PURPOSE: To examine the association of blood antioxidants with cataract. METHODS: Cross-sectional study of people aged >or=50 years identified from a household enumeration of 11 randomly sampled villages in North India. Participants were interviewed for putative risk factors (tobacco, alcohol, biomass fuel use, sunlight exposure, and socioeconomic status) and underwent lens photography and blood sampling. Lens photographs (nuclear, cortical, and posterior subcapsular) were graded according to the Lens Opacities Classification System (LOCS II). Cataract was defined as LOCS II grade >or=2 for any opacity or ungradable, because of dense opacification or history of cataract surgery. People without cataract were defined as LOCS II <2 on all three types of opacity, with absence of previous surgery. RESULTS: Of 1443 people aged >or=50 years, 94% were interviewed, 87% attended an eye examination, and 78% gave a blood sample; 1112 (77%) were included in the analyses. Compared with levels in Western populations, antioxidants were low, especially vitamin C. Vitamin C was inversely associated with cataract. Odds ratios (OR) for the highest (>or=15 micromol/L) compared with the lowest (<or=6.3 micromol/L) tertile were 0.64, (95% confidence interval [CI] 0.48-0.85; P < 0.01). Tertiles of zeaxanthin (P < 0.03), alpha-carotene (P < 0.05), and retinol (P < 0.02) were associated with decreased odds of cataract. In analysis of continuous data, significant inverse associations were found for vitamin C, zeaxanthin, lutein, lycopene, alpha- and beta-carotene, and beta-cryptoxanthin, but not for alpha- or gamma-tocopherol. CONCLUSIONS: Inverse associations were found between cataract and blood antioxidants in an antioxidant-depleted study sample.