© 2020 Background: This study aimed to examine associations of personality with generalized anxiety disorder (GAD) and physical activity (PA), PA with GAD, and PA mediates associations between personality and incident GAD. Methods: Participants aged ¿50 years completed the 60-item NEO-Five Factor Inventory questionnaire to assess personality and short-form International Physical Activity Questionnaire at baseline, and the Composite International Diagnostic Interview ¿ Short Form to clinically assess GAD at baseline and 2, 4, and 6 years later. Participants who had GAD at baseline or reported having ever been told by a doctor that they had anxiety were excluded from analyses. Binary logistic regression quantified associations of the `Big Five¿ personality traits with PA and incident GAD, and associations of PA with incident GAD (i.e., GAD at any point during follow-up). The `counterfactual approach¿ identified potential mediating effects of PA in the associations between personality traits and incident GAD. Results: Participants (n = 4582; 53.7% female) were aged 64.38±8.88 years. Incidence of GAD was 2.95% (n = 135). Extraversion (OR=1.160, 95%CI=1.087¿1.237), openness (1.113, 1.043¿1.188), and conscientiousness (1.083, 1.015¿1.155) were positively associated with physical activity. Neuroticism was positively (2.335, 1.945¿2.803), and extraversion (0.700, 0.563¿0.797), conscientiousness (0.826, 0.693¿0.985), and PA (0.655, 0.451¿0.952) were inversely, associated with the incident GAD. Approximately 8.7% of the effect of extraversion and 8.8% of the effect of conscientiousness on GAD was due to mediation by PA only. Limitations: PA was self-reported Conclusions: Personality screening may help to identify older adults at-risk of anxiety who would benefit from participation in physical activity interventions.