How rural is All of Us? Comparing characteristics of rural participants in the National Institute of Health's All of Us Research Program to other national data sources

AbstractPurposeThe National Institute of Health's All of Us Research Program represents a national effort to develop a database to advance health research, especially among individuals historically underrepresented in research, including rural populations. The purpose of this study was to describe the rural populations identified in the All of Us Research Program using the only proxy measure currently available in the dataset.MethodsCurrently, the All of Us Research Program provides a proxy measure of rurality that identifies participants who self‐reported delaying care due to far travel distances associated with living in rural areas. Using the All of Us Controlled Tier Dataset v6, we compared sociodemographic and health characteristics of All of Us rural participants identified via this proxy to rural US residents from nationally representative data sources using chi‐squared tests.Results3.1% of 160,880 All of Us participants were rural, compared to 15%‐20% of US residents based on commonly accepted rural definitions. Proportionally more rural All of Us participants reported fair or poor health status, history of cancer, and history of heart disease (P<.01).ConclusionsThe All of Us measure may capture a subset of underserved participants who live in rural areas and experience health care access barriers due to distance. Researchers who use this proxy measure to characterize rurality should interpret their findings with caution due to differences in population and health characteristics using this proxy measure rural compared to other commonly used rural definitions.