Summary Background Tamoxifen and raloxifene reduce the risk of breast cancer in women at elevated risk of disease, but the duration of the effect is unknown. We assessed the effectiveness of selective oestrogen receptor modulators (SERMs) on breast cancer incidence. Methods We did a meta-analysis with individual participant data from nine prevention trials comparing four selective oestrogen receptor modulators (SERMs; tamoxifen, raloxifene, arzoxifene, and lasofoxifene) with placebo, or in one study with tamoxifen. Our primary endpoint was incidence of all breast cancer (including ductal carcinoma in situ) during a 10 year follow-up period. Analysis was by intention to treat. Results We analysed data for 83 399 women with 306 617 women-years of follow-up. Median follow-up was 65 months (IQR 54–93). Overall, we noted a 38% reduction (hazard ratio [HR] 0·62, 95% CI 0·56–0·69) in breast cancer incidence, and 42 women would need to be treated to prevent one breast cancer event in the first 10 years of follow-up. The reduction was larger in the first 5 years of follow-up than in years 5–10 (42%, HR 0·58, 0·51–0·66; p<0·0001 vs 25%, 0·75, 0·61–0·93; p=0·007), but we noted no heterogeneity between time periods. Thromboembolic events were significantly increased with all SERMs (odds ratio 1·73, 95% CI 1·47–2·05; p<0·0001). We recorded a significant reduction of 34% in vertebral fractures (0·66, 0·59–0·73), but only a small effect for non-vertebral fractures (0·93, 0·87–0·99). Interpretation For all SERMs, incidence of invasive oestrogen (ER)-positive breast cancer was reduced both during treatment and for at least 5 years after completion. Similar to other preventive interventions, careful consideration of risks and benefits is needed to identify women who are most likely to benefit from these drugs. Funding Cancer Research UK.