Affect variability and inflammatory markers in midlife adults

Objective: Higher affect variability (the extent to which individuals vary in their affect over time) has been associated with poorer health indicators, but associations with inflammation are less well understood. The purpose of the present study was to examine whether affect variability was associated with inflammation in ways consistent with the stability theory or the fragile positive affect theory, and whether associations were linear or nonlinear. Method: In a racially diverse sample (N = 231; Aged 25-65; 65% female; 62% Black; 25% Hispanic), we examined whether positive affect (PA) and negative affect (NA) variability exhibited linear or quadratic associations with circulating inflammatory cytokines (a composite measure comprised of IL-1β, IL-4, IL-6, IL-8, IL-10, TNF-α, IFN-γ), and C-reactive protein (CRP) and whether person-mean affect moderated these associations. Affective states were assessed using ecological momentary assessments (EMAs) 5 times per day for 2 weeks, with a blood draw at the end of the EMA period. Individual standard deviations of affective states indexed affect variability. Results: A quadratic association indicated that moderate NA variability was associated with lower CRP. There was evidence of significant moderation by linear associations with PA only: For those with higher person-mean PA, PA variability was positively associated with the cytokine composite. Both personmean PA and person-mean NA moderated quadratic associations, such that for those with high person-mean affect, both high and low affect variability was associated with systemic inflammation. Conclusion: Results are in line with fragile affect theory suggesting that associations between affect variability and health indicators may vary by person-mean affect.