Longitudinal associations between mental health and social environment in older adults: a multilevel growth modeling

Abstract

Objectives This study aimed to assess longitudinal relationships between social environment indicators (social connectedness, social engagement, social contribution) and mental health indicators (depression and anxiety) among community-dwelling adults age 55 years and older. Methods Data were drawn from 3-waves of the national longitudinal survey of Midlife Development in the United States (MIDUS) (N = 2,020; age range = 55-94 years). We developed multilevel growth models to ascertain the relationships of interest, controlling for sociodemographic and physical health factors. Results Over the 20-year period of study, lower levels of emotional social support, social integration and social contribution significantly predicted depression and anxiety, whereas social network and social engagement were not significant predictors of these mental health outcomes in older adults. The models also indicated a moderation effect of the number of chronic conditions on the slopes of depression and anxiety. Discussion Considering our findings, interventions to enhance social contribution and social connectedness could be effective to help older adults maintain positive mental health, as well as programs that facilitate older adults’ connections with their families, communities and health care providers. These interventions must also account for multiple chronic conditions since functional limitations drive declining integration in the community and participation in social activities.