The BARI-hoods Project: neighborhood social determinants of health and postoperative weight loss using integrated electronic health record, census, and county data.

Abstract

While social determinants of health (SDoH) have gained attention for their role in weight loss following bariatric surgery, electronic health record (EHR) data provide limited information beyond demographics associated with disparities in weight loss. To integrate EHR, census, and county data to explore disparities in SDoH and weight loss among patients in the largest populous county of Ohio. Seven hundred seventy-two patients (82.1% female; 37.0% Black) who had primary bariatric surgery (48.7% gastric bypass) from 2015 to 2019 at Ohio State University. EHR variables included race, insurance, procedure, and percent total weight lost (%TWL) at 2/3, 6, 12, and 24 months. Census variables included poverty and unemployment rates. County variables included food stores, fitness/recreational facilities, and open area within a 5- and 10-minute walk from home. Two mixed multilevel models were conducted with %TWL over 24 months, with visits as the between-subjects factor; race, census, county, insurance, and procedure variables were covariates. Two additional sets of models determined within-group differences for Black and White patients. Access to more food stores within a 10-minute walk was associated with greater %TWL over 24 months (P = .029). Black patients with access to more food stores within a 10-minute (P = .017) and White patients with more access within a 5-minute walk (P = .015) had greater %TWL over 24 months. Black patients who lived in areas with higher poverty rates (P = .036) experienced greater %TWL over 24 months. No significant differences were found for unemployment rate or proximity to fitness/recreational facilities and open areas. Close proximity to food stores is associated with better weight loss 2 years after bariatric surgery. Lower poverty levels did not negatively affect weight loss in Black patients.