were adjusted for special program payments, Medicare spending estimates were not—an admitted limitation of our study. Expenditures reported in the Medicaid Statistical Information System file include payments made to providers on behalf of dual eligibles for Medicare cost sharing, but they do not include payments to Medicare for premiums. To account for these payments, we calculated premiums paid by Medicaid to Medicare for each acute and subacute care service, reduced those amounts from their respective spending categories in Medicare, and added them to Medicaid spending categories.
Also, for dual eligibles enrolled in managed care, the Medicaid Statistical Information System file includes capitation payments only. No information on spending by type of service is available for these dual eligibles. Although this absence of data constitutes a limitation, fewer than 12 percent of dual eligibles were enrolled in capitated managed care arrangements in 2009.15 In addition to spending data, the Medicaid Statistical Information System file contains information on the personal characteristics of Medicaid beneficiaries, such as date of birth, sex, race and ethnicity, state and county of residence, and eligibility information that specifies why a person is enrolled in Medicaid. To identify whether the beneficiary was a full-benefit dual eligible or a Medicare Savings Plan beneficiary, we used the Medicaid Statistical Information System variable that describes a beneficiary’s most recent or “last-best” monthly eligibility code. Methods Subacute and long-term care services are divided into institutional and community-based care. Institutional care is provided by nursing facilities, intermediate care facilities for the mentally retarded, and institutions for mental disease. Because of the fundamental difference in the nature and duration of these benefits, we separated the services covered by Medicare and Medicaid in our calculations. Communitybased care was divided into home health services covered by both Medicaid and Medicare, personal care services covered by Medicaid only, May 2 012
3 1:5
Downloaded from content.healthaffairs.org by Health Affairs on April 20, 2012 at LAW OFFICES OF LUCIEN WULSIN
H e a lt h A f fai r s
3
Web First
◀
29,868
and long-term care provided under Medicaid’s home and community-based service waiver programs. All analyses presented were weighted using the Medicare Current Beneficiary Survey crosssectional weight. This weight is intended for use in cross-sectional statistics involving the total (combined) national sample; however, it also can be used for analyzing representative subgroups. Given the relatively high match rate between the Medicare Current Beneficiary Survey and the Medicaid Statistical Information System, we believe that the combined data file provided a representative sample of the national dual-eligible population.
$
Spending on dual eligibles
Study Results
In 2007 Medicare and Medicaid combined spent $29,868 per dual eligible—more than four times the per capita spending on other Medicare beneficiaries.
Health Care Service Use And Costs Given their health and functional status, we found that dual eligibles had a higher level of service use in all of the categories examined when compared to other Medicare beneficiaries (data not shown). These differences were particularly large for outpatient hospital services, inpatient hospital services, and institutional long-term care. Corresponding to this higher utilization, dual eligibles were more costly to care for than non– dual eligibles (Exhibit 1). In 2007 Medicare alone spent $15,850 per capita on dual eligibles—more than twice the per capita spending on other Medicare beneficiaries ($7,226; not shown). Medicaid per capita spending on dual eligibles was $14,018. The combined average per capita Medicare and Medicaid spending for dual eligibles was, therefore, $29,868—more than four times the per capita spending for nondual-eligible Medicare beneficiaries. It is important to remember that in this study we focused on public spending on dual eligibles and included
Exhibit 2 Total Medicare And Medicaid Spending By Dual Eligibles And Other Medicare Beneficiaries, 2007 State Medicaid spending Federal Medicaid spending Federal Medicare spending
Dual eligibles
SOURCE Medicaid Statistical Information System and Medicare Current Beneficiary Survey 2007 linked file.
4
Health Affairs
May 2 012
3 1:5
only those services paid for by Medicare or Medicaid. Health care services paid by other sources, such as private health insurance or out of pocket, were excluded. It is this large spending difference that makes the dually eligible population such an important policy group. But these results reflect average spending. An often overlooked fact is that not all dual eligibles are high cost. In 2007 about twofifths of dual eligibles had combined Medicaid and Medicare annual spending that was less than the average $7,226 per capita Medicare spent on non–dually eligible beneficiaries in that year (data not shown). About 37 percent of these low-cost beneficiaries were “partial duals”; the remaining 63 percent were full duals. Collectively, Medicare and Medicaid spending for the 8.9 million dual-eligible beneficiaries totaled $265.7 billion in 2007 (Exhibit 2). For the 35.4 million non-dual-eligible Medicare beneficiaries, Medicare spending totaled $256.0 billion. Although Medicare and Medicaid financed similar shares of spending for the dual-eligible population—$141.0 billion and $124.7 billion, respectively—four-fifths of all dollars spent on dual eligibles were federal funds. Between the two programs, the federal government paid $210.4 billion in 2007 on the dual-eligible population—$141.0 billion through Medicare and another $69.4 billion through federal Medicaid matching payments. State Medicaid spending for dual eligibles totaled $55.3 billion. Share Of Spending By Type of Service Although Medicaid and Medicare each financed about half of the total spending on the dualeligible population, their shares varied considerably by type of service (Exhibit 1). Overall, Medicare paid for the bulk of acute care services, whereas Medicaid paid for 20 percent. Medicaid’s share, however, ranged from a low of 2.8 percent for prescription drugs and 5.4 percent for inpatient hospital care to a high of 79.6 percent for other acute care services. Many acute care services not covered by Medicare—such as dental and vision care—are paid for by state Medicaid programs. For subacute care, Medicare again financed the vast majority of these services; Medicaid paid 18.3 percent. In contrast, Medicaid clearly dominated financing of dual eligibles’ institutional and communitybased long-term care. These findings were consistent with the different coverage mandates of the two programs. Medicare covers many acute care services, and Medicaid provides wrap-around coverage by paying Medicare cost sharing for beneficiaries who are either fully or partially eligible. And for “full duals,” Medicaid provides additional benefits,
Downloaded from content.healthaffairs.org by Health Affairs on April 20, 2012 at LAW OFFICES OF LUCIEN WULSIN
Exhibit 3 Dual Eligibles’ Enrollment And Total Medicare And Medicaid Spending, By Spending Groups, 2007
Percent
including some acute and subacute care services beyond what Medicare covers, such as home health and, importantly, long-term care services. Looking at per capita and per user spending for dual eligibles (Exhibit 1), we found little difference for acute care services—$17,227 and $17,504, respectively. This reflected the fact that virtually all dual eligibles used acute care services, particularly ambulatory care and prescription drugs. For subacute and long-term care services, however, a substantial difference between per capita and per user spending was found, highlighting the more limited use of these services. Among users, the most expensive service, by far, was institutional long-term care, which averaged $42,410 per user. Although lower than institutional long-term care, per user costs for community-based long-term care—$21,242— were also high. Who Are High-Cost Dual Eligibles ? As is the case for health spending generally, a relatively small portion of dual eligibles was responsible for the majority of the population’s total spending: 20 percent of dual eligibles accounted for 60.6 percent of combined Medicaid and Medicare spending on the population (Exhibit 3). This skewed spending suggests that policies aimed at high-cost dual eligibles will provide the best value and derive the greatest efficiencies. But within this high-cost group, there were critical subgroups—a fact that has important policy implications. Owing to the way services are covered by Medicare and Medicaid, what makes dual eligibles high cost in one program does not necessarily make them high cost in the other. Indeed, we found a relatively small overlap in the highest spenders in the two programs. Looking at the top 10 percent of dual eligibles in Medicaid spending, defined as more than $45,180, fewer than 15 percent were also in the top 10 percent of the dual eligibles for Medicare spending, defined as more than $44,348. Just over 80,000 dual eligibles—less than 1 percent of all dual-eligible beneficiaries—were in the highest 10 percent of the spending distribution in both Medicaid and Medicare. It is these high-cost dual eligibles who would particularly benefit from initiatives aimed at coordinating care and aligning incentives between the two programs. That said, high-cost MedicareMedicaid dual eligibles accounted for only about 4.6 percent of total spending on the population in 2007 ($12.3 billion of the $265.7 billion spent on dual eligibles overall). Much larger shares of Medicare and Medicaid funding were devoted to the top spenders in each of the programs (Exhibit 4). For the 900,000 dual eligibles who were in the top 10 percent
Spending
SOURCE Medicaid Statistical Information System and Medicare Current Beneficiary Survey 2007 linked file. NOTES “Top 10 percent of Medicaid spenders” is Medicaid spending greater than $45,180. “Top 10 percent of Medicare spenders” is Medicare spending greater than $44,348.
of Medicaid spending, $80.1 billion in Medicare and Medicaid funds were spent on this group, accounting for nearly 30.1 percent of overall spending on the dual-eligible population. And for the 900,000 dual eligibles who were in the top 10 percent of Medicare spending, Medicare and Medicaid spending totaled $93.1 billion, or
Exhibit 4 Total Medicaid And Medicare Spending By High-Cost Dual Eligibles’ Spending Distributions, By Type Of Service, 2007
SOURCE Medicaid Statistical Information System and Medicare Current Beneficiary Survey 2007 linked file. NOTES “Top 10 percent of Medicaid spenders” is Medicaid spending greater than $45,180. “Top 10 percent of Medicare spenders” is Medicare spending greater than $44,348.
May 2 012
3 1:5
Downloaded from content.healthaffairs.org by Health Affairs on April 20, 2012 at LAW OFFICES OF LUCIEN WULSIN
H e a lt h A f fai r s
5
Web First 35.0 percent of overall spending on the dualeligible population. Given the sizable resources devoted to these two groups, care improvement and efficiency strategies aimed at high-cost Medicare dual eligibles and high-cost Medicaid dual eligibles would be especially worthwhile. However, different strategies may be needed, as there is little overlap in the types of services that the two groups of dual eligibles use. For the high-cost Medicaid dual eligible, more than 73 percent of spending was for long-term care services, while just 23 percent was for acute care services (Exhibit 4). Moreover, Medicaid financed 82.8 percent of the spending on high-cost Medicaid dual eligibles (data not shown). In contrast, for high-cost Medicare dual eligibles (Exhibit 4), more than 78 percent of spending went for acute care services, and less than 10 percent was for long-term care services. Medicare paid for 86.3 percent of this group’s spending. We compared the demographic and health characteristics of high-cost Medicare and Medicaid dual eligibles to those of lower-cost dual eligibles—in other words, to those in the bottom 50 percent of the combined Medicaid and Medicare spending distribution, with spending of less than $13,009. The results of this comparison suggested several factors that distinguish highcost dual eligibles from their lower-cost counterparts (Exhibit 5). For example, we found that living in an institution and, closely related, having three or more limitations in daily living activities, such as bathing or using the toilet, come close to completely
determining who is a high-cost Medicaid dual eligible. Being over age eighty or having diabetes were other important risk factors for having high Medicaid costs. For those high-cost Medicare dual eligibles, living in an institution was not particularly predictive of cost status. However, having three or more limitations in activities of daily living and a diagnosis of heart disease were predictive of high-cost Medicare dual-eligible status. Finally, self-reported mental illness did not appear to be predictive of high cost for either Medicare or Medicaid dual eligibles. Instead, mental illness was equally prevalent across the three spending groups, with about one-third of each reporting a mental illness.
Discussion Our analysis of linked Medicare and Medicaid data confirmed the well-known fact that, on average, dual eligibles were among the most costly beneficiaries in both programs. Results of our analysis advance the understanding of this population by showing that dual eligibles are a highly diverse group in terms of their spending, and that being a dual eligible is not necessarily synonymous with high spending. For example, we found that nearly 40 percent of dual eligibles had lower average per capita spending than other non-dual-eligible Medicare beneficiaries. At the same time, a sizable share of the dualeligible population had very high costs and accounted for a disproportionate share of spending on this population. About 20 percent of dualeligible beneficiaries accounted for more than
Exhibit 5
Percent
Selected Characteristics Of High- And Low-Cost Dual Eligibles, 2007
Living in an institution
Diabetes
Heart disease
Mental illness
SOURCE Medicaid Statistical Information System and Medicare Current Beneficiary Survey 2007 linked file. NOTES “Top 10 percent of Medicaid spenders” is Medicaid spending greater than $45,180. “Top 10 percent of Medicare spenders” is Medicare spending greater than $44,348. “Bottom 50 percent of spenders” is Medicaid and Medicare spending less than $13,009. ADL is activities of daily living.
6
H ea lt h A f fai r s
M ay 2 0 1 2
31:5
Downloaded from content.healthaffairs.org by Health Affairs on April 20, 2012 at LAW OFFICES OF LUCIEN WULSIN
It may be important to reach dual eligibles who are high cost in one program but not the other.
60 percent of combined Medicaid and Medicare spending on the population. This distribution is consistent with general health care spending— that is, a small fraction of the population drives much of the spending on dual eligibles as it does for the population at large. It is important to note that our study is a national analysis. Given the variation among state Medicaid programs in terms of coverage, eligibility, and spending, study findings will vary by state. High-cost dual eligibles, however, are not a homogenous group.We found, rather, that there are distinct subgroups—a finding that has important implications for developing policies for this population. We found, for example, that only a very small number of dual eligibles were high-cost beneficiaries in both Medicare and Medicaid. Fewer than 1 percent of dual eligibles nationally were in the highest 10 percent of the spending distribution in both programs. Although spending on the group was considerable, collectively highcost dual-eligible beneficiaries in both Medicare and Medicaid accounted for just 4.6 percent of overall spending on dual eligibles in 2007. Instead, much of the spending on dual eligibles was concentrated in two distinct groups. The top 10 percent of spenders in Medicaid made up nearly one-third of the combined MedicareMedicaid spending, and the top 10 percent of spenders in Medicare accounted for a slightly larger share of combined spending, 35 percent. We found very little overlap between the two groups in both service use and financing. For top-spending Medicaid dual eligibles, the vast majority of their spending was for long-term care services, most of which were paid for by Medicaid. For top-spending Medicare dual eligibles, most spending was for acute care, which was overwhelmingly paid for by Medicare.
Conclusion In summary, we found that dual eligibles were a highly diverse population in terms of their
spending and service-use patterns. There was a substantial portion of the population with no extraordinary spending. There was another smaller share that had high Medicaid spending. There was a largely separable group, again small, that had high Medicare spending. Finally, there was a small fraction with both high Medicaid and high Medicare spending. These findings suggest that decision makers need to take a multiprong policy approach to improving the value, efficiency, and quality of care provided to dual-eligible beneficiaries. For example, carefully designed care coordination and integration initiatives are likely to be a promising strategy for dual eligibles who are both high-cost Medicare and high-cost Medicaid beneficiaries. By working together, Medicare and Medicaid could improve the efficiency of care provided for dual eligibles who rely heavily on both programs. For other strategies that are focused on improving how each program works for dual eligibles—including Medicare’s delivery of acute and postacute services and Medicaid’s delivery of long-term care services—it may be important to reach dual eligibles who are high cost in one program but not the other. For example, the bulk of high-cost Medicaid dual eligibles use long-term care services, and many live in nursing homes. For institutionalized dual eligibles, it would make sense, for example, to have centralized coordination and care management efforts at the nursing home. Activities could focus on, among other things, preventing avoidable hospitalizations or further deterioration in physical functioning. For high-cost Medicaid dual eligibles living in the community, coordination of activities could be centered with a community-based provider, such as a primary care provider or social worker. These coordinated activities should ensure that beneficiaries receive needed medical services, medications, and supportive care and social services. Particular effort should be directed at avoiding premature or unnecessary nursing home placement. For dual eligibles with high Medicare costs but not extraordinarily high Medicaid costs, other care coordination approaches might be more effective. Given that many of these dual eligibles have chronic health conditions and are heavy users of acute care services, this group would benefit from efforts such as medical homes or accountable care organizations. These efforts encourage primary care, care coordination, and data sharing across several providers to reduce the inappropriate use of costly services, including emergency department, inpatient, and nursing home care. M ay 2 0 1 2
31:5
Downloaded from content.healthaffairs.org by Health Affairs on April 20, 2012 at LAW OFFICES OF LUCIEN WULSIN
H e a lt h A f fai r s
7
Web First Recognizing that there are many relatively inexpensive dual eligibles is also important when developing policies. For example, managed care can result in higher-than-necessary payments for low-cost dual eligibles if capitation rates are not appropriately adjusted. Support for this analysis was provided by the Kaiser Commission on Medicaid and the Uninsured, Kaiser Family
By carefully designing programs to meet the specific needs of different types of dual eligibles, the Medicare and Medicaid programs can improve efficiency and value while maintaining and perhaps improving access to needed services. ▪
Foundation. The authors thank the Centers for Medicare and Medicaid Services for providing the information
needed to link data sets required to do the analysis. [Published online April 18, 2012.]
NOTES 1 Centers for Medicare and Medicaid Services. Fact sheets: people enrolled in Medicare and Medicaid [Internet]. Baltimore (MD): CMS; 2011 May 11 [cited 2012 Mar 15]. Available from: http://www.cms .gov/apps/media/press/factsheet .asp?Counter=3954 2 House Energy and Commerce Committee. Dual-eligibles: understanding this vulnerable population and how to improve their care: hearing before the Subcommittee on Health of the House Committee on Energy and Commerce, 112th Cong., 1st Sess. [Internet]. Washington (DC): The Committee; 2011 Jun 21 [cited 2012 Apr 10]. Available from: http:// democrats.energycommerce.house .gov/index.php?q=hearing/hearingon-dual-eligibles-understandingthis-vulnerable-population-andhow-to-improve-care 3 Adamy J. Overlapping health plans are double trouble for taxpayers. Wall Street Journal. 2011 Jun 27. 4 Medicare Payment Advisory Committee. Report to the Congress: Medicare and the health care delivery system. Washington (DC): MedPAC; 2011 Jun. 5 Rousseau D, Clemans-Cope L, Lawton E, Langston J, Connolly J, Howard J. Dual eligibles: Medicaid enrollment and spending for Medicare beneficiaries in 2007. Washington (DC): Kaiser Commission on Medicaid and the Uninsured; 2010 Dec. 6 Centers for Medicare and Medicaid Services. Overview: about the Medicare-Medicaid coordination office
8
Health A ffairs
May 2 012
3 1:5
7
8
9
10
11
[Internet]. Baltimore (MD): CMS; [cited 2012 Apr 10]. Available from: http://www.cms.hhs.gov/MedicareMedicaid-Coordination/Medicareand-Medicaid-Coordination/ Medicare-Medicaid-CoordinationOffice/index.html Guterman S, Davis K, Stremikis K, Drake H. Innovation in Medicare and Medicaid will be central to health reform’s success. Health Aff (Millwood). 2010;29(6):1188–93. Walsh EG, Freiman M, Haber S, Bragg A, Ouslander J, Wiener JM. Cost drivers for dually eligible beneficiaries: potentially avoidable hospitalizations from nursing home facility, skilled nursing facility, and home and community-based service waiver programs: report to the Centers for Medicare and Medicaid Services. Waltham (MA): RTI International; 2010 Aug. Segal M. Dual eligible beneficiaries and potentially avoidable hospitalizations [Internet]. Baltimore (MD): Centers for Medicare and Medicaid Services; 2011 [cited 2012 Mar 15]. (Policy Insight Brief). Available from: https://www.cms.gov/InsightBriefs/downloads/PAHInsight Brief.pdf Medicare Payment Advisory Committee. Report to the Congress: aligning incentives in Medicare. Washington (DC): MedPAC; 2010 Jun. There are several ways in which Medicare beneficiaries can qualify for some level of Medicaid benefits and become a dual eligible. Dual eligibles include elderly and disabled
12
13
14
15
Downloaded from content.healthaffairs.org by Health Affairs on April 20, 2012 at LAW OFFICES OF LUCIEN WULSIN
beneficiaries who receive cash assistance through the Supplemental Security Income program, the federal program that provides cash assistance to low-income aged and disabled people. They also include elderly Medicare beneficiaries who spend down their financial resources in nursing homes and qualify for Medicaid. The dual-eligible population also includes poor elderly Medicare beneficiaries who rely on Medicaid’s community-based longterm care services. Although most dual eligibles receive full Medicaid benefits, some receive assistance with Medicare premiums and other cost-sharing requirements only. People who qualify for this level of assistance generally have limited income and resources but do not meet the state’s Medicaid eligibility standards. Here we refer to these beneficiaries as “partial duals.” Medicare Payment Advisory Committee. Report to the Congress: Medicare and the health care delivery system. Washington (DC): MedPAC; 2010 Jun. Kaiser Commission on Medicaid and the Uninsured. Dual eligibles: Medicaid’s role for low-income Medicare beneficiaries. Washington (DC): The Commission; 2010 May. To access the Appendix, click on the Appendix link in the box to the right of the article online. Centers for Medicare and Medicaid Services. 2009 Medicaid Managed Care enrollment report. Baltimore (MD): CMS; 2009 Jun.
ABOUT THE AUTHORS: TERESA A. COUGHLIN, TIMOTHY A. WAIDMANN & LOKENDRA PHADERA
Teresa A. Coughlin is a senior fellow at the Urban Institute Health Policy Center.
In this month’s Health Affairs, Teresa Coughlin and coauthors report on their analysis of dual eligibles—the nearly nine million people who receive both Medicare and Medicaid—and their findings of key differences among duals in terms of service use and overall health spending. Notably, in 2007, 20 percent of dual eligibles accounted for more than 60 percent of combined Medicaid and Medicare spending, while relatively few were high cost in both Medicare and Medicaid. Although all dual eligibles are poor, says Coughlin, “our analysis shows that not all are expensive to care for, a fact that tends to get overlooked in the current policy discussion.” The authors write that their analysis should inform decisions of policy makers in shaping specific reform initiatives to account for the subpopulations of dual eligibles.
Coughlin, a senior fellow at the Urban Institute Health Policy Center, researches and analyzes a range of health issues with particular emphasis on Medicaid, managed care, the health care safety net, and state health policy. In her current research, she is conducting two studies on dualeligible enrollees. One jointly examines Medicare and Medicaid spending and use patterns for duals; the other looks at state Medicaid spending variation for them. Coughlin holds a master’s degree in public health from the University of California, Berkeley.
Timothy A. Waidmann is a senior fellow at the Urban Institute Health Policy Center.
Timothy Waidmann is also a senior fellow at the Urban Institute Health Policy Center. He has more than twenty years’ experience designing and conducting studies in a variety of health policy topics, including disability and health among the elderly, Medicare and
Medicaid policy, disability and employment, public health and prevention, and health status and access to health care in vulnerable populations. Waidmann, who received a master’s degree and a doctorate in economics from the University of Michigan, won the Urban Institute President’s Award for Outstanding Research Publication, Policy Research, in 2010.
Lokendra Phadera is a research associate at the Urban Institute Health Policy Center.
Lokendra Phadera is a research associate at the Urban Institute Health Policy Center. His research has focused on state health reform evaluations, spending and service utilization patterns of people who are in both Medicaid and Medicare, and national health care spending and utilization trends. Phadera received his bachelor’s degree in economics from Middlebury College.
M ay 20 1 2
31:5
Downloaded from content.healthaffairs.org by Health Affairs on April 20, 2012 at LAW OFFICES OF LUCIEN WULSIN
Health Affa irs
9
