BACKGROUND: Recent information suggests that the Age-Related Eye Disease Study (AREDS) supplement, enhanced intake of docosahexaenoic acid (DHA) and eicosapentaenoic acid (EPA), and reducing dietary glycaemic index (dGI) are protective against advanced age-related macular degeneration (AMD). METHODS: Dietary information was collected at baseline, and fundus photograph grades were obtained during the 8-year trial period from 2924 eligible AREDS AMD trial participants. Using the eye as the unit of analysis and multifailure Cox proportional-hazards regression, the risk of AMD progression was related to dietary intake in the four arms of the trial. RESULTS: Independent of AREDS supplementation, higher intakes of DHA (> or =64.0 vs <26.0 mg/day) (hazard ratio (HR) = 0.73, 95% confidence interval (CI), 0.57 to 0.94), EPA (> or =42.3 vs <12.7 mg/day) (HR = 0.74, 95% CI 0.59 to 0.94), and lower dGI (dGI, <75.2 vs > or =81.5) (HR = 0.76, 95% CI 0.60 to 0.96) were associated with a lower risk for progression to advanced AMD. Participants consuming a lower dGI and higher DHA or EPA had the lowest risk (p value for synergistic interaction <0.001). Only participants in the "placebo" (p value for antagonistic interaction = 0.006) benefited from a higher DHA intake against early AMD progression (HR = 0.58, 95% CI 0.37 to 0.92; P(trend) = 0.01). CONCLUSIONS: The findings show an association of consuming a diet rich in DHA with a lower progression of early AMD. In addition to the AREDS supplement, a lower dGI with higher intakes of DHA and EPA was associated with a reduced progression to advanced AMD. Trial registration number: NCT00000145.