Printer-friendly page

Home and Community-Based PAS > Reports > Medicaid Home and Community-Based Services: Program Data, 1992-2001

Medicaid Home and Community-Based Services: Program Data, 1992-2001

Part 1

(a) Medicaid HCBS Programs

For most of the 20th century, the bulk of formal (paid) long-term care (LTC) was provided in institutions such as nursing homes. Little use was made of HCBS such as personal care, home delivered meals and independent living.¹ Still, in the 21st, the only Federal Medicaid LTC benefits required of states are those provided by institutions, and HH services provided to those who would otherwise be in an institution.^1-5 Medicaid regulations do, however, allow states the option of using combinations of three programs (waivers, HH, and PCS) to provide HCBS either directly, or through a variety of contractual arrangements. Figure 2 compares key elements of the three programs and each program is introduced briefly below.

Figure 2: Summary Comparison of Medicaid HCBS Programs²

Medicaid 1915(c) HCBS Waivers

Since Congressional authorization in 1981, the Medicaid 1915(c) HCBS waiver program offers to the states, federal matched funding to expand HCBS and accelerate movement away from LTC services provided from institutions.^2-5 This optional program allows the Department of Health and Human Services to ‘waive’ certain Medicaid statutory requirements so that states can receive federal funds to expand HCBS and reduce existing institutional care in three main ways. First, states can target HCBS towards specific geographical areas (i.e. a county), populations (e.g. the elderly), and conditions (e.g. traumatic brain injury). Second, they can provide HCBS services not otherwise covered by the Medicaid program (e.g. respite care, home modifications, case management, homemaker services, personal care, and adult day care). Third, states can set a limit on the numbers of waiver ‘slots’ available in order to control program costs.^2-5 As a Medicaid program, the 1915(c) waiver initiative encourages states to expand and target the provision of HCBS services toward the elderly and those with disabilities who would otherwise be in an institution.^3,4 By l999, all states, except Arizona had one or more 1915(c) waiver programs for LTC services. Arizona operates its Medicaid LTC program under a capitation arrangement using an 1115 waiver.⁴

Medicaid Personal Care Services (PCS)

Since 1975, states have the option (through Medicaid Title XIX) of offering PCS as a Medicaid state plan optional benefit.⁶ States have considerable discretion in defining PCS but the programs typically involve hands-on assistance with activities of daily living (e.g., bathing and eating) to non-institutional residents of all ages with disabilities and chronic conditions. In 2001, 28 states offered a PCS benefit but they varied considerably in the amount and scope of the services provided. States can provide PCS in the home, outside the home, or in both locations. They cannot, however, reimburse legally responsible relatives (typically parents of disabled minor children and spouses). While the PCS optional benefit must be statewide and available to all Medicaid categorically eligibility groups, states can opt to make it available to the medically needy (who spend down to the state standard because of medical expenses). In contrast to the waiver program, PCS state optional benefits cannot use the nursing home eligibility standard for the categorically needy.^6,7

Medicaid Home Health (HH)

The Medicaid program requires that HH services be offered as a mandatory benefit to those individuals entitled to nursing facility care under the state’s Medicaid plan.^7,8 For other groups, Medicaid HH services are optional and states can vary the amount, scope and duration of benefits offered (so long as they remain sufficient to reasonably achieve their purpose and are the same for all eligibility groups). While some advocates argue that access to Medicaid HH services is overly restrictive in some states, there is little information about the extent of these services, their impact, or the recipients.⁸

(b) The Current Policy Context of Medicaid HCBS

Against a backdrop of the Bush Administration’s plans to reform Medicaid and 43 states reporting budget deficits in 2003,⁹ all states face three major pressures to expand their Medicaid HCBS programs. First, the public increasingly displays a preference for LTC provided at home rather than from institutions.¹⁰ Second, mounting concerns regarding the quality of institutional LTC add further pressures on state governments to extend HCBS.¹¹ Third, extended Medicaid HCBS are anticipated by legislation such as the 1990 Americans with Disability Act (ADA) and legal judgments including the 1999 Olmstead Supreme Court decision. The importance of HCBS increased with the passage of the ADA because it outlawed certain practices of private and public entities that unreasonably restrain the participation of individuals with disabilities in society.^6-8 The Olmstead ruling requires that states must not discriminate against persons with disabilities by refusing to provide community services when these are available and appropriate.¹² In summary, states are caught between the rock of increasing requirements to provide Medicaid HCBS, and the hard place of fiscal crisis.

(c) Data Sources

While it was known that Medicaid HCBS spending was increasing, less was know about the participants, expenditures and services in waiver, HH and PCS programs across the states.^13-19 Three sources of data are combined in this descriptive longitudinal analysis of Medicaid HCBS programs: (1) Centers for Medicaid and Medicare Services (CMS [formally HCFA]) Form 372 reports for waivers, (2) research surveys of PCS and HH, and (3) CMS Form 64.

(1) CMS Form 372. Since 1994, a University of California, San Francisco (UCSF) research team has collected annually, the CMS Form 372 on which states report participant and expenditure data for each of their waivers. Although Form 372s were submitted to CMS, they were not compiled and computerized by CMS. Thus, original Form 372 data had to be collected directly from states. Each year, between 3-5 telephone and fax calls were made by study researchers, to every state Medicaid program in order to collect the 372 reports for the previous year and any outstanding reports (the 1994 effort collected reports from 1992 onward). If the final ‘lag’ reports were unavailable, state officials were asked to provide interim ‘initial’ reports or estimated data. The data are coded and entered into a SAS database.

As in previous years, five issues emerged during the collection of Form 372 data. First, there were some reporting discrepancies between expenditures by service and total waiver expenditures. These were rectified by a telephone call to the responsible state official. Second, the classification of waivers by recipient group was not always straightforward because some states issue waivers titles that do not indicate the name of the recipient group. In order to classify the waiver by recipient group, we matched each waiver number with the name of the recipient group that the state reported to CMS in their initial waiver request. Then, we categorized the recipient groups into eight categories.

Third, reporting of waiver service categories are determined by each state. Participants by service are duplicated since most participants receive more than one category of service. Since waivers may offer many different services and because states do not have common names for these services, coding the service data is a challenge. A list of 30 separate service codes was compiled from a review of the data. Then waiver services were coded in the category that appeared to be most similar to the coding category. A miscellaneous category is used for those services not covered by the categories. The 30 service categories were collapsed into six categories to present in this report.

Fourth, when states did not provide a Form 372 for a waiver (24 cases in 2001), total expenditure and recipient figures (but not service level data) were estimated by the researchers using linear interpolation from previous years’ data. Tables in Part 2 indicate all estimated data with an asterisk. Because no attempt was made to estimate service-level data within the 24 waivers for which no 372 report was available, these waivers are not included in Tables 7-9 reporting service category data for 2001.

Fifth, it is recognized that discrepancies persist between waiver data reported on Form 372 and the Form 64 quarterly financial returns that states submit to CMS.²⁰ This study uses Form 372 data for three reasons: (a) they are compiled and reviewed by state program staff before submission to CMS, (b) Form 64 returns are compiled by state staff to request reimbursement from CMS, and (c) Form 372 is the only source of participant data for the waiver program.

(2) UCSF Research Survey of PCS and HH. Until 1998, state Medicaid programs reported HH and PCS expenditures (not participants) on CMS Form 2082. Since then, states report the expenditures (plus participants) separately but local Medicaid officials reported to us that these data suffer from problems of accuracy and consistency. Thus, primary data regarding HH and PCS participants and expenditures for 1999-2001 were collected directly by the investigators from states through annual telephone and fax surveys conducted in 2001-2003. Where such data were not available, we rely on Form 2082 data and Form 64 expenditure data²⁰ as the best available. In such cases, data are marked with an asterisk in the reported tables. The HH and PCS data from these sources are coded and entered into a SAS database.