Social network typologies and mental health among older adults.

In this study, we test the robustness of previous social network research and extend this work to determine if support quality is one mechanism by which network types predict mental health. Participants included 1,669 adults aged 60 or older from the Americans' Changing Lives study. Using cluster analysis, we found diverse, family, and friends network types, which is consistent with the work by Litwin from 2001. However, we found two types of restricted networks, rather than just one: a nonfamily network and a nonfriends network. Depressive symptomatology was highest for individuals in the nonfriends network and lowest for individuals in the diverse network. Positive support quality partially mediated the association between network type and depressive symptomatology. Results suggest that the absence of family in the context of friends is less detrimental than the absence of friends in the context of family, and that support quality is one mechanism through which network types affect mental health.