Predictors of 30-year mortality in depressed and comparison samples.

Although higher rates of depression and lower rates of social behaviors (i.e., social support and activities) and physical activities are associated with mortality, the independent contribution of each of these factors needs examination.A prospective sample of 848 individuals (55% female) was used; half were clinically depressed at baseline; the other half comprised a comparison sample matched on census tract, gender, and marital status. Depressive symptoms, social behaviors, and physical activities were examined as time-varying predictors of mortality over a 30-year period using multiple imputation for missing data and Cox proportional hazards regression, controlling for demographic factors and health risk factors.By the end of the study, 137 individuals from the depressed sample and 99 individuals from the comparison sample had died. Although the mortality rate is higher in the depressed sample, after controlling for demographic, health risk factors, social behaviors, and physical activity, there was no significant difference in mortality between the depressed and control samples. Among participants in the depressed sample, reduced participation in social activities was significantly associated with a higher risk of mortality.Frequency and intensity of activities were not assessed and all data except for mortality were self-report.Promoting social engagement through activities may hold promise for delaying mortality among individuals who are depressed. Potential methods to promote social engagement and factors such as positive emotions that should be considered in future studies are discussed.