OBJECTIVE: To evaluate the association between refractive errors and age-related macular degeneration (AMD). MAIN OUTCOME MEASURES: A clear understanding of the relationship between refractive error and AMD provides insights into the pathophysiology of AMD. METHODS: We searched PubMed and Embase from their inception to July 2012 for population-based studies with data on refractive error and AMD assessed from retinal photographs at baseline and follow-up. We performed separate meta-analyses for cross-sectional studies and cohort studies using adjusted odds ratios (ORs) and hazard ratios (HRs) under random effects models, respectively. RESULTS: Analysis of the 6 cross-sectional studies showed that hyperopia was associated with higher odds of prevalent AMD (pooled OR hyperopia vs. emmetropia: 1.16; 95% confidence interval [CI], 1.04-1.29) and that myopia was associated with lower odds of prevalent AMD (pooled OR myopia vs. emmetropia: 0.75; 95% CI, 0.61-0.92). Analysis from the 3 cohort studies showed nonsignificant associations. Analysis of the 5 cross-sectional and 2 cohort studies showed that each diopter increase in spherical equivalent was associated with increased odds of both prevalent (pooled OR, 1.09; 95% CI, 1.06-1.12) and incident (pooled HR, 1.06; 95% CI, 1.02-1.10) AMD. In 3 cross-sectional studies with data on axial length, each millimeter increase in axial length was associated with a decreased odd of prevalent AMD (pooled OR, 0.76; 95% CI, 0.69-0.85). CONCLUSIONS: Refractive error is associated with AMD, although a temporal relationship cannot be determined on the basis of current evidence. Ophthalmologists should be aware that risk of AMD clinically seems to vary by refractive status. FINANCIAL DISCLOSURE(S): The author(s) have no proprietary or commercial interest in any materials discussed in this article.