U.S. Department of Health and Human Services

A Profile of Medicaid Institutional and Community-Based Long-Term Care Service Use and Expenditures Among the Aged and Disabled Using MAX 2002: Final Report

Executive Summary

Audra T. Wenzlow, Robert Schmitz and Kathy Shepperson

Mathematica Policy Research, Inc.

January 18, 2008

This report was prepared under contract #HHS-100-97-0013 between the U.S. Department of Health and Human Services (HHS), Office of Disability, Aging and Long-Term Care Policy (DALTCP) and Mathematica Policy Research, Inc. For additional information about this subject, you can visit the DALTCP home page at http://aspe.hhs.gov/_/office_specific/daltcp.cfm or contact the ASPE Project Officer, John Drabek, at HHS/ASPE/DALTCP, Room 424E, H.H. Humphrey Building, 200 Independence Avenue, S.W., Washington, D.C. 20201. His e-mail address is: John.Drabek@hhs.gov.

The opinions and views expressed in this report are those of the authors. They do not necessarily reflect the views of the Department of Health and Human Services, the contractor or any other funding organization.

Since 1982, states have increasingly utilized Section 1915(c) waivers and optional state community-based programs to shift long-term care for the aged and disabled from institutions to the community. New rules introduced under the Deficit Reduction Act (DRA) of 2005 provide states with even more flexibility to provide home and community-based long-term care services to their low-income populations. Two overarching goals underlie these policies: (1) to provide long-term care services more cost-effectively; and (2) to give aged and disabled people more options in how they receive their care. As baby boomers enter their senior years and increase the need for long-term care services nationally, information about how Medicaid community long-term care programs have functioned in the past will be critical for assisting states in choosing how to utilize the new options provided under the DRA. Until recently, only limited aggregate data and some national surveys have been available to examine Medicaid community-based long-term care service use and compare it with use of institutional care. The Medicaid Analytic eXtract (MAX) data system produced by Centers for Medicare & Medicaid Services now enables much more detailed analyses of long-term care utilization and expenditures at the person level.

This study evaluates the potential of using MAX Person Summary files to examine how successfully states have rebalanced their long-term care systems and how Medicaid enrollees who utilize community-based long-term care services differ from people in institutions. For 37 states we believe have reliable MAX long-term care data, we: (1) compare utilization-based measures of the balance of community versus institutional long-term care with traditional expenditure-based measures; (2) contrast patterns of long-term care service utilization and expenditures of aged and disabled subgroups; (3) examine the detailed service types that compose community-based long-term care; and (4) summarize other services used and costs incurred by long-term care users. Each analysis highlights the utility of using person-level data available in MAX to extend our knowledge of how home and community-based long-term care services are used across the country.

Key Findings About Medicaid Long-Term Care Service Use in 2002

The findings presented in this report suggest that there is significant variation across measures, across states, and across population subgroups in patterns of institutional and community-based long-term care use and expenditures. While our national estimates are based on MAX data from only 37 states, our results suggest that further person-level analyses are warranted. In comparing expenditure with utilization-based measures, we find that:

• Only 34 percent of Medicaid long-term care expenditures paid for persons served were for community-based services in 2002, while almost 59 percent of long-term care users used community-based services.

• National estimates mask significant variation across states. Community-based services accounted for over 60 percent of long-term care expenditures in Alaska and New Mexico but less than 12 percent in the District of Columbia and Mississippi. Use of community-based services among long-term care users ranged from 87 percent in Alaska to 23 percent in Indiana.

• While the utilization-based measure was larger than the expenditure measure in every state, there was significant variation across states in how the two measures compared. Illinois and Alabama ranked 31st and 32nd out of the 37 states in the percent of expenditures that were for community-based services but ranked 10th and 8th, respectively, in the percent of long-term care users who used community-based services. In contrast, the two measures of long-term care balance were most similar in New Mexico, which ranked 1st and 3rd in expenditures and use of community-based services.

Our subgroup analyses for aged and disabled enrollees suggest that:

• Institutional and community long-term care expenditures were much more balanced among young disabled Medicaid enrollees than their aged counterparts in 2002. Over half of long-term care expenditures were for community-based services among disabled enrollees but less than 20 percent were for community-based care among those over 65. Community-based service expenditures as a share of total long-term care expenditures ranged from 50 percent for people under age 65, 31 percent for people between ages 65 and 74, 21 percent for people between ages 75 and 84, and 13 percent for those age 85 and older. Rates of community-based service utilization were higher but followed a similar pattern by age.

• The primary distinguishing factor between people using community-based and those using institutional long-term care was age. Eighty percent of people using only institutional care were over age 65, compared with 63 percent of those using both types of services, and 43 percent of those using community services only. Compared with people using community-based services, a higher percentage of people in institutions were non-Hispanic White, female, dual Medicare and Medicaid enrollees, and enrolled in Medicaid for only part of the year--all factors associated with age.

Service type decompositions suggest that MAX data could be used to gain much better understanding of the types of community-based services that are used in Medicaid:

• For example, residential care--community-based services provided in residential settings (excluding home health, adult day care, and private duty nursing)--made up over 6 percent of total long-term care and about 23 percent of community-based long-term care expenditures reported as service types in 2002. However, because waiver services are often not reported as specific service types but grouped with all “Other” services in MAX, the usefulness of MAX for detailed analyses of the composition of Medicaid waiver services is limited.

Finally, our examination of non-long-term care Medicaid service use provides a broader perspective of the types of individuals that use Medicaid long-term care:

• People using both institutional and community-based services (6 percent of long-term care users) had higher average total Medicaid expenditures ($46,055) than users of institutional care only ($38,844) or community care only ($24,966). The high overall expenditures for people using both types of long-term care were due to hospitalization: almost half used Medicaid inpatient services in 2002 compared with about a quarter of other long-term care users. Because short nursing facility stays for acute conditions after hospitalization are covered by Medicare but may include Medicaid cost-sharing, use of both community and institutional care among dual enrollees may reflect stays primarily paid by Medicare rather than Medicaid.

• Overall, aged and disabled enrollees using Medicaid long-term care services accounted for 7.7 percent of all full-benefit Medicaid enrollees in our 37 sample states but represented over 50 percent of their total Medicaid expenditures (including fee-for-service and managed care premiums paid).

The Potential of MAX Data for Understanding Patterns of Long-Term Care

While more current data are available at the aggregate level, MAX 2002 provides the most detailed and current person-level information on all Medicaid enrollees and the services they used during a calendar year. As illustrated by our study findings, MAX can be used to address Medicaid long-term care policy questions that require knowledge about patterns of utilization and expenditures incurred by individuals. MAX could potentially be used to examine whether new enrollees, new long-term care users, or others are utilizing community-based services, and with the use of claims files, how individuals’ use of community services is associated with their future patterns of care.

Several factors that limit the usefulness of MAX data--its timeliness, its completeness, and reporting errors--have greatly improved in recent years and are expected to continue to do so. However, the utility of MAX to examine Medicaid institutional and community-based long-term care services nationally will greatly depend on improvements made by states in the accuracy and detail of long-term care data they report to the Medicaid Statistical Information System (the source data for MAX) and enhancements to MAX coding of community long-term care. We recommend the development of more detailed service-level information for services covered under Section 1915(c) waivers and state plans in future editions of MAX.