Area Deprivation Index (ADI) Neighborhood Atlas

General description

1. Database primary purpose - To make measures of neighborhood disadvantage available to the public, including educational institutions, health systems, non-profit organizations, and government agencies, in order to use these metrics in research, program planning, and policy development.
2. Overall data type - Demographics, socioeconomics
3. Dataset type - Cross-sectional
4. Data source - Census
5. Data level - patient level
6. Geographic location of the data collection sites - United States
7. Sponsor, manager, or home institution - Center for Health Disparities Research, University of Wisconsin
8. Date range - 2015 and 2020
9. Geolocation data - The 9-digit zip code ID and the block group Census ID
10. Clinical areas of interest - all
11. Number of records - 2015: 220,333 records for block group files and 76,802,374 records for ZIP+4 files. 2020: 242,335 records for block group files and 68,864,917 records for ZIP+4 files
12. Variables that are uniquely present in this dataset - National percentile of block group ADI score and state-specific decile of block group ADI score
13. Database caveats and limitations - (1) The ADI is limited insofar as it uses the American Community Survey (ACS) Five-Year Estimates in its construction; then, all limitations of the source data will persist throughout the ADI. (2) The choice of geographic units will also influence the ADI value. In the case of the ADI, the Census Block Group is the geographic unit of construction, as the Census Block Group is considered the closest approximation to a "neighborhood." As such, we can only recommend linking the ADI to census block groups, as other geographic units (including 5-digit ZIP codes, ZCTA, and others) will not be valid. (3) Construction of the version 3 2018 ADI includes suppression of block groups containing any of the following: less than 100 people, less than 30 housing units or more than 33% of the population living in group quarters, and Census data labeled as N/A or missing in the core component variables.

Applicable methods

1. Association methods, such as Cox proportional-hazards regression (1), logistic regression (2, 3), linear mixed model (4)
2. Univariate analyses (2)
3. Bivariate analyses (2)
4. Machine learning (5)

High-impact designs

1. Evaluate whether estimates of active and disabled life expectancy differ on the basis of neighborhood disadvantage after accounting for individual-level socioeconomic characteristics and other prognostic factors (6)

Data dictionary

To access the ADI Neighborhood Atlas data dictionary, click here

Variable categories

1. ADI scores: national percentile of block group ADI score and state-specific decile of block group ADI score