Admissions to High‐Quality Nursing Homes from Community: Racial Differences and Medicaid Policy Effects

Research Objective To examine racial differences in admissions to high‐quality nursing homes (NHs) among residents with Alzheimer's disease and related dementias (ADRD), and whether such racial differences can be influenced by dementia‐related state Medicaid add‐on policies. Study Design Multiple CY2011‐2017 national data were linked: The Minimum Data Set (MDS) 3.0, Medicare Beneficiary Summary File (MBSF), Medicare Provider Analysis and Review (MedPAR), and nursing home compare (NHC). We also collected information on state dementia‐related Medicaid add‐on policies and the 2015 national Area Deprivation Index (ADI). For each identified resident, we constructed a choice set of NHs, defined as a set of NHs within a certain driving distance radius from home (15 miles in urban areas and 20 miles in rural areas), based on the 5‐digit zip‐code of a NH and individuals' community residence. We excluded individuals who did not choose a NH within their choice set and individuals with only one NH within their choice set. The dependent variable was whether a NH is “chosen” or not (binary). Key independent variables included the interaction of NH quality (measured by 5‐star rating) and race (i.e. black versus white), and interaction of NH quality, race, and state Medicaid dementia‐related add‐on policies. Individual‐ and facility‐level covariates (e.g. distance from residence to NHs) were identified. A set of McFadden's choice (conditional logit) models were estimated. Population Studied We identified Medicare enrolled NH residents with ADRD who were newly admitted from community (i.e. no NH episodes or hospitalizations in the prior 30 days) between 01/01/2011 and 12/31/2017. In total, 874,526 individuals in 15,755 NHs were identified. Principal Findings In the study sample, 89% were whites and 11% were blacks. Overall, 50% of whites and 35% of blacks were admitted to high‐quality NHs (4‐ or 5‐star). Blacks were less likely to have aggressive behaviors (14.7% vs. 16.6%, P < 0.01) but more likely to have severe cognitive impairment (9.6% vs. 6.8%, P < 0.01) than whites. Blacks were more likely to live in social‐economically disadvantaged neighborhoods and be Medicare‐Medicaid dually eligible. Results from the McFadden's model suggested that blacks were less likely to be admitted to a high‐quality NH than whites (OR = 0.7, P < 0.01). Although residents with low socio‐economic status, cognitive impairment and behavioral problems were less likely to enter high‐quality NHs, these characteristics did not fully explain the observed racial differences in entering high‐quality NHs. However, the racial difference was reduced in states with dementia‐related add‐on policies, compared to states without these policies (OR = 1.2, P < 0.01). Conclusions Blacks with ADRD were less likely to be admitted to high‐quality NHs than whites. Individuals' health conditions and social‐economic status could not fully explain this racial difference. State Medicaid policies that provided additional subsidies to ADRD residents may reduce this racial difference. Implications for Policy or Practice It is important to deliver high quality NH care for individuals with ADRD. Racial minorities still face barriers in accessing high‐quality NHs. Medicaid add‐on policies appear to reduce the racial difference in entering high‐quality NHs. Primary Funding Source National Institutes of Health.