Abstract
OBJECTIVES: Decades of research indicate that volunteering is associated with better health for the volunteer beyond the selection effects based on health. However, little is known about potential heterogeneity in health outcomes associated with volunteering in the context of good or poor health. This study addresses this gap by focusing on the frailty index (FI) to investigate the volunteering-health nexus across the population frailty distribution ranging from fit to frail. METHODS: Using nationally representative data from the Health and Retirement Study (person N = 34,986; 198,218 person-wave observations), we estimated unconditional quantile regression models with panel fixed effects to estimate changes in FI associated with changes in the share of volunteers in the population across the frailty distribution observed across the study period (1998-2020). RESULTS: Our findings demonstrated that the volunteering-FI association was heterogeneous across the frailty distribution. The association was the most potent at the higher end of the frailty distribution, suggesting that efforts to promote volunteering may yield greater benefits for older adults experiencing high levels of frailty. DISCUSSION: The current study findings provide unique and compelling evidence in support of earlier calls for considering volunteering as a public health intervention. The study findings are discussed in the context of population health outcomes and health disparities.