Dementia care in diverse older adults in the Deep South and the rest of the United States

Abstract

AbstractBackgroundPrevalence of Alzheimer’s disease and related dementia (ADRD) is increasing especially in the southeastern US. For older adults with ADRD, gaps exist in utilization of specialists (neurologists, psychiatrists, geriatricians) and ADRD drugs, e.g., across racial/ethnic (R/E) patient groups. Because some R/E groups, e.g., African Americans (AAs), are overrepresented in the Southeast, closer scrutiny of their ADRD care is needed. Within the Southeast, the Deep South (DS) includes some of the poorest US states (Alabama, Mississippi, Louisiana, Georgia, and South Carolina). We examined R/E disparities in specialist and drug utilization in DS vs. non‐DS states, and examined individual and context‐level factors that impact utilization in each region.MethodsPopulation, outcomes, and individual‐level covariates were identified from claims for fee‐for service Part A/B (hospital, outpatient) and D (prescription drugs) Medicare beneficiaries in 2013‐2015 (66 and older). For context‐level covariates, we combined Area Health Resource File county‐level data in exploratory factor analysis (EFA). We conducted adjusted analyses for DS and non‐DS with outcomes: 1) ≥ 1 specialist visit; and 2) ≥ 1 prescription for donepezil, galantamine, rivastigmine, or memantine. R/E groups included AAs, Non‐Hispanic Whites, Hispanic, Asian/Pacific Islander, and Other.ResultsWe identified 127,512 Medicare beneficiaries with ADRD (9.4% in DS). Specialist utilization ranged from 37% AAs to 42% Whites in DS, 45% Whites to 50% AAs in non‐DS. ADRD drug utilization ranged from 48% AAs to 55% Hispanics in DS, 42% AAs to 48% Hispanics in non‐DS. EFA identified 4 context‐level factors with two common to DS and non‐DS. R/E disparities were not significant, except for non‐DS Asian/Pacific Islanders vs. Whites (RR 0.81, CI = 0.777‐0.849, p<.0001). In DS and non‐DS, comorbidities, poverty, and EFA factor “Medical Resources Availability” were associated with utilization outcomes. No other factor was associated with outcomes in DS. In non‐DS, factors “Population Health”, “Health Insurance Market” were associated with outcomes.ConclusionADRD specialist and drug utilization is relatively low in both DS and non‐DS, with no significant R/E differences. Research should continue to examine factors that may drive medical care decisions of older adults with ADRD in US regions with different socio‐economic and medical resource contexts.