| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||
Correspondence: Address correspondence to Mauro Hernandez, PhD, Department of Social and Behavioral Sciences, University of California, San Francisco, 3333 California Street, Suite 455, San Francisco, CA 94118. E-mail: Mauro.Hernandez{at}ucsf.edu
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
Key Words: Medicaid • Home- and community-based services • Waivers • Long-term care • Adult foster care • Nursing homes
Throughout the 1990s, Oregon's AL model received considerable attention in the mainstream media (McCarthy, 1992), gerontological textbooks (Wilson, 1993), research reports (Kane & Wilson, 1993), and state policy reports (Mollica, Ladd, Dietsche, Wilson, & Ryther, 1992). Recent Oregon studies have examined outcome trajectories and placement preferences for AL and nursing facility residents (Frytak, Kane, Finch, Kane, & Maude-Griffin, 2001; Reinardy & Kane, 2003), how Medicaid dollars are used to pay for AL services (O'Keeffe, O'Keeffe, & Bernard, 2003), and how providers incorporate and negotiate AL values (e.g., independence, choice, and privacy) into daily practice and marketing materials (Carder, 2002a, 2002b; Carder & Hernandez, 2004). The aim of this article is to describe Oregon state policy and supply developments among licensed group long-term-care settings, as well as use by Medicaid clients. The article focuses on apartment-style AL facilities (ALFs) and more traditional residential care facilities (RCFs) while including selected findings regarding nursing facilities and smaller adult foster homes.
Licensed Settings in Oregon
The Oregon Department of Human Services, Division of Seniors and Persons with Disabilities, currently licenses four long-term-care settings. Adult foster homes are private residences or purpose-built dwellings licensed to provide care for up to five residents. (Excluded for purposes of this study are relative adult foster homes, which may be considered a form of in-home care. In such arrangements, a resident lives in the home of and receives services from a Medicaid-contracted, nonspouse relative. Such settings typically serve no more than one resident who must be Medicaid eligible.) ALFs and RCFs serve six or more residents and currently provide a similar range of personal care and health-related services but with different physical design requirements. Nursing facilities provide nursing care on a 24-hr basis in institutions that meet requirements for Medicare and Medicaid nursing homes.
State Policies
As noted above, several reports provide extensive reviews of Oregon's long-term-care policy environment, including current ALF and RCF regulatory and reimbursement policies (Mollica & Johnson-Lamarche, 2005). Following are summaries of selected policies relevant to the scope of this study.
Financing
The Oregon Housing and Community Services Department (OHCS) administers various programs that finance low-income housing projects. The Department's Elderly and Disabled Loan Program provides loans for projects that can include senior independent apartments, congregate care, RCFs, and ALFs. These loans provide below-market interest rates and are financed through the issuance of tax-exempt general obligation bonds (OHCS, 2003). Eligible projects must be multi-unit housing with apartment-style individual units that are either new construction or acquisitions with rehabilitation. Low-income set-aside requirements include designating either (a) 20% of units for residents at or below 50% of the area median income or (b) 40% of units for residents at or below 60% of the area median income.
Medicaid Reimbursement
As the first state to obtain a 1915(c) home- and community-based services waiver from the federal government in 1981, Oregon uses Medicaid dollars to subsidize residential care service costs for nursing-home-eligible residents. Monthly payments include all services provided directly by the licensed residential care provider, excluding room and board costs.
Reimbursement policies vary by provider type. ALF payments have been tiered by resident impairment level since 1990, when the highest service level was set at about 75% of the lowest nursing facility payment. The state's goal was to provide greater financial incentives for ALFs to accept and retain more impaired residents, particularly as their needs increased. Since 1996, there have been several unsuccessful attempts to reduce ALF service payments in order to increase parity with other residential settings and, in recent years, to reduce state Medicaid expenditures. Compared to ALFs, adult foster home rates have been tiered by resident need since 1991 but using different methodologies for determining service levels. RCF payments were based on licensed bed capacity until 1995, when RCFs were brought under the adult foster home reimbursement system. Payment rates and service level determination methods for adult foster homes and RCFs were substantively modified in 1998 and 2002. These changes occurred partly in response to changes in regulatory requirements and recognition of increasing resident frailty levels in these settings. Another notable policy change has been the use of special contract rates that could be negotiated with individual providers since the late 1990s. This has allowed providers serving specified populations—typically RCFs with designated Alzheimer's care units (ACUs)—to negotiate a single payment rate equivalent to the highest ALF rate.
ALF and RCF Licensing
ALF licensing rules became effective in 1990, authorized under preexisting RCF statutes. Regulations established distinct criteria for personal living spaces (e.g., private apartments with kitchenettes, roll-in shower, lockable door); higher service capacity than typically provided in RCFs at the time (e.g., incontinence, behavior management, nursing services); and a philosophical orientation emphasizing privacy, choice, and aging in place, among other values (Wilson, 1990). Substantial revisions became effective in 1999, and other incremental changes have been made since then. RCF statutes and corresponding rules were first adopted in 1977 when these settings were known as Homes for the Aging. Major revisions took place in 1994 and 2002 that responded to concerns about rising impairment levels and the desire to make RCF standards more consistent with updated ALF rules. Such changes included adopting similar language regarding philosophical orientation, minimum service requirements, service delivery approaches, and contract requirements. Currently, ALFs are primarily distinguished from RCFs by the private apartment requirements noted above. A moratorium on issuing licenses for new ALF and RCFs took effect in 2001.
|
Methods |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
Developing a historical inventory of licensed long-term-care settings in Oregon involved using multiple sources to accurately identify changes in licensed bed capacity, as well as initial licensing and closure dates. There was no single data source that contained complete and accurate historical supply data for all licensed settings during the study period. Sources included electronic files, facility lists, onsite file reviews, and historical records obtained primarily from state licensing officials and other state agencies. I used standard protocols to ensure the accuracy of any estimates for missing data. Procedures varied by setting type but included cross-checking information from multiple sources, consulting with state agency staff, and contacting facilities directly, as described in greater detail elsewhere (Hernandez, 2006).
For example, three ALFs in the state licensing agency's database had initial licensing years that preceded the adoption of ALF licensing regulations. According to agency staff, these facilities were initially licensed as RCFs. Therefore, I created dummy RCF records for these facilities using the recorded initial licensing year and a closure (i.e., RCF to ALF conversion) year of 1989. Because ALF licensing rules did not become effective until 1990, this provided a more accurate count of Oregon's RCF supply prior to 1990. I then adjusted the initial licensing year for these three ALFs to 1990. Because the licensing agency's database maintains current but not historical bed capacity data, I reviewed administrative files on site for all currently open ALFs to identify when changes in size occurred as recorded on copies of licenses or other file documentation. When examining ALF and RCF Medicaid contracting trends, I used initial licensing dates because actual contracting dates were not available. The working assumption was that providers had not changed their Medicaid contracting status since initial licensing based on state agency staff reports that such changes were rare among ALFs and RCFs during the study period.
I aggregated individual facility data into a separate county-level file for examining population-adjusted rural ALF and RCF supply availability. I coded counties as metropolitan or nonmetropolitan based on U.S. Department of Agriculture Rural–Urban Continuum Codes for Metro and Nonmetro Counties from 1995. I standardized beds by the county population aged 65 or older in 1,000s. The study excluded one county (Sherman) that had no ALFs, RCFs, or nursing facilities throughout the study period. I obtained older population data from the U.S. Census Bureau and Portland State University's Population Research Center.
Adult foster home supply data came from periodic reports compiled by the licensing agency's central office from its local offices showing estimated total homes and beds from 1991 to 2004. I made estimates for previous years using figures reported by Kane, Illston, Kane, and Nyman (1990) for the number of adult foster home beds in 1990 and homes in 1988.
Other secondary data sources included loan information for projects financed by Oregon's Elderly and Disabled Loan Program, Medicaid reimbursement rates obtained from the state's Medicaid agency, and previously published nursing facility rates (Swan, Harrington, Grant, Luehrs, & Preston, 1993). I estimated adult foster home and RCF payment rates for 1986 using the average annual increase from 1987 to 1990. Medicaid caseload data came from Seniors and Persons with Disabilities, which administers Oregon's Medicaid program.
|
Results |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
|
Medicaid Reimbursement
Provider payments varied by licensing category and over time, reflecting reimbursement policy changes for each of these settings. As shown in Figure 2, ALF rates were set relatively high upon program inception and received steady, gradual increases in most years. Adult foster homes and RCFs received relatively low payments initially and experienced marked stepped increases following reimbursement policy changes. As a result, payment gaps favoring ALFs narrowed. Nursing facilities received the highest reimbursement throughout the study period, with a widening gap between these and other licensed settings in recent years. By 2004, the daily Medicaid reimbursement rate for nursing facilities was about $140 compared to a maximum of $64 for ALFs and $52 for RCFs and adult foster homes.
|
Oregon's nursing facility bed supply decreased steadily throughout the study period from about 15,500 beds at the end of 1986 to 12,600 beds at the end of 2004 (Figure 3). During this period, 67 organizations ceased to operate as licensed nursing facilities and only 18 new facilities were established.
|
ALF and RCF supply growth seemed to level in the most recent years despite an extensive list of proposed projects that were exempted from Oregon's 2001 moratorium policy. State agency records showed that, in the weeks prior to the policy's adoption, 76 ALFs with more than 3,000 total units and 62 RCFs with more than 1,300 total beds submitted plans for development. Most of these facilities had not begun construction by the end of 2004. Key informants reported shared concerns about excess supply conditions in most Oregon markets.
Rural ALF and RCF Supply
The ALF sector had a larger proportion of providers in nonmetropolitan counties (41%) compared to RCFs (30%). This was partly due to early ALF development efforts favoring nonmetro areas. By 1995, the majority of ALFs (54%) were located in nonmetro counties. Interviews with providers and developers suggested that such markets were perceived to have fewer market entries and development barriers. Reported advantages included lower land and labor costs, a lack of desirable supportive housing, and no other ALFs. By comparison, the proportion of nonmetro RCFs changed little over time.
Adjusting for population size, the county-level ALF bed supply was higher and favored nonmetro areas, whereas RCF bed supply was lower, favoring metro areas. In 2004, there were about 33 ALF beds per 1,000 older adults in nonmetro counties compared to 30 in metro areas. The population-adjusted RCF bed supply was higher in metro (19 beds per 1,000 older adults) than nonmetro areas (13 beds per 1,000 older adults).
ALF and RCF Medicaid Participation
In 2004, more than 2 in 3 (69%) RCFs were contracted Medicaid providers compared to most ALFs (89%) and nursing facilities (93%). ALFs licensed since 2001 were less likely to be Medicaid providers (82%) than those licensed in earlier years (92%). Fewer than half of the 35 RCFs licensed since 2002 (49%) served Medicaid residents. Assuming no change in contracting status for individual RCFs, participation rates were likely higher in 1995 (77%) and even higher in 1986 (84%). Location may have influenced Medicaid contracting decisions for both provider types. ALFs and RCFs located in metropolitan counties were much more likely to forgo Medicaid contracting (17% and 37%, respectively) than those in nonmetropolitan counties (3% and 17%, respectively).
ALF and RCF Specialization
Organizations choosing to develop a specialized ACU typically used an RCF license rather than an ALF license partly because of regulatory and reimbursement policies. In 2004, a much larger proportion of RCFs (38%) had designated ACUs than ALFs (1%). Some specialized RCFs were reportedly purpose built in recent years, whereas other ALFs and RCFs chose to specialize several years after initial licensing by designating part or all of the facility. Nevertheless, RCFs that were licensed in more recent years were more likely to have a designated ACU. A majority (52%) of RCFs licensed since 2000 had an ACU designation, compared to less than a third (30%) of those licensed in prior years. ALFs choosing to specialize in this way developed an RCF-licensed ACU on the same or adjacent property, either during initial construction (n = 27), as a later addition (n = 9), or through RCF conversion (n = 1). Interviewees reported that ACU operators preferred the RCF licensing category because ALF licensing requirements for single occupancy units, private bathrooms and kitchenettes were neither cost-effective nor desirable for residents with dementia. Another reported motivating factor was the ability of ACUs to negotiate higher Medicaid contract rates than normally available to RCFs.
Medicaid Caseloads in Licensed Settings
Between 1990 and 2004, the number of Medicaid clients served in licensed long-term-care settings remained stable relative to growth in the older population, whereas monthly caseload distribution shifted in favor of noninstitutional settings. As shown in Table 1, nursing facilities represented the dominant licensed setting, followed by adult foster homes.
|
|
Discussion |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
First, the study identified state policy differences for residential care settings that created favorable conditions for ALF supply growth. More generous reimbursement rates may have attracted more providers to become Medicaid contractors and serve eligible clients while also providing greater resources for organizational growth and survival. Recent policy changes have narrowed reimbursement gaps between settings, recognizing higher resident impairment levels and increasingly similar service capacity across provider types. The state's loan program also facilitated early ALF development at a time when conventional lenders were more reluctant to finance new projects. Requirements for full apartments and low-income set-aside units combined with favorable Medicaid reimbursement rates steered most loan applicants to the ALF licensing category. Licensing policies may have also favored ALF development. Adopting separate regulations allowed ALFs to differentiate themselves from RCFs, which were then considered a less favorable model of care. With more consumer-friendly language, design features, and services, these new rules may have provided a greater level of external legitimacy for attracting new residents, their families, and investors. Further comparative state studies are needed to evaluate the relationship of state policies and supply developments between and within states.
Second, supply trends also differed across residential care categories. The smaller adult foster homes were the dominant setting in most years; however, adult foster home bed supply has contracted somewhat since peaking in 1995. Speculation about possible causes includes higher operating costs attributed to more impaired residents and licensing requirements, inadequate payment rates, and growing competition from ALFs. These and other conditions may adversely impact the long-term survival of this smaller form of residential care (Ball et al., 2005; Morgan, Eckert, Gruber-Baldini, & Zimmerman, 2004). Compared to the ALF sector, RCF supply grew less rapidly, partly due to higher closure rates throughout the study period. RCFs also tend to be much smaller than ALFs—37 versus 66 licensed beds for the latter. Interviewees attributed recent RCF resurgence to the development of formalized standards for ACUs and more favorable reimbursement rates for such specialized providers.
Third, the population-adjusted supply of ALF beds favored nonmetropolitan counties, whereas RCF bed supply favored metropolitan counties. This contrasts with earlier findings from a national study that suggested an undersupply of AL in rural areas, particularly those categorized as high privacy and high service (Hawes, Phillips, Holan, & Sherman, 2003). Greater supply of apartment-style ALFs in Oregon's rural communities may be due to unique state policy and market conditions described above. RCF expansion into rural markets has been more gradual over time. Future studies might examine factors contributing to entry and survival rates for different provider types and communities.
Finally, Medicaid contracting rates for ALFs and RCFs are relatively high in Oregon compared to other states (Mollica & Johnson-Lamarche, 2005). Other studies have found a much smaller proportion of facilities that would either accept (Hawes, Phillips, & Rose, 2000) or retain (Zimmerman, Sloane, & Eckert, 2001) Medicaid residents. Of interest is the fact that the newer apartment-style ALFs are more likely to be Medicaid providers compared to the smaller RCFs that allow shared occupancy. In 2004, Medicaid residents also represented a larger proportion of the available ALF bed supply. There is, however, some indication that more recently opened ALFs and RCFs are choosing not to become Medicaid providers. Further study is needed to determine whether lower participation rates may be related to concerns about the gradual erosion of payment levels that providers report have not kept pace with operating cost increases. Increased uncertainty may also influence recent contracting decisions for newer providers familiar with Oregon's budget deficits and past attempts to cut payment rates.
|
Footnotes |
|---|
1 Department of Social and Behavioral Sciences, University of California, San Francisco.
Decision Editor: Susan L. Hughes, DSW
Received for publication August 10, 2006. Accepted for publication March 16, 2007.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
| ||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
