Association for Public Policy Analysis & Management

Prior research has demonstrated that the distribution of disability varies geographically, with disproportionately high concentrations of high disability prevalence rates in the Appalachian and Southeastern regions of the United States known as the “Disability Belt.” The purpose of this investigation was: 1) to validate spatially contiguous geographic areas with high disability prevalence using measures of both self-reported disability and disability claims; 2) to compare spatial patterns and clustering across these two distinct sources of disability data; and 3) to assess whether variation in county economic composition accounts for observed geographic variation in disability. To accomplish this, disability data from the 2010 pooled five-year American Community Survey (ACS) and the Social Security Administration’s (SSA) Old-Age, Survivors, and Disability Insurance (OASDI) program participant data were combined with spatial data from US Census TIGERLINE files for 3,019 counties in the 48 contiguous United States. Moran’s I and Local Indicator of Spatial Autocorrelation (LISA) statistics were used to assess global and local clustering, respectively. A lagged spatial regression model was used to assess associations between disability prevalence and county-level socioeconomic indicators. All analysis was conducted in GeoDa (v.1.8.14). The prevalence of disability in the US according to both self-reported functioning difficulties and disability claims was significantly clustered (Moran’s I=0.60, p<0.001 and I=0.70, p<0.001, respectively). Significant local clustering at p < 0.01 was observed in 20.6% of the total counties, with contiguous areas of high disability prevalence identified largely in the Appalachian, Southeastern, and lower Midwest regions. Contiguous areas of significantly low disability prevalence were identified in the upper Midwest and Rocky Mountain areas as well as the New York metro and California coastal areas. County-level low median household income as well as high proportions of unemployed residents, public assistance recipients, and persons with low educational attainment were significantly associated with increased disability prevalence across both data sources. The findings from this analysis more clearly establish the existence of a definable Disability Belt in the United States that is congruent with prior reports as well as reports of geographic clustering of other chronic health conditions. Geographic variation is partially accounted for by differences in area socioeconomic composition. More research is needed to identify other area level characteristics, such as land usage and healthcare access, which may also contribute to geographic disparities in disability prevalence. Such information has implications for how a variety of services and supports are provided at the state and local level, including employment services, health care, housing assistance, income assistance, rehabilitation services, and transportation.