CONTEXT: Previous research has suggested that persons with cataract have an increased risk of death. OBJECTIVE: To compare the mortality experience of patients with cataract who elect surgery, patients with cataract who do not elect surgery, and patients without cataract independent of potentially confounding risk factors. DESIGN: Cohort study. SETTING: Ophthalmology and optometry clinics affiliated with the Callahan Eye Foundation Hospital in Birmingham, Alabama. PATIENTS: 384 persons with and without cataract. MAIN OUTCOME MEASURE: Mortality. RESULTS: Of the 384 study subjects, 286 had cataract, of whom 200 elected to have cataract surgery. Patients with cataract who did and did not elect surgery had significantly higher mortality compared to those without cataract (crude mortality rate ratio (MRR) 3.9 (95% confidence interval (CI) 1.5-9.8) and 7.3 (95% CI 2.8-19.1), respectively). After adjustment for age, gender, race, education, chronic medical conditions, smoking, drinking, depression, and cognitive status, the no-surgery cataract group had an elevated mortality rate (MRR 3.2 (95% 1.2-9.0)), compared to the no-cataract group, with a borderline elevation in MR for the surgery group (MRR 2.0 (95% 0.8-5.9). Limiting the study population to non-diabetics or those without concurrent eye conditions (glaucoma, maculopathy, retinopathy) did not materially influence the adjusted MRRs although the precision of the estimates was reduced. CONCLUSIONS: The results suggest that older persons with cataract, in particular those who decline surgery, have an increased risk of death, supporting the hypothesis that age-related cataract reflects systemic as well as localized ocular disease.