Data were analyzed from 3834 participants (mean [SD] age at baseline, 64.0 [7.4] years; 2082 [54.3%] women) with a mean (SD) follow-up of 12.25 (2.60) years. Adjusting for demographic characteristics, baseline illness, health behaviors, and mean level and instability in negative affect, Cox proportional hazards regression showed that high positive affect instability was associated with greater mortality, with a hazard ratio of 1.25 (95% CI, 1.04-1.49; P = .02). Associations did not differ by age, suggesting that the increased mortality risk associated with high positive affect instability was not restricted to older ages.
