Different intervention strategies for preventing type 2 diabetes mellitus in China: A systematic review and network meta-analysis of randomized controlled trials.

Different strategies are increasingly used for early intervention in prediabetes in China, but the effects of these strategies on incident diabetes have not yet been confirmed. The aim of the present study was to assess systematically the effects of different strategies for preventing diabetes, aimed at Chinese people with prediabetes. Seven electronic databases were searched to identify eligible trials published from inception to September 20, 2016. Randomized controlled trials with a minimum follow-up duration of 6 months were included. Standard pairwise meta-analysis with a random-effects model and network meta-analysis with a frequentist framework were performed. A total of 63 studies, including 11 intervention strategies, were included. Compared with placebo, all strategies, except for lipid-affecting drugs and sitagliptin, reduced the rate of incident diabetes with different levels of effectiveness, ranging from 0.18 (95% confidence interval [CI] 0.12, 0.27) to 0.39 (95% CI 0.20, 0.75). Ranking probability analysis indicated that metformin and β-cell-stimulating drugs reduced the risk of diabetes most, with probabilities of 87.4% and 81%, respectively. Ethnicity and cultural factors should be considered for diabetes prevention. Most of the included trials were of poor methodological quality, however, and the results should be interpreted with caution.