Objective: the aim of this study was to examine the bidirectional association of vision loss (VL) and hearing loss (HL) with anxiety over a 12-year period. Design: this was a prospective population-based study. Setting: community-dwelling French adults. Participants: the study included 3,928 adults aged 65 and above from the Three-City study. Methods: the relationships of VL, as assessed by near visual acuity and self-reported HL to a diagnosis of generalised anxiety disorder (GAD) were assessed over 12 years. A further objective was to explore whether sensory loss has a differential relationship with GAD than with anxiety symptoms, assessed by the Spielberger's State-Trait Anxiety Inventory. Results: at baseline, HL [odds ratio (OR) = 1.41, 95% confidence interval (CI) 1.02-1.96, P = 0.04], but not mild or moderate to severe VL, was associated with self-reported anxiety symptoms (OR = 1.07 95% CI 0.63-1.83, P = 0.80; OR = 0.66 95% CI 0.12-2.22, P = 0.50, respectively). Neither vision nor HL was significantly associated with incident GAD. Baseline GAD was related to increased risk of incident HL (OR = 1.17, 95% CI 1.07-1.28, P < 0.001), but not mild or moderate to severe vision loss (OR = 1.01, 95% CI 0.96-1.06, P = 0.81; OR = 0.97, 95% CI 0.89-1.05, P = 0.45, respectively). Conclusions: increased anxiety symptoms were observed in older adults with HL, whereas we found no evidence for an association between VL and anxiety. Anxiety was prospectively associated with increased risk of reporting HL. Improved detection of anxiety in older adults with HL may improve quality of life.