Nine RCTs (13 articles) were included (n=182,323, range 1,621 to 39,876).
There was no statistically significant difference between the beta-carotene group and the placebo group in the incidence of all-site cancer (eight RCTs) and of stomach (seven RCTs), pancreas (four RCTs), colon-rectum (seven RCTs), prostate (five RCTs), breast (four RCTs) and skin (six RCTs) cancers. The risk of lung cancer was significantly higher in the beta-carotene than the placebo group (RR 1.13, 95% CI 1.04 to 1.24; eight RCTs). There was no statistically significant heterogeneity for any of these analyses.
In subgroup analyses, among smokers and asbestos workers beta-carotene significantly increased the risk of all-site cancer (RR 1.08, 95% CI 1.01 to 1.15; two RCTs), lung cancer (RR 1.20, 95% CI 1.07 to 1.34; two RCTs) and stomach cancer (RR 1.54, 95% CI 1.08 to 2.19; one RCT). Beta-carotene at high doses significantly increased the risk of lung cancer (RR 1.16, 95% CI 1.06 to 1.27; six RCTs) and stomach cancer (RR 1.54, 95% CI 1.08 to 2.19; four RCTs) cancer. At low doses beta-carotene significantly decreased the risk of stomach cancer (RR 0.84, 95% CI 0.71 to 1.00; two RCTs).
Other findings were reported in the review.