Background Anxiety has been shown to often precede depression in children and young adults. Only a small number of investigations have examined this form of comorbidity in older adults and the temporal relationship of these syndromes remains unclear. The objective was to verify whether trait anxiety predicts incident/recurrent depressive symptomatology in older adults independently of variables susceptible to explain this relationship in this population, such as cognitive complaints, subjective health, and baseline depressive symptoms. Methods A random sample of 4,649 individuals aged 65 years or older from the Three-City Study, a prospective longitudinal study with a 10-year follow-up, was used. Incident and recurrent depressive symptomatology were determined by Center for Epidemiological Studies Depression Scale cutoff scores. Anxiety was measured using the trait scale of the State-Trait Anxiety Inventory. Cox proportional hazards models were used to determine the independent risk of depressive symptomatology for baseline anxiety, cognitive complaints, subjective health, and depressive symptoms, adjusting for sociodemographic, mental health, and physical health covariates. Results Incident depressive symptomatology was independently predicted by baseline anxiety, depressive symptoms, cognitive complaints, and subjective health. Recurrent depressive symptomatology was independently predicted by baseline anxiety and depressive symptoms, but not by cognitive complaints and subjective health. Anxiety was associated with a higher risk of incident depressive symptomatology only in participants without a history of a major depressive episode, and with a higher risk of recurrent depressive symptomatology in men than in women. Conclusions Trait anxiety constitutes an important independent risk factor for subsequent depressive symptomatology in older adults.< Réduire