Abstract
Reconstructing Longitudinal Address Histories to Assess Health Impacts of Place-based Stressors in Older Adults
Laura McGuinn, PhD, Assistant Professor at the University of Chicago
A growing body of literature highlights the significant role of an individual’s residential environment in shaping health and economic outcomes throughout their life course. Variation in the social and economic conditions across neighborhoods can produce place-based social and environmental stressors that affect health, including higher levels of air pollution, noise, physical disorder, crime, and lower levels of green space and walkability. However, research on the impact of neighborhood environments on the health of older adults has been limited to measuring current environments using the patients’ current addresses from health care systems datasets. To date, there is limited research assessing the feasibility and reliability of using residential histories that are sometimes available in electronic health record (EHR) data, or linking data on residential histories from external, commercial sources, especially for individuals who may move frequently due to housing and/or financial insecurity and older populations. To address this gap, we propose a pilot study to evaluate the feasibility of constructing residential histories for aging populations. We will compare address data from EHR data with those from a commercial database, InfoUSA, and assess the reliability of these sources for exposure assessment. InfoUSA, published by the company Data Axle, contains information on households based on an amalgamation of data sources including address lists, marketing datasets, and publicly available records. This dataset provides a new source of address histories for individuals that appear in the EHR data from the University of Chicago Medical Center (UCMC). The pilot’s specific aims are to:
- Construct detailed, 10-year residential histories for 10,976 UCHP patients using both UCMC’s EHR data and InfoUSA data.\
- Assess the concordance of the two constructed residential history measures, evaluating differences by demographic and economic factors including sex, race/ethnicity, insurance status, age, marital status, renter/owner status, and urban/rural status, consistent with the NIA Health Disparities Research Framework.
- Link both residential history sources to environmental and social exposures to assess differences in exposures estimates across data sources.