Social network composition and inflammation at midlife: A socioemotional selectivity theory perspective.

The Socioemotional Selectivity Theory of aging posits that the narrowing of social networks with age may provide emotional benefits. It is unknown whether these benefits extend to measures of physical health, such as markers of inflammation. In a longitudinal midlife cohort, we (1) characterized age-related changes in social networks and inflammation, (2) tested associations between changes in social network composition and changes in inflammation, and (3) explored whether changes in positive affect or social support mediated these associations. At waves 1 and 2, ∼15 years later, participants from the Adult Health and Behavior study (N = 404, 42% male, 89% white, W1 mean age=45.5 years) self-reported their social network composition, positive affect, and social support and provided blood samples to assess IL-6, CRP, TNF-α, and suPAR. Social network composition was characterized using a ratio of the number of close to peripheral members. The social network ratio increased over time and close relationships were more likely to be retained than peripheral (p = .007). Inflammatory markers IL-6, CRP, and suPAR increased over time, but TNF-α slightly decreased (ps < .027). An increase in the social network ratio, indicating more narrowing, was associated with lower levels of TNF-α at W2 (b=-.019, SE=.008, p = .013) and retaining more peripheral relationships was associated with higher levels of CRP (b=.015, SE=.006, p = .015), adjusting for W1 inflammation. Positive affect and social support did not operate as mediators. Our findings provide initial support that network narrowing may be physiologically beneficial in the transition from midlife to late adulthood.