Congratulations on your New Publication!
Congratulations Dr. Melissa Armstrong on the publication of ” Social determinants of healthy aging: An investigation use the All of Us Cohort,” which appears in the March issue of PLOS One.
This retrospective cohort study used data from 99,935 participants aged 50 and older from the All of Us Research Program to examine how social determinants of health (SDOH) relate to healthy aging. Researchers trained three machine learning models—logistic regression, multi-layer perceptron, and extreme gradient boosting (XGBoost)—to predict healthy aging, defined by a composite measure including comorbidities, cognitive conditions, and mobility function. Among the models, XGBoost with random oversampling performed best (AUROC 0.793), demonstrating balanced predictive accuracy across race and sex groups. Feature importance analysis showed that health insurance type was the strongest predictor of healthy aging, followed by employment status, substance use, and insurance coverage. Overall, the findings highlight the significant influence of health insurance and other SDOH factors in predicting and potentially supporting healthy aging outcomes.