The aim of this network meta-analysis is to compare bone mineral density (BMD) changes among different osteoporosis prevention interventions in postmenopausal women. We searched MEDLINE, Embase and Cochrane Library from inception to 24 February 2019. Included studies were randomised controlled trials (RCT) comparing the effects of different treatments on BMD in postmenopausal women. Studies were independently screened by six authors in three pairs. Data were extracted independently by two authors and synthesised using Bayesian random-effects network meta-analysis. The results were summarised as mean difference in BMD and surface under the cumulative ranking (SUCRA) of different interventions. A total of ninety RCT (10 777 participants) were included. Ca, vitamin D, vitamin K, oestrogen, exercise, Ca + vitamin D, vitamin D + vitamin K and vitamin D + oestrogen were associated with significantly beneficial effects relative to no treatment or placebo for lumbar spine (LS). For femoral neck (FN), Ca, exercise and vitamin D + oestrogen were associated with significantly beneficial intervention effects relative to no treatment. Ranking probabilities indicated that oestrogen + vitamin D is the best strategy in LS, with a SUCRA of 97·29 % (mean difference: +0·072 g/cm2 compared with no treatment, 95 % credible interval (CrI) 0·045, 0·100 g/cm2), and Ca + exercise is the best strategy in FN, with a SUCRA of 79·71 % (mean difference: +0·029 g/cm2 compared with placebo, 95 % CrI –0·00093, 0·060 g/cm2). In conclusion, in postmenopausal women, many interventions are valuable for improving BMD in LS and FN. Different intervention combinations can affect BMD at different sites diversely.
