We conducted a randomized, double-blind, controlled trial to examine the efficacy of retinol supplementation on the incidence of first new nonmelanoma skin cancer in moderate-risk subjects. A total of 2297 free-living subjects were enrolled; subjects resided in Arizona (median age, 63 years) and had a history of more than 10 actinic keratoses and at most 2 squamous cell carcinoma (SCC) or basal cell carcinoma (BCC) skin cancers. Subjects were randomly assigned to receive oral retinol (25,000 IU) or placebo supplementation daily for up to 5 years. The primary end points for the trial were time to first new SCC or BCC. During a median follow-up time of 3.8 years, we found that 526 subjects had a first new skin cancer. Comparing retinol-supplemented subjects with placebo-supplemented subjects showed a hazard ratio for first new SCC of 0.74 (95% confidence interval, 0.56-0.99; P = 0.04). The hazard ratio of first new BCC for the retinol-supplemented subjects compared with those receiving placebo was 1.06 (95% confidence interval, 0.86-1.32; P = 0.36). Potentially adverse symptoms that were judged to be associated with retinol were rare (approximately 1% higher in the retinol group than in the control group). Therefore, we concluded that daily supplementation with 25,000 IU of retinol was effective in preventing SCC, although it did not prevent BCC.