Cohort Survey-CMS Data Linkage for Multi-level Modeling and Novel Risk Prediction of Suicide in Diverse Older Adults

Midlife and older adults have the highest rates of suicide in the U.S., and these rates vary widely by race and ethnicity. Complex social, behavioral and health factors can either increase suicide risk or enhance resilience. These factors also vary by race and ethnicity, but existing suicide prediction models do not perform equally across all groups. We can improve suicide prevention efforts by developing more sophisticated risk frameworks and creating new machine-learning-based prediction models applicable to diverse older adults.