Home
HOME ARCHIVE SEARCH TABLE OF CONTENTS

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
PubMed
Right arrow PubMed Citation
The Gerontologist 46:165-172 (2006)
© 2006 The Gerontological Society of America

Residential Care Provision in Medicaid Home- and Community-Based Waivers: A National Study of Program Trends

Martin Kitchener, PhD1, Mauro Hernandez, PhD(c)1, Terence Ng, MA1 and Charlene Harrington, PhD1

Correspondence: Address correspondence to Martin Kitchener, Associate Adjunct Professor, Department of Social and Behavioral Sciences, University of California, San Francisco, 3333 California Street, Suite 455, San Francisco, CA 94118. E-mail: Martin.Kitchener{at}ucsf.edu


    Abstract
 TOP
 Abstract
 Methods
 Results
 Discussion and Conclusions
 References
 
Purpose: While state policy and market factors are known to have contributed to the increased supply of residential care, little is known about efforts to accommodate demand from lower-income consumers. This study describes participation and expenditure trends for residential care services funded by Medicaid waivers and examines variation across programs. Design and Methods: We collected annually reported Center for Medicare and Medicaid Services (CMS) Form 372 data from state officials for each waiver that provides residential care services for the period 1995 to 2002. Descriptive statistics examined waiver program participation and expenditures while adjusting for population changes and inflation. Results: Between 1995 and 2002, national Medicaid waiver-funded residential care participants increased by almost threefold to 120,000 and expenditures more than quadrupled to $2.3 billion. However, Medicaid waiver program participation and spending varied considerably by state, by target population, and by level of care. Implications: This study highlights three important policy concerns: (a) Medicaid-supported residents remain underrepresented in residential care, (b) large interstate variation persists in Medicaid residential care service provision, and (c) state policy choices favor Medicaid spending on residential care for persons with mental retardation or developmental disabilities.

Key Words: Residential care • Medicaid • Home- and community-based services • Waivers • State policy


The federal-state Medicaid program is the second largest budget item for most states, and in 2003 it paid for 44% of the nation's $151 billion total long-term-care expenditures (Smith, Cowan, Sensenig, & Catlin, 2005). While institutional care (e.g., nursing homes) consumed 67% of Medicaid long-term-care expenditures (Burwell, Sredle, & Eiken, 2004), pressures, including rising public demand and the 1999 Supreme Court Olmstead decision, require that states develop alternative home- and community-based service (HCBS) programs (Kitchener & Harrington, 2004). By 2001, Medicaid HCBS had expanded to serve 2.1 million participants at a cost of $22 billion (Kitchener, Ng, Miller, & Harrington, 2005).

Residential care is one HCBS option that has experienced dramatic growth in the last two decades, although primarily as a private-pay option. Unlike mandatory Medicaid long-term-care services (e.g. nursing homes), states may choose to offer residential care as an optional benefit. Federal guidelines allow states considerable discretion in defining both residential care services and eligible settings (Carlson, 2005; Mollica & Johnson-Lamarche, 2005). As a result, local political dynamics have produced huge interstate variation in both the nature of waiver-funded residential care, and the types of eligible service providers.These may be known as boarding homes, homes for the aged, assisted living facilities, adult foster care, rest homes, and other terms (Mollica & Johnson-Lamarche).

Previous studies have identified factors contributing to low residential care participation rates and possible barriers to expansion of Medicaid-funded residential care programs (Mollica & Johnson-Lamarche, 2005; O'Keeffe, O'Keeffe, & Bernard, 2003), while others have reported considerable variation in HCBS expenditures and participants (Burwell et al., 2004; Kitchener, Carrillo, & Harrington, 2003; Kitchener et al., 2005). This article provides the first national study of Medicaid waiver expenditures and participation trends for residential care services.

Findings from this study extend knowledge of Medicaid HCBS waiver services provided in residential care facilities in two ways. First, the study draws from the most recent and comprehensive available dataset to present national program (participation and expenditure) trends for all waivers that provided residential care services for the period 1995 to 2002. Second, it reports important differences in program trends for targeted populations while describing considerable variation between states.

Residential Care
In pursuit of Medicaid long-term-care goals such as redressing the bias toward funding institutional care and controlling costs (Kitchener, Beynon, & Harrington, 2004), states have increasingly adopted policies that facilitate access to residential care facilities by lower-income residents. Such efforts may also reflect states responding to the 1999 Olmstead Supreme Court decision and related federal guidelines (Mollica & Johnson-Lamarche, 2005). Although many nursing home residents might be served in residential care (Newcomer et al., 2001; Spector, Reschovsky, & Cohen, 1996), low- and moderate-income residents may face access barriers to using these facilities (Hawes, Phillips, & Rose, 2000; Spillman, Liu, & McGilliard, 2002).

In the absence of any uniform definition among policy makers and researchers, there is general agreement that residential care facilities are nonnursing home, residential settings that provide room, board, assistance with activities of daily living (ADLs), and 24-hr oversight (Assisted Living Workgroup, 2003). These settings are often conceived as falling somewhere in the middle of a continuum of long-term-care options between nursing home and in-home care. Because regulatory policy decisions for residential care facilities occur primarily at the state level, local political and structural factors may shape licensing requirements. As a result, residential care regulations vary in terms of the services that are required versus permitted, the types of residents that can be admitted or retained, as well as how and where services can be provided (Mollica & Johnson-Lamarche, 2005).

Efforts to describe the residential care facility supply and resident population have been limited by the lack of uniform definitions and data (Wunderlich & Kohler, 2001). When counting all types of residential care facilities in the United States (including small foster care homes and facilities serving nonaged populations), there were 55,141 licensed facilities and 1,026,397 beds in 2002. Between 1990 and 2002, the residential care industry experienced significant growth both in terms of the number of facilities (57%) and licensed beds (97%; Harrington, Chapman, Miller, Miller, & Newcomer, 2005). Interviews with Medicaid directors and state officials suggest that residential care bed supply may be inadequate, particularly for low-income residents (Harrington, LeBlanc, Wood, & Satten, 2000).

National studies have attempted to describe the facility and resident characteristics of residential care facilities serving primarily an older population. Such facilities vary considerably in terms of size, age, price, private units, available services, and admission and discharge criteria (Hawes, Rose, & Phillips, 2000; Zimmerman, Sloane, & Eckert., 2001). A typical resident is a White, widowed woman in her mid-80s (Hawes, Rose, et al.; Spillman et al., 2002). Residential care facility residents have significant long-term-care needs but are much less impaired than nursing facility residents. Nationally, one in two residents(52%) needs help with three or more ADLs, compared to three in four (74%) nursing facility residents (Spillman et al.). A more or less impaired residential care population is likely the product of complex interactions between state-level (licensing, reimbursement, etc.), facility-level (organizational characteristics and service capacity), and individual-level (resources, functional status, etc.) factors (Bernard, Zimmerman, & Eckert, 2001).

Low- and moderate-income residents seem to make up a significant, though possibly decreasing, proportion of the residential care facility population. The proportion of residents receiving Medicare and earning less than $10,000 per year declined from 55% in 1992 to 48% in 1998 (Spillman et al., 2002). Lower-income individuals seem to be more likely to live in smaller, lower-priced homes with fewer residential and safety-enhancing design features (Newcomer, Breuer, & Zhang, 1994; Salmon, Hyer, Hedgecock, Zayac, & Engh, 2004).

Residential Care Funding Options
Most residential care residents use their personal income, often supplemented by other financial assets, to pay for monthly housing and service costs. Other private sources of payment include family supplementation and, to a lesser degree, private long-term-care insurance policies (National Center for Assisted Living, 2001). Probably the most common public-payment source for low-income facility residents is provided through the federal Supplemental Security Income (SSI) program, which is available in every state. States may also provide a supplemental payment above the monthly SSI payment to cover additional room, board, and some service costs (Mollica & Johnson-Lamarche, 2005). For example, individuals living in California-licensed Residential Care Facilities for the Elderly receive a federal monthly SSI payment of $564 and an additional $400 in State Supplemental Payments (California Registry, 2004). Considering that base monthly rates may average $2,524 nationally, SSI payments, even when supplemented with optional state payments where available, may be inadequate to cover room, board, and service costs for residents with moderate to high service needs.

The two major options used by states to purchase residential care services using federally matched Medicaid dollars are the Medicaid 1915(c) HCBS waiver program and the optional state plan personal care services benefit (Kitchener et al., 2005). Unlike the personal care benefit, 1915(c) waivers allow states to: (a) specify how many individuals the program will serve and maintain waiting lists when that cap is reached; (b) use more restrictive, functional eligibility criteria; and (c) use less restrictive income criteria, such as 300% of SSI (Smith et al., 2000). Functional eligibility requirements for HCBS waiver programs specify the applicable service-level and population-targeting criteria. Individuals must be certified as requiring a level of care provided by a hospital, nursing facility, or an intermediate care facility for people with mental retardation (ICF-MR). The costs of providing services in each of these institutional settings are used to determine the HCBS waiver program's cost-effectiveness (Smith et al., 2000). States may also use HCBS waivers to specify target populations, such as those who are aged, have a physical disability, have traumatic brain injuries, or have mental retardation or developmental disability (MRDD; Kitchener, Ng, & Harrington, 2003).

States may also use the personal care services benefit to purchase covered services for residential care facility residents. Those states that offer the personal care benefit may not restrict the number of eligible individuals, although income criteria may be more restrictive because the income threshold is much lower than for the HCBS 1915(c) waiver program (Kitchener, Ng, & Harrington, 2004). As a personal care benefit, individuals are not required to meet the institutional level of care admission criteria, and states are not required by federal statute to demonstrate cost effectiveness in relation to institutional care spending (Smith et al., 2000).

Although federal guidelines provide working definitions for residential care services under HCBS waivers, states have considerable flexibility in defining the eligible settings and the covered range of services (Smith et al., 2000). HCBS waiver applications allow states to select from two predefined residential care service categories, "adult foster care" and "assisted living." Both include the provision of personal care services, homemaker, chore, attendant care, companion services, and medication oversight. Under the assisted living benefit, states may specify other optional services (e.g. medication administration, skilled nursing, transportation, etc.). However, states have a third option to define their own residential care benefit and corresponding range of covered services. Such policy choices may be determined by regulatory or statutory limitations, local supply characteristics, or the relative influence of various interest groups (O'Keeffe et al., 2003). For example, in some states, contracted residential care facilities may be required to provide limited nursing services that might not be required or permitted in other states or in smaller group homes within the same state (Mollica & Johnson-Lamarche, 2005). Medicaid payments may not be used in any state to cover residential care room-and-board costs and other facility-maintenance costs. Such costs are paid through the resident's personal resources with additional state supplementation where available.

While previous studies have reported the growth in national waiver participants and expenditures (Kitchener, Ng, et al., 2003), little is known about trends in the provision of waiver services to persons living in residential care facilities. This article focuses on state Medicaid programs providing coverage for residential care as a fee-for-service benefit under a 1915(c) HCBS waiver program. We did not consider personal care services under the state optional plans because state reporting procedures for personal care services do not provide reliable estimates that allow distinctions by population categories or settings (Mollica, 2002). Residential care facility participant and expenditure data for individuals enrolled in Medicaid 1115 demonstration programs and 1915(b) managed care initiatives were also unavailable. For example, the Arizona Long-Term Care System (ALTCS) operates under a Medicaid 1115 waiver and contracts with residential care facilities, which served approximately 3,076 enrolled residents in 2004 (Mollica & Johnson-Lamarche, 2005).


    Methods
 TOP
 Abstract
 Methods
 Results
 Discussion and Conclusions
 References
 
Data Sources and Sample
This study conducted descriptive analyses of a unique dataset compiled from CMS Form 372 waiver reports for 1995 to 2002.

CMS Form 372
Since 1994, the authors have collected annually, from state officials, the CMS Form 372s that report unduplicated participant and expenditure data for each waiver program (N = 252 in 2002). In 2004, as in previous years, we made between three and five requests to every state Medicaid program in order to collect the 2002 reports, plus any missing from previous years. We considered waivers for inclusion in this analysis if they provide residential care or foster care as a service. States use a variety of terms to describe such services in CMS Form 372 reports. These may differ from licensing categories provided in state assisted living and residential care policy reports (Mollica & Johnson-Lamarche, 2005). In addition to including data for service categories and waiver programs that matched these lists, we included several terms that clearly referred to services provided in a congregate setting (e.g. staffed residence, residential group home, etc.). We also reviewed Medicaid provider manuals, waiver program information, and 1915(c) waiver applications available on state web sites for descriptions of more ambiguous terms (e.g., supported living, alternate living).

We included data for those waivers in which the service category was described as being provided in a group residential setting rather than to individuals residing in their own homes. This process left us confident that the sample comprised all HCBS waivers providing services in residential care facilities (N = 65 in 2002). By including residential care services provided to nonaged populations or in smaller group home settings, this study includes program data excluded in other studies (Mollica & Johnson-Lamarche, 2005). Although the dataset includes pre-1995 residential care waiver data, this analysis begins in 1995 because prior years included few (35 or less) residential care waivers. For missing CMS Form 372 reports for residential care waivers (three in 2002), we estimated data from the previous years' data (participants were held constant and expenditures were increased by consumer price index [CPI]).

Analysis
We coded data from the CMS Form 372 reports using a standardized protocol and entered it into an Excel database that allowed us to conduct four descriptive analyses. First, we produced three sets of state and national program data concerning the residential care waiver programs for the period 1995 to 2002: (a) participants and expenditures by program, (b) participants per 1,000 population, and (c) CPI-adjusted expenditures per participant. It should be noted that residential care provided in waivers did not cover room and board. Second, we identified participant and expenditure data by three target groups (MRDD, aged or disabled, and other) and level of care (nursing home, ICF-MR, hospital).


    Results
 TOP
 Abstract
 Methods
 Results
 Discussion and Conclusions
 References
 
Residential Care Services in Waivers: Program Trends 1995–2002
Nationally, residential care participants and expenditures in Medicaid 1915(c) waiver programs grew consistently between 1995 and 2002 (see Table 1). While the total number of HCBS-waiver programs increased by 33% (Kitchener et al., 2005), those offering residential care services increased by 71%. As a result, the proportion of waiver programs providing services in residential care facilities increased from 20% in 1995 to 26% in 2002. Coverage for residential care services increased at a higher rate for waiver programs using nursing facility level of care criteria compared to those using an ICF-MR or hospital level of care. In 2002, only a very small number of waivers (three) offered residential care using the hospital level of care, possibly due to state licensing requirements that generally restrict admitting or retaining residents with acute health conditions (Mollica, 2002).


View this table:
[in this window]
[in a new window]
 
Table 1. Residential Care Services in Waivers: Summary Program Trends 1995–2002.

 
Between 1995 and 2002, the number of residential care facility residents served by Medicaid waiver programs increased by 176%. Annual participant growth rates peaked at 23% between 1999 and 2000, but slowed to 10% between 2001 and 2002. Adjusting for inflation, annual residential care waiver expenditure growth rates have climbed steadily, peaking at 34% between 1998 and 1999. Annual growth was more modest in subsequent years, with a rate of 15% between 2001 and 2002.

Residential Care Services in Waivers, Standardized Program Trends: 1995–2002
Adjusting for population growth and inflation over the study period, Figure 1 compares annual national residential care waiver participants per 1,000 U.S. population with CPI-adjusted expenditures per waiver participant. The number of waiver participants per 1,000 population grew by 156% during the study period with the highest increase of 22% between 1999 and 2000. Annual growth rates for CPI-adjusted expenditures (i.e., in 2002 dollars) per participant increased in most years except for a 4% decline between 1996 and 1997 and a 7% decline between 2000 and 2001. By 2002, CPI-adjusted expenditures per participant of $19,369 were still slightly below the peak of $19,874 in 2000 (see also Table 1).


Figure 01
View larger version (25K):
[in this window]
[in a new window]
 
Figure 1. Residential care services in waivers, standardized program trends, 1995–2002. RC = residential care; CPI = consumer price index. The data presented are the authors' analysis of CMS Form 372 waiver data. CPI-adjusted expenditures per participant are reported in constant 2002 dollars. Population figures are based on U.S. population excluding Arizona

 
Residential Care Services in Waivers by Target Group: 2002
Expenditures for Medicaid waiver services in residential care facilities vary greatly by targeted population groups. In 2002, individuals in MRDD waivers providing residential care represented 48% of the 119,894 residential care service recipients and accounted for 76% of the $2.3 billion in waiver expenditures (Figure 2), or $30,136 per MRDD participant. In contrast, 51% of residential care waiver participants were in aged or disabled programs while their expenditures represented only 23% of all waiver residential care service expenditures, or $8,925 per aged or disabled participant. While 30 states provided Medicaid reimbursement for residential care services in waiver programs targeting aged or disabled adults, only 15 states targeted MRDD participants. Additionally, 1,621 participants in waivers providing residential care (1%) were in programs targeting persons with AIDS (1 waiver), children (1 waiver), mental health problems (1 waiver), and traumatic brain injury or spinal cord injury (7 waivers). For those target populations combined, the average annual waiver cost was $21,169 per participant.


Figure 02
View larger version (20K):
[in this window]
[in a new window]
 
Figure 2. Residential care services in waivers, program data by target group, 2002. The data presented are the authors' analysis of CMS Form 372 data. Other = AIDS (n = 1), children (n = 1), mental health (n = 1), and traumatic brain injury or spinal cord injury (n = 7) target group waivers

 
Residential Care Services in Waivers, Level of Care and State Variation: 2002
Comparing population-adjusted participation and expenditures by types of waivers reveals considerable variation between states and levels of care. Among the 65 waiver programs in 38 states that provided residential care services in 2002, there were 0.42 participants per 1,000 U.S. population with per-participant expenditures of $19,369 (See Table 2). Among states providing residential care services under a nursing facility level of care waiver, participants represented 0.22 per 1,000 U.S. population with expenditures of $9,153 per participant. For waivers using the ICF-MR or hospital levels of care, residential care participants represented 0.21 per 1,000 U.S. population with per-participant expenditures of $30,196.


View this table:
[in this window]
[in a new window]
 
Table 2. Participant and Expenditure Rankings for States Offering Waivers With Residential Care, 2002.

 
State rankings varied when comparing programs by levels of care, population-adjusted participants, and expenditures. Among nursing facility level of care waivers, the highest ranked state for residential care expenditures per participant was Kentucky ($30,736), and the lowest ranked state was Nevada ($3,236). When ranking nursing facility waiver programs providing residential care by the participants per 1,000 population, Oregon (4.68) was highest, and Kentucky was lowest (0.004). Among waiver programs using an ICF-MR or hospital level of care, Massachusetts ($58,995) had the highest residential care waiver expenditures per participant, while Missouri ($2,619) had the lowest. For ICF-MR and hospital waivers providing residential care, South Dakota (2.41) had the highest number of waiver participants per 1,000 population while South Carolina had the lowest (0.001).


    Discussion and Conclusions
 TOP
 Abstract
 Methods
 Results
 Discussion and Conclusions
 References
 
This article presents a comprehensive analysis of national residential care participation and expenditure trends across Medicaid 1915(c) waiver programs during an 8-year period while examining variation by state, target population, and level of care. Despite the growing number of states with Medicaid HCBS services provided to residential care facility residents and increasing participant and spending rates, this study presents three sets of findings that may interest policy makers and suggests several areas for further research.

First, although 1915(c) HCBS waivers are the most popular Medicaid funding mechanism used by states to purchase residential care services, waiver program participants remain underrepresented in residential care facilities. In 2002, the 119,894 waiver participants represented less than 12% of the licensed residential care facility bed supply. By comparison, a much larger proportion of nursing home residents (59%) rely on Medicaid as a primary source of payment (Jones, 2002). Total Medicaid residential care participation rates may be somewhat higher with the inclusion of residents receiving services through optional state-plan personal-care-services programs and capitated long-term-care programs such as the ALTCS (Mollica & Johnson-Lamarche, 2005). However, comparable data are not available for these programs.

Possible explanations for the relatively low proportion of Medicaid-supported residential care facility residents nationwide include the lack of available funding, low reimbursement levels, and lack of Medicaid coverage for residential care (Spillman et al., 2002). Interviews and policy discussions suggest that residential care providers and developers may not be attracted to Medicaid reimbursement due to the possibility of insufficient payment, frozen payment levels over time, reduced profit margins, concerns about increased regulatory oversight, and higher client-service costs (Mollica, 2002; O'Keeffe et al., 2003). Other possible barriers to expanding Medicaid-waiver-funded residential care facility programs include state budget pressures, inadequate room-and-board coverage, and restrictive financial eligibility criteria in some states (Mollica; O'Keeffe et al.). Waiver program size is also limited by enrollment caps, and this produces waiting lists in some states (Kitchener, Ng, et al., 2004). More focused state studies might examine Medicaid payments in relation to residential care private-market rates, participant waiting lists in waivers with residential care services, and prevailing room-and-board costs to identify programmatic or policy adjustments needed to improve provider and resident participation in particular states.

Second, this study confirms that considerable interstate variation persists in Medicaid service provision to residential care facility residents among the growing number of states using HCBS waivers to pay for such services. Further research is required to examine Medicaid waiver program policies and state licensing factors that may contribute to interstate variation in residential care participants and expenditure trends. Although beyond the scope of this study, future work should examine the effect of reimbursement rates and licensing requirements that may restrict access to residential care facility services. Other studies have identified several supply-and-demand, state policy, state resource, and political measures with significant associations and positive predictive values for HCBS-waiver participation and expenditures (Kitchener, Carrillo, & Harrington, 2002; Miller, Harrington, & Goldstein, 2002; Miller, Harrington, Ramsland, & Goldstein, 2002).

Third, this study indicates that state policy choices favor higher amounts of residential care spending for MRDD groups. Of the 38 states that provide residential care services in Medicaid waiver programs, four out of five did so for the aged or disabled, while less than two in five provided such coverage for the MRDD population. However, residential care waiver expenditures per participant were more than three times higher for MRDD waivers than for aged or disabled waivers. This likely results from a combination of factors including the more intensive staffing levels needed to serve MRDD clients in residential care facilities, the higher institutional rate that is used to measure the cost-effectiveness of MRDD waivers, and the political strength of the MRDD lobby compared with the elder advocates. Future studies are needed to better understand the cost differences for comparable residential care services provided to different populations, as well as the relationship of reimbursement rates to access and quality.

Finally, some reports have raised questions about the quality of care and the adequacy of state oversight for Medicaid HCBS waiver programs (General Accountability Office [GAO], 2003) and residential care facilities for the aged or disabled population (GAO, 1997, 1999; Wunderlich & Kohler, 2001). Yet little is known about the quality of care in waiver programs and residential care facilities in general. Studies of quality should be a focus for future research along with the relationship of Medicaid payment rates to quality of care.


    Footnotes
 
This research was funded in part by the Kaiser Commission on Medicaid and the Uninsured (Grant # 00-1355C) and the National Institute for Disability and Rehabilitation Research (Grant # H133B031102-04). The views expressed in this article are those of the authors and do not necessarily reflect those of either sponsor. Back

1 Department of Social and Behavioral Sciences, University of California, San Francisco. Back

Decision Editor: Linda S. Noelker, PhD

Received for publication April 29, 2005. Accepted for publication August 15, 2005.


    References
 TOP
 Abstract
 Methods
 Results
 Discussion and Conclusions
 References
 




This article has been cited by other articles:


Home page
Journal of Health Politics, Policy and LawHome page
E. A. Miller and V. Mor
Balancing Regulatory Controls and Incentives: Toward Smarter and More Transparent Oversight in Long-Term Care
Journal of Health Politics Policy and Law, January 1, 2008; 33(2): 249 - 279.
[Abstract] [PDF]


This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
PubMed
Right arrow PubMed Citation


HOME ARCHIVE SEARCH TABLE OF CONTENTS