Predictor Selection Data from the AHRQ SDOH 2022 Dataset Used to Estimate County-Level Social Isolation

<p dir="ltr">This repository contains supporting predictor-selection data for the study “County-level social isolation in the United States: Estimating missing prevalence and identifying structural correlates of geographic variation.” The analysis used county-level variables from the 2022 Agency for Healthcare Research and Quality (AHRQ) Social Determinants of Health (SDOH) Database.</p><p dir="ltr">The first file contains the 331 candidate SDOH predictors considered in the regression modeling process after exclusion of identifier columns and variables without complete county-level data.

The second file documents predictors excluded during the subsequent predictor-selection process and the corresponding reasons for exclusion.</p><p dir="ltr">These files are provided to improve transparency and reproducibility of the predictor-selection workflow used to develop the primary county-level social isolation regression model.</p>