Volunteering as a predictor of all-cause mortality: what aspects of volunteering really matter?

Background: This study evaluates the predictive effects of different aspects of volunteering (e.g. volunteering status, number of hours, number of years, and type of volunteering activity) on all-cause mortality. Methods: A seven-year follow-up dataset of a nationally representative sample of Israelis, 60 years and older was used. Results: As expected, volunteering was associated with a reduced mortality risk even after adjusting for age, gender, education, baseline mental health and physical health, activity level, and social engagement. Those who volunteered for 10 to 14 years had a reduced mortality risk relative to non-volunteers. In addition, those who volunteered privately, not as part of an official organization, also had a reduced mortality risk compared to non-volunteers. The number of hours of volunteering was not a significant predictor of all-cause mortality in the fully adjusted model. In additional sensitivity analyses limited to those who volunteered, none of the various aspects of volunteering was associated with a reduced mortality risk. Conclusions: Results suggest that not all aspects of volunteering have the same predictive value and that the protective effects of length of volunteering time and type of volunteering are particularly important. However, whether or not volunteering is the most consistent predictor of mortality and whether once a person volunteers the various aspects of volunteering are no longer associated with mortality risk.