LAUSR

PURPOSE Age-related eye disease may be associated with cognitive decline, but the scientific literature has not been consistent. Furthermore, no studies have been able to explain the relationship. Our objective was to assess whether older adults with age-related macular degeneration (AMD) or glaucoma performed worse on 6 cognitive tests compared with older adults with normal vision and, if so, to understand why. DESIGN Cross-sectional analysis of hospital-based study (Maisonneuve-Rosemont Hospital Ophthalmology Clinics, Montréal, Canada). PARTICIPANTS Three hundred thirty-six adults 65 years of age or older with either AMD, glaucoma, or normal vision. METHODS Cognition was measured with 6 cognitive tests administered orally. Activity levels were measured using the Victoria Longitudinal Study Activity Lifestyle Questionnaire. Visual acuity and visual field were measured. Multiple linear regression was used. Mediation was assessed using structural equation modeling. MAIN OUTCOME MEASURES Results of the verbal fluency test (animal and letter versions), the digit span test (forward and backward versions), and the logical memory test (immediate and delayed recall). RESULTS People with glaucoma showed lower scores on 3 cognitive tests than the group with normal vision: the digit span forward and backward tests (β = -0.8 [95% confidence interval (CI), -1.5 to -0.2] and β = -0.7 [95% CI, -1.3 to -0.1], respectively) and the logical memory test with immediate recall (β = -1.3 [95% CI, -2.4 to -0.2]). Activity levels statistically significantly mediated the relationship between glaucoma and the digit span forward test (P = 0.043; percentage of the total effect mediated, 17%). CONCLUSIONS People with glaucoma showed lower scores on cognitive tests that may depend on verbal working memory and encoding. If confirmed in longitudinal studies, interventions should be developed that are appropriate for a visually impaired population to slow this cognitive decline.