Revisiting the Area Deprivation Index

Abstract Objective To re-estimate and re-validate the Area Deprivation Index to address recent criticism of the existing index, which is calculated and distributed by Neighborhood Atlas. Data Sources To calculate the updated Area Deprivation Index (ADI), we obtained 17 census measures from the 2018–2022 American Community Survey (ACS) 5-year data that reflected poverty, housing, employment, and education within census block groups, census tracts, and counties. To validate the association of the updated index with health outcomes, we obtained the most recent census tract-level life table data from the U.S. Small-area Life Expectancy Estimates Project (USALEEP). Study Design We used principal components analysis to estimate new factor weights associated with the primary component summarizing the 17 selected census measures. The resulting scores were ranked into percentiles to arrive at the updated ADI. We compared this updated ADI to the existing ADI distributed by the Neighborhood Atlas. We calculated average life expectancy and age-specific mortality rates by groups defined by the updated ADI. Principal Findings The principal components analysis identified a single factor that explained 46% of variance at the census tract level. There were noticeable differences in factor loadings for the unemployment rate and the proportion of households without a motor vehicle compared to the original loadings. Results were similar at the county and block group levels. Compared to existing ADI values, there were substantive changes in the updated ADI values for many geographic ares with high home values, but low employment and educational attainment. The updated ADI demonstrated robust associations with age-specific mortality and life expectancy. Conclusions The updated ADI better summarized the 17 underlying census variables than the current ADI. The updated index was strongly correlated with life expectancy and mortality. Callout Box What is known on this topic ADI is a measure of area-level deprivation that summarizes 17 Census measures regarding poverty, housing, employment, and education. The original ADI demonstrated associations with mortality and clinical outcomes. The current version of the ADI made available by the Neighborhood Atlas relies on old factor weights and appears to have flaws in its construction that lead to overreliance on home value and income variables. What this study adds To address these concerns, this study re-estimates the ADI properly from source data to ensure that the final index reflects a combination of all 17 census measures, and not just home value and income. The updated ADI more accurately reflects the distribution of deprivation in neighborhoods across the country and is highly correlated with life expectancy and age-specific mortality rates. The updated ADI is available publicly and should be used by researchers who would like to utilize a broad measure of neighborhood deprivation.