Â鶹´«Ã½

Data Sources

The major data sources used in this report are Medicaid and Medicare administrative data for calendar year 2005. Medicaid data for the 50 States and the District of Columbia are from the Medicaid Analytic eXtract (MAX) dataset compiled by the Centers for Medicare & Medicaid Services (CMS) from the Medicaid Statistical Information System (MSIS) data (CMS, 2010b). MAX data were used to identify the universe of people in the home and community-based services (HCBS) population, the population of interest for this report (N=2,234,716).

MAX data provided information about the characteristics of the HCBS population and HCBS users (1915(c) waiver and State plan services). The data also included personal attributes of HCBS participants (age, gender, race/ethnicity, managed care enrollment, 1915(c) waiver enrollment, and county and State of residence). MAX data on the HCBS population (minus exclusions as noted below) were also used as outcome indicator denominators in Tables 7-19.

For people eligible for Medicaid only, MAX data on hospital inpatient stays were used to construct the numerator data used in Tables 7-19. For HCBS participants dually eligible for Medicare and Medicaid, the Medicare MedPAR data (CMS, 2010d) on hospital inpatient stays were used to calculate the numerator in the outcome indicators used in Tables 7-19. Both MAX and MedPAR data were used to categorize HCBS participants into the HCBS clinical subpopulations used in this report.

The Medicare Denominator File data (CMS, 2010d) were used to provide information on people who were dually eligible for Medicare and Medicaid. HCBS participants linked to the Medicare Denominator File were designated as dually eligible for Medicare and Medicaid for the entire year. The Medicare Denominator File was also used to check the accuracy of the MAX data on the percentage of dually eligible HCBS participants enrolled in Medicare Advantage Plans. The numbers reported in Table 3 were derived from MAX data only, but the two sources provided effectively the same results.

The CMS report, MAX 2005 Waiver Crosswalk, provided a list of all waivers (1115, 1915(b), 1915(c), and 1915(b/c)) operating in 2005. We selected 1915(c) waivers from this list and used that as the basis for developing our database of 1915(c) waiver characteristics. For each waiver listed, we abstracted information about the qualifying level of care, populations covered, financial eligibility, and covered services from the CMS online database of approved waiver applications (CMS, 2010e).

Information for 160 of 292 waivers was available from the CMS Web site. Information for the remaining 132 waivers was obtained from the University of California-San Francisco Personal Assistance Services Web site (2010) and State web sites. Information from waiver application form version 3.5 or later was standardized. Information from earlier 1915(c) applications on the CMS Web site or from other sources was not standardized, and data extraction depended on careful review and conceptual understanding by the people extracting the data. Frequent communication among those individuals helped to maintain reliability of data extraction. Information from sources other than waiver applications may be incomplete.

Additional data sources were used to describe the service system characteristics (e.g., types of services covered, waiver program characteristics) and context (e.g., supply, area health characteristics, State service characteristics). State-specific information on the availability of various Medicaid State plan HCBS (Tables 1 and 9) was abstracted from the CMS publication "Medicaid-At-a-Glance 2005" (CMS, 2010b).

Other contextual variables were obtained from the Area Resource File (ARF), Census Bureau data sources (e.g., Small Area Income and Poverty Estimates, American Community Survey [ACS]), Centers for Disease Control and Prevention's Behavioral Risk Factor Surveillance System (BRFSS), and other published reports (Burwell, et al., 2006; Grabowski, et al., 2009; Mollica & Reinhard, 2005). All contextual data used in this report were linked to HCBS participants based on the person's county or State of residence as reported on the MAX Person Summary (PS) file. Unless noted otherwise, contextual measures were constructed for calendar year 2005. Variable-specific details are provided in the Definitions section of this appendix.

Definitions

Age. Age of HCBS individuals was based on date of birth (as reported on the MAX PS file) and calculated as of January 1, 2005.

Age 65+, percentage. For purposes of Table 15, the percentage of the population age 65 and older was measured at the county and State level (based on the HCBS participant's place of residence). Source data for the percentage 65+ were from the Census Bureau's Population Estimates (Census County Characteristics File) as of July 1, 2005, and as reported on the 2008 ARF.

Age 85+, percentage. For purposes of Table 15, percentage age 85+ was defined as the percentage of an area's (county and State) population age 85 or older. Source data were from the Census Bureau's 2000 Decennial Census, as reported on the ARF (2008).

African American, percentage. For purposes of Table 15, percentage African American was defined as the percentage of an area's (county and State) population that was identified as non-Hispanic black. Source data were from the Census Bureau's population estimates (Census County Characteristics File) for July 1, 2005 (ARF, 2008).

Difficulty going outside of the home, percentage of State's population 65+ with. Measured with the 2005 ACS (Census Bureau, 2010a). ACS respondents were asked if they had a physical, mental, or emotional condition lasting 6 months or more that made it difficult "going outside the home alone to shop or visit a doctor's office."

Disability, percentage of State's population 18-64 with any. Disability was measured as any of the following six types of disability: sensory disability, physical disability, self-care disability, mental disability, difficulty going outside of the home, or a work disability as measured with the 2005 ACS (Census Bureau, 2010b).

Disability, percentage of State's population 65+ with any. Disability was measured as any of the following five types of disability: sensory disability, physical disability, self-care disability, mental disability, or difficulty going outside of the home as measured with the 2005 ACS (Census Bureau, 2010a).

Dual eligibility status. HCBS participants were classified into two categories: those dually eligible for Medicare and Medicaid and those eligible for just Medicaid. MAX HCBS participants linked to the Medicare Denominator File at any point during 2005 were classified as dually eligible for Medicare and Medicaid. For the outcome indicators, which are based on quarterly denominators, dually eligibly status was defined on a quarterly basis determined by the date of Medicare enrollment during 2005. Individuals were classified as dually eligible for the entire quarter during the quarter of initial enrollment.

Federally qualified health center (FQHC), supply of. FQHCs and rural health centers (RHCs) are "safety net" providers such as community health centers and clinics. FQHC supply was measured as the total number of FQHCs plus the total number of RHCs per capita (State and county level). Source data on safety net providers were for 2005 and from the CMS (ARF, 2008); population data were from the Census Bureau (ARF, 2008).

HCBS population. Using 2005 MAX data, we defined people as being in the HCBS population if at least one of the following conditions was satisfied:

1. The individual was enrolled in a 1915(c) program during the year (or quarter). This was defined as any monthly waiver enrollment flag during the year (or quarter) that indicated the person was enrolled in a 1915(c) for aged and disabled people; aged people only; disabled people only; people with brain injuries; people with HIV/AIDS; people with mental retardation or developmental disabilities; people with mental illness; people who are technology dependent; or people in an unspecified waiver (on the PS file, MAX_ WAIVER_ TYPE_ 1_MO_1 through MAX_WAIVER_TYPE_3_ MO_12 [any of three waivers per month for the year or quarter with a value = G, H, I, J, K, L, M, N, or O]).
2. The individual showed evidence of State plan HCBS use during the year (or quarter). State plan use was defined as Medicaid fee-for-service use of any of the following: personal care (PS file CLTC_FFS_ PAYMT_AMT_11>0), adult day care (PS file CLTC_FFS_ PAYMT_AMT_13>0), home health care exceeding 90 days (total number of home health days [based on SRVC_BGN_DT and SRVC_END_DT] was summed across all records on the OT file where CLTC_FLAG=14), residential care (PS file CLTC_FFS_ PAYMT_AMT_15>0), at-home private duty nursing (OT file CLTC_FLAG=12 and PLC_OF_SRVC_CD =12), or at-home hospice care (OT file CLTC_FLAG=19 and PLC_OF_SRVC_CD=12). Quarterly versions of State plan use were subsequently based on service begin and end dates on the OT file (SRVC_BGN_DT and SRVC_END_DT).
3. The individual showed evidence of use of 1915(c) waiver services during the year (or quarter). As defined in the MAX, 1915(c) waiver services are defined as: personal care, private duty nursing, adult day care, home health care, residential care, rehabilitation, targeted case management, transportation, hospice care, durable medical equipment, or other waiver services not listed. If any of these services had a Medicaid fee-for-service payment amount greater than zero, the individual was defined as a user of 1915(c) waiver services (CLTC_FFS_PAYMT_AMT_30, 31, 32, 33, 34, 35, 36, 37, 38, 39, or 40 are >0). Quarterly versions of waiver service use were subsequently based on service begin and end dates on the OT file (SRVC_BGN_DT and SRVC_END_DT).

For purposes of calculating the outcome indicator denominators used in Tables 7-19, the HCBS population was constructed as previously described on a quarterly basis and exclusions were applied. Exclusions were: (1) individuals younger than age 18; (2) individuals with only institutional use in a given quarter (details below); (3) individuals enrolled in Medicaid managed acute or managed long-term care plan for any month of the year (or quarter) (if any monthly PS file EL_PHP_TYPE_1_1 through EL_PHP_TYPE_4_12 = 01, 05, or 06); (4) individuals dually eligible and enrolled in a comprehensive Medicare managed care plan at any point during the year (or quarter) (on the Medicare Denominator File, Medicare entitlement/buy-in indicator BUYIN01 through BUYIN12 [value >0 for any month] and HMO indicator HMOIND01 through HMOIND12 [value >0 for any month]); and (5) all persons in States with severe data limitations (Arizona, Maine, Wisconsin, and Washington).

People were determined to have only institutional use in a quarter if claims from the MAX Long-Term Care (LT) file at any point during the quarter were found with any of the following values: (1) ICF-MR day count >0; (2) nursing facility day count >0; (3) inpatient psychiatric facility for individuals under 21 day count >0; (4) mental hospital for aged day count >0; or (5) MAX type of service code (MAX_TOS) for any of the following: 02 (mental hospital for aged), 04 (inpatient psychiatric facility for individuals under 21), 05 (ICF-MR), or 07 (nursing facility services) and the person did not have any community-based long-term care claims during the quarter. Quarterly versions of institutional use were based on service begin and end dates on the LT file (SRVC_BGN_DT and SRVC_END_DT).

HCBS (clinical) subpopulations. Using MAX and MedPAR 2005 data, we defined the following four subpopulations:

• Intellectual/developmental disabilities (I/DD). HCBS participants were classified into the I/DD group if any one of four criteria (all measured for 2005) were met: (1) enrollment in an I/DD waiver (if any of the monthly MAX Waiver Type Code variables [MAX_WAIVER_TYPE_1, 2, or 3] on the MAX PS file equals L), (2) use of an intermediate care facility for people with mental retardation (ICF-MR) (on the MAX PS file, persons with positive fee-for-service� payment amount for an ICF-MR [FFS_PYMT_AMT_05 >0]), (3) relevant diagnosis codes on the person's hospital inpatient stay records (for the dual eligible population, MedPAR diagnosis code variables = DGNSCD1through DGNSCD10 ; for the Medicaid-only population, MAX IP diagnosis code variables DIAG_CD_1 through DIAG_CD_9), or (4) relevant diagnosis codes on the person's MAX HCBS OT service records (MAX OT diagnosis code variables DIAG_CD_1 or 2).

  Relevant diagnoses and associated International Classification of Diseases, Ninth Revision (ICD-9) codes were: Down's syndrome (758.0), developmental delay-chromosomal abnormalities (330.0-330.3, 330.8-330.9, 758.1-758.2, 758.31, 758.33, 758.39, 758.7, 759.5, 759.81-759.83, or 759.89), developmental delay-severe brain injury of childhood (768.5, 768.73, 773.4, 774.7, or 779.7), developmental delay-cerebral palsy/epilepsy/spina bifida/physical disabilities (343.2-343.4, 343.8-343.9, 345.00-345.01, 345.10-345.11, 345.2-345.3, 345.40-345.41, 345.50-345.51, 345.60-345.61, 345.70-345.71, 345.80-345.81, 345.90-345.91, 741.00-741.03, 741.90-741.93), developmental delay-fetal alcohol syndrome (760.71), or developmental delay-other major cognitive disabilities (autistic disorder, mental retardation) (299.00-299.01, 299.10, 299.80, 299.90, 314.1, 318.0-318.2, or 319).

• Serious mental illness (SMI). HCBS participants were classified into the SMI group if any one of four criteria (all measured for 2005) were met: (1) enrollment in a mental illness/serious emotional disturbance waiver (if any of the monthly MAX Waiver Type Code variables on the MAX PS file equal M), (2) use of an inpatient psychiatric facility or a mental hospital for the aged (on the MAX PS file, persons with a positive FFS payment amount for either of the two institutions [FFS_PYMT_AMT_02 >0 or FFS_PYMT_AMT_04 >0]), (3) relevant diagnosis codes on the person's hospital inpatient stay records (for the dual eligible population, MedPAR diagnosis code variables DGNSCD1 through DGNSCD10; for the Medicaid-only population, MAX IP record diagnosis code variables DIAG_CD_1 through DIAG_CD_9), or (4) relevant diagnosis codes on the person's MAX HCBS OT service records (MAX OT diagnosis code variables DIAG_CD_1 or DIAG_CD_2).

  Relevant diagnoses and associated ICD-9 codes were: psychoses except in presence of affective disorders (schizophrenia, schizoaffective disorder, paranoid state, delusional disorder, shared psychotic disorder (295.00-295.05, 295.10-295.15, 295.20-295.25, 295.30-295.35, 295.40-295.45, 295.50-295.55, 295.60-295.65, 295.70-295.75, 295.80-295.85, 295.90-295.95, 297.0-297.3, 297.8, or 297.9), major affective disorders (bipolar I disorder, manic disorder, major depressive disorder) (296.00, 296.02-296.04, 296.10, 296.12-296.14, 296.20, 296.22-296.24, 296.30, 296.32-296.34, 296.40, 296.42-296.44, 296.50, 296.52-296.54, 296.60, 296.62-296.64, 296.7, 296.80-296.82, or 296.89), major anxiety disorders (panic disorder, obsessive-compulsive disorder) (300.01, 300.21, or 300.3), or other major disorders due to medical conditions (293.83, 294.0, 294.8, or 310.1).

• 65+. All HCBS participants who are age 65 or older.
• Under 65 with physical disabilities. All people who are under age 65 and are not classified as I/DD or SMI (includes, for example, traumatic brain injury and HIV waiver participants).

These subpopulations were defined to be non-mutually exclusive, so an individual could be classified as I/DD, SMI, and 65+ simultaneously. The under 65 with physical disabilities group is a residual category, needed in order to assign all HCBS participants to a subpopulation. Thus, there is no overlap between the under 65 group and the other three subpopulations.

Health maintenance organization (HMO). For purposes of Table 3, HCBS participants were classified as being enrolled in an HMO (managed acute care plan) if the MAX data indicated enrollment in a medical or comprehensive managed care plan at any point during the year. (If any of the PS file plan variables EL_PHP_TYPE_1_1 through EL_PHP_TYPE_4_12 equal 01, the person was designated as being enrolled in an HMO). Data on managed care enrollment among dually eligible people were checked against HMO enrollment as recorded in the Medicare Denominator File. Therefore, we interpret the HMO variable derived from MAX as including both Medicaid and Medicare managed care plans.

Hispanic, percentage. For purposes of Table 15, percentage Hispanic was defined as the percentage of an area's (county and State) population that was identified as Hispanic or Latino, regardless of race. Source data were from the Census Bureau's population estimates (Census County Characteristics File) for July 1, 2005 (ARF, 2008).

Home health agencies, supply of. Supply of home health agencies was measured at the county and State level as the total number of home health agencies certified by CMS as a Medicare or Medicaid home health agency, per population age 65 and older. Source data on home health agencies were from CMS (ARF, 2008); population data were from the Census Bureau (ARF, 2008).

I/DD. See HCBS subpopulations.

Intermediate care facility-mentally retarded (ICF-MR), supply of. ICF-MRs are certified by CMS to provide an institutional level of care to people with an intellectual or developmental disability. The supply of ICF-MR facilities was measured as the per capita number of ICF-MRs in an area (county and State measures). ICF-MR source data were the CMS Online Survey and Certification Reporting system as reported on the institutional 2005 Quarter 2 Provider of Service file (CMS, 2010a); population data were from the Census Bureau (ARF, 2008).

Long-term care funds spent on HCBS, percentage of a State's. For purposes of Table 13, the percentage of long-term care funds spent on HCBS is calculated as the following:

Medicaid expenditures for community-based long-term care services
_____________________________________________________________________________________

Medicaid expenditures for community-based long-term care + institutional long-term care expenditures

Medicaid community based long-term care expenditures include expenditures for 1915(c) waivers, Medicaid personal care services, and Medicaid home health care services. Medicaid institutional expenditures include expenditures for nursing facility services and ICF-MR services. Expenditures are for fiscal year 2005 and are from CMS Form 64 data as reported by Burwell, et al. (2006).

Managed long-term care plan, enrollment in a. For purposes of Table 3, HCBS participants were classified as being enrolled in a managed long-term care plan if the MAX data indicated enrollment in a managed long-term care plan or in a Program for All-Inclusive Care for the Elderly (PACE) plan at any point during the year. The MAX PS file captures enrollment in four prepaid plans for each month of the year; if any of the plan variables EL_PHP_TYPE_1_1 through EL_PHP_TYPE_4_12 = 05 or 06, the person was designated as being enrolled in a managed long-term care plan.

Median income, area's. Median household income information is derived from the Census Bureaus' Small Area Income and Poverty Estimates (SAIPE) program, which uses the American Community Survey as a primary data source. County-level information for 2005 was obtained from ARF (2008); State-level information was obtained directly from the Census Bureau (2010b) and measured income as of 2003.

Medically needy eligibility criteria, State's. The Medicaid option for medically needy people allows States to provide Medicaid to individuals who are ineligible because of excess income but have high medical expenses. Whether a State had a program for medically needy people in 2005 was abstracted from "Medicaid At-a-Glance 2005" (CMS, 2010c). States with a program for medically needy people were classified into one of two categories: more restrictive eligibility criteria and less restrictive eligibility criteria.

The categories were based on a State's 2005 income and assets eligibility criteria for the program for medically needy people. States were ranked based on income and asset criteria separately, and the rankings across the two were averaged. States that were ranked as less restrictive across income and assets combined were then classified as less restrictive and the remaining States as more restrictive.

States without programs for medically needy people were Alabama, Alaska, Colorado, Delaware, Idaho, Indiana, Mississippi, Missouri, Nevada, New Mexico, Ohio, Oklahoma, Oregon, South Carolina, South Dakota, and Wyoming. States with more restrictive programs for medically needy people were California, Kansas, Louisiana, Minnesota, New Hampshire, New York, Pennsylvania, and Rhode Island, as well as the District of Columbia. All other States were classified as less restrictive.

Mental disability, percentage of State population 65 and older with a. Mental disability was measured with the 2005 ACS (Census Bureau, 2010a). ACS respondents were asked if they had a physical, mental, or emotional condition lasting 6 months or more that made it difficult "learning, remembering, or concentrating."

Metropolitan statistical area (MSA). HCBS individuals are designated as living in an MSA if their county of residence was classified as an MSA by the Office of Management and Budget in June 2003; otherwise, people are classified as living in a nonmetropolitan area (non-MSA). This was determined using the ARF data for the Rural/Urban Continuum Codes (codes 01, 02, and 03 were used to code a county as an MSA).

Nursing home bed supply. Bed supply was measured at the county and State level as the total number of nursing home beds (skilled nursing facility beds plus nursing facility beds) per population age 65 and older. Source data on 2005 nursing home beds were from CMS (ARF, 2008); population data were from the Census Bureau (ARF, 2008).

Nursing home level of care eligibility criteria for HCBS. Federal regulations require that States limit eligibility for 1915(c) waiver services to Medicaid beneficiaries who meet the State's criteria for admission to a Medicaid institution (in addition to meeting the State's financial eligibility criteria). Criteria are set for a nursing home level of care and for ICF-MR level of care. States are responsible for establishing the criteria, which may be strict or flexible. A strict criterion limits the number of people who can receive HCBS.

A 2002 review of 45 States found that States use four approaches (or a combination of approaches) to set the nursing home level of care criteria (Mollica & Reinhard, 2005) and that the criteria can be arrayed on a continuum. Specifically, nursing home "admissions based solely on impairments in one or two out of five to six ADLs would be placed on the low end of the spectrum, those based on ADLs and medical criteria in the middle, and those based on medical criteria on the high end." (ADLs are activities of daily living and refer to tasks such as bathing, eating, and dressing.)

For purposes of Table 13, and using the continuum developed by Mollica and Reinhard (2005), California, Delaware, Kansas, New Hampshire, Ohio, Oregon, Rhode Island, Washington, and Wyoming were classified as having low criteria. Alaska, Arkansas, Colorado, Connecticut, Florida, Georgia, Idaho, Illinois, Indiana, Iowa, Louisiana, Massachusetts, Michigan, Minnesota, Mississippi, Missouri, Montana, Nebraska, New Jersey, New Mexico, North Dakota, Oklahoma, Pennsylvania, South Carolina, Texas, Vermont, and Wisconsin were classified as having middle criteria. Alabama, Arizona, Hawaii, Maine, Maryland, Tennessee, Utah, and Virginia were classified as having high (strict) criteria.

Physical disability, percentage of State population 65 and older with a. Physical disability was measured with the 2005 ACS (Census Bureau, 2010a). ACS respondents were asked if they had "a condition that substantially limits one or more basic physical activities such as walking, climbing stairs, reaching, lifting, or carrying."

Poverty, percentage of area's population in. Poverty information is derived from the Census Bureaus' SAIPE program, which uses the ACS as a primary data source. County and State-level information (for 2005) was obtained from the ARF (2008).

Prevalence of asthma, State's. Data are from the BRFSS (CDC, 2010) for 2005. Prevalence estimates are based on adult BRFSS respondents who responded yes to the question: "Have you ever been told by a doctor, nurse, or other health professional that you had asthma?"

Prevalence of cardiovascular disease, State's. Data are from the BRFSS for 2005 and are based on the questions: "Has a doctor, nurse, or other health professional EVER told you that you had any of the following? (Ever told) you had angina or coronary heart disease?"

Prevalence of diabetes, State's. Data are from the BRFSS for 2005 and are based on the question: "Have you ever been told by a doctor that you have diabetes?"

Prevalence of high blood pressure, State's. Data are from the BRFSS for 2005 and are based on the question: "Have you ever been told by a doctor, nurse, or other health professional that you have high blood pressure?"

Prevalence of serious mental illness, State's. Numerator data used to calculate the prevalence of serious mental illness were based on the work of Hozler, et al. (1981), as reported by Grabowski and colleagues (2009). A statistical model was used to estimate the number of people with serious mental illness in 2005, by State, based on State risk factors. The definition is based on a minimum impairment score; a minimum number of disability days; and a range of chronic conditions including bipolar I and II, mania, major depression with hierarchy, dysthymia hierarchy, generalized anxiety, hypomania, major depressive episode, panic disorder, posttraumatic stress disorder, agoraphobia with or without panic, social phobia, and specific phobia. Denominator data used to calculate the State prevalence rates for serious mental illness were Census Bureau population data (ARF, 2008).

Psychiatric beds, supply of. As defined by the American Hospital Association (AHA), the supply of psychiatric beds is the number of psychiatric care beds set up in short-term general hospitals. Bed supply was measured as the number of psychiatric beds per area population (at State and county level). Source data were AHA information for 2005 (as reported on the 2007 ARF); population data were from the Census Bureau (ARF, 2007).

Race/ethnicity. Classification of HCBS participants by race/ethnicity is based on the following five racial/ethnic groups: white only non-Hispanic, black only non-Hispanic, Hispanic, other non-Hispanic races or multiple races, and unknown (non-Hispanic) race. Data were from CMS administrative data provided on the MAX PS file (RACE_CODE_1 through RACE_CODE_5 and ETHNICITY_CODE).

Self-care disability, percentage of State's population 65 and older with a. This percentage was measured with the 2005 ACS (Census Bureau, 2010a). ACS respondents were asked if they had a physical, mental, or emotional condition lasting 6 months or more that made it difficult "dressing, bathing, or getting around inside the home."

Sensory disability, percentage of State's population 65 and older with a. This percentage was measured with the 2005 ACS (Census Bureau, 2010a). ACS respondents were asked if they had "blindness, deafness, severe vision or hearing impairment."

Short-term acute care hospital beds, supply of. Defined by AHA as short-term general hospitals, these hospitals are general medical and surgical hospitals; most patients stay fewer than 30 days. Supply of hospital beds was measured as the number of hospital beds per capita (at the State and county level). Source data for 2005 bed supply were from AHA as reported on the ARF (2008); population data were from the Census Bureau (ARF, 2008).

SMI. See HCBS subpopulations.

Social Security Disability Insurance (SSDI), percentage of State's population 18-64 with. Numerator data for the percentage of people identified by the Social Security Administration as disabled and ages 18-64 are as reported in Table 8 of the Annual Statistical Report on the Social Security Disability Program 2005 (Social Security Administration, 2010). Population data for the denominator were for the 2005 resident population ages 18-64 and are from the Census Bureau (ARF, 2008).

SSDI persons in State with diagnosis of mental retardation as a percentage of State's population. Numerator data for the percentage of people identified by the Social Security Administration as disabled and with a diagnosis of mental retardation are from Table 10 of the Annual Statistical Report on the Social Security Disability Program 2005 (Social Security Administration, 2010); population data for the denominator were for 2005 and are from the Census Bureau (ARF, 2008).

State plan services offered. For purposes of Table 9, optional State plan services offered were abstracted from "Medicaid-At-a-Glance 2005" (CMS, 2010c). The data were merged to HCBS participants based on the person's State of residence as reported on the MAX PS file. Services offered are for home health therapies (physical therapy, occupational therapy, speech therapy, or hearing services), private duty nursing, personal care, targeted case management, hospice care, transportation, adult day care, and residential care.

State plan services, users of. For purposes of Tables 5 and 11, users of optional State plan services are defined as people in the HCBS population who have a positive community long-term care Medicaid fee-for-service payment for nonwaiver services at any point during 2005. Services include private duty nursing, personal care, targeted case management, hospice care, transportation, adult day care, and residential care (MAX PS file CLTC_FFS_PAYMT_AMT_11, 12, 13, 15, 17, 18, or 20 are >0).

Users of home health therapies are defined as people with fee-for-service payment for physical therapy, occupational therapy, speech therapy, or hearing services greater than zero (MAX PS file FFS_PYMT_AMT_34 >0). For purposes of Table 5, users of mandatory State plan home health services are defined as people with a positive value for community long-term care nonwaiver home health fee-for-service payment (CLTC_FFS_PAYMY_AMT_14 >0).

Underserved by mental health providers. People living in a county designated by the Health Resources and Services Administration (as of November 2004) as a Health Professional Shortage Area for mental health professionals were categorized as being underserved based on the number of professionals per capita. Mental health providers include psychiatrists, clinical psychologists, clinical social workers, psychiatric nurse specialists, and marriage and family therapists. Data are from the 2008 ARF.

Underserved by primary care providers. People living in a county designated by the Health Resources and Services Administration (as of November 2004) as a Health Professional Shortage Area for primary care practitioners were categorized as being underserved based on the number of primary care providers per capita. Primary care practitioners are defined as non-Federal doctors of medicine or osteopathy who provide direct patient care and who practice principally in one of four specialties: general or family medicine, general internal medicine, pediatrics, and obstetrics and gynecology. Data are from the ARF (2008).

Users of HCBS State plan and 1915(c) waiver services. For the purposes of Table 3, HCBS participants were categorized into one of three categories based on the fee-for service community-based long-term care (CLTC) payment amount variables provided on the MAX PS file.

• Used only HCBS State plan services. HCBS State plan services were defined as nonwaiver services at any point during 2005 for personal care, private duty nursing, adult day care, home health care, residential care, rehabilitation for aged or disabled enrollees, targeted case management for aged or disabled enrollees, transportation for aged or disabled enrollees, or hospice care for aged or disabled enrollees. If any of these services had a Medicaid fee-for-service payment amount greater than zero and the individual used no 1915(c) waiver services, then the individual was categorized as a user of only HCBS State plan services (on the MAX PS file, CLTC_FFS_PAYMT_AMT_11, 12, 13, 14, 15, 16, 17, 18, or 19 are >0) and CLTC_FFS_PAYMT_AMT_30, 31, 32, 33, 34, 35, 36, 37, 38, 39, and 40 have no positive values).
• Used only 1915(c) waiver services. Waiver services were defined as 1915(c) waiver services at any point during 2005: personal care, private duty nursing, adult day care, home health care, residential care, rehabilitation, targeted case management, transportation, hospice care, durable medical equipment, or other waiver services not listed. If any of these services had a Medicaid fee-for-service payment amount greater than zero and the individual used no HCBS State plan services, then the individual was categorized as a user of only 1915c) waiver services (CLTC_FFS_PAYMT_AMT_30, 31, 32, 33, 34, 35, 36, 37, 38, 39, or 40 are >0 and CLTC_FFS_PAYMT_AMT_11, 12, 13, 14, 15, 16, 17, 18, and 19 have no positive values).
• Used both HCBS State plan and 1915(c) waiver services. Individuals who used both HCBS State plan services and 1915c) waiver services at any point during 2005 were categorized into this category (CLTC_FFS_PAYMT_AMT_11, 12, 13, 14, 15, 16, 17, 18, or 19 are >0) and CLTC_FFS_PAYMT_AMT_30, 31, 32, 33, 34, 35, 36, 37, 38, 39, or 40 are >0).

Waiver plan enrollment groups, 1915(c). States classify 1915(c) waivers into categories that indicate the target population served by the HCBS waiver. The target population enrollment groups are aged and disabled people, aged, physically disabled people, people with brain injuries, people with HIV/AIDS, mentally retarded/developmentally disabled people, people with mental illness/serious emotional disturbance, and technology dependent/medically fragile people. For the purposes of Table 4, HCBS participants enrolled in a 1915(c) waiver at any time during 2005 were classified by their respective waiver target population group (an unknown residual category is also reported). For the few persons enrolled in more than one 1915(c) waiver, persons are counted for each of their respective waivers. Waiver target population enrollment groups are as defined by the States in their MSIS as reported on the MAX PS file (any MAX_WAIVER_TYPE_1_MO_1 through MAX_WAIVER_ TYPE_3_MO_12 equal to F, G, H, I, J, K, L, M, N, or O, for their respective waiver population enrollment group).

Waiver services offered by the State. For purposes of Table 10, waiver services were defined as 1915(c) waiver services used at any point during 2005 for: personal care, case management, adult day/health care, residential care, durable medical equipment/supplies, or transportation. To determine if an individual was enrolled in a 1915(c) waiver plan that provided the subject service, we abstracted information on waiver plans taken from approved State waiver applications available from the CMS Web site (2010e).

For waivers without an electronic application on the CMS Web site, we compiled data from the University of California-San Francisco Personal Assistance Services Web site (University of California-San Francisco, 2010) and State Web sites. The waiver application plan level information on availability of the individual services was then linked to HCBS participants (by MAX_WAIVER_TYPE_1, 2, or 3 plus MAX_WAIVER_ID_1, 2, or 3). For the few people enrolled in more than one 1915(c) waiver, persons are linked to services for both of their respective waivers. Enrollment in waiver plans that provided residential care or durable medical equipment/supplies were obtained from the waiver application field "Other Services" and extracted for analytical purposes.

Waiver services, users of 1915(c). For purposes of Tables 6 and 12, waiver services were defined as 1915(c) waiver services for personal care, case management, adult day/health care, residential care, durable medical equipment, or transportation. If any of these waiver services had a Medicaid community long-term care fee-for-service payment amount greater than zero (at any point during 2005), the individual was categorized as a user of that service (MAX PS file CLTC_FFS_PAYMT_AMT_ 31, 33, 35, 37, 38, or 40 >0).

White, percentage. For purposes of Table 15, the per-capita white population was defined as the percentage of an area's (county and State) population that was identified as non-Hispanic white. Source data were from the Census Bureau's population estimates (Census County Characteristics File) for July 1, 2005, as reported on the 2008 ARF.

Appendix References

Administration on Aging. AGing Integrated Database (AGID). Available at: . Accessed June 2010.

Area Resource File (ARF). Rockville, MD: Health Resources and Services Administration, Bureau of Health Professions; 2008.

Burwell B, Sredl K, Eiken S. Medicaid long-term care expenditures in FY 2005. Cambridge, MA: Chronic Care and Disability Group, Thomson Reuters; July 7, 2006.

Crowley J. Medicaid medically needy programs: an important source of Medicaid coverage. Washington, DC: Kaiser Commission on Medicaid and the Uninsured; January 2003.

Census Bureau. American Community Survey 2005 public use microdata sample files. Accessed June 2010a.

Census Bureau. Small Area Income and Poverty Estimates program. Available at: . Accessed July 2010b.

Centers for Disease Control and Prevention. Behavioral Risk Factor Surveillance System. Available at: . Accessed June 2010.

Centers for Medicare & Medicaid Services (CMS). Health plans, reports, files and data: institutional-2005, POS Institutional Report. Available at: . Accessed June 2010a.

CMS. Medicaid Analytic eXtract (MAX) general information. Available at: . Accessed November 2010b.

CMS. Medicaid At-a-Glance 2005. Available at: . Accessed June 2010c.

CMS. Research Data Assistance Center (ResDAC) Medicare Data Documentation. Available at: . Accessed November 2010d.

CMS. Medicaid state waiver program demonstration projects: Medicaid waivers and demonstrations list. Available at: . Accessed April 2010e.

Grabowski DC, Aschbrenner KA, Feng Z, et al. Mental illness in nursing homes: variations across states. Health Aff 2009;28(3):689-700.

Holzer CE, Jackson DJ, Tweed D. Horizontal synthetic estimation: a strategy for estimating small area health related characteristics. Eval Program Plann 1981;4(1):29-34.

Mollica R, Reinhard S. Establishing "nursing home level of care" criteria: how states vary. (Prepared under CMS Grant No.11-P-92015/2-01 by National Academy for State Health Policy (NASHP) and the Rutgers Center for State Health Policy (CSHP) for the Community Living Exchange at Rutgers/NASHP). Issue Brief. New Brunswick, NJ: Rutgers CSHP; October 2005. Available at: (Word® file). Accessed June 2010.

Social Security Administration. Annual Statistical Report on the Social Security Disability Program, 2005. Available at: . Accessed June 2010.

University of California-San Francisco. Personal Assistance Services Web site: Medicaid waiver data. Available at: . Accessed June 2010.