Aims To evaluate the association between glucagon‐like peptide‐1 receptor agonists (GLP‐1 RAs) and the risk of bone fracture in patients with type 2 diabetes mellitus (T2DM). Materials and methods We conducted a systematic literature search in PubMed, Embase, the Cochrane Library, and Web of Science from inception to 28 February 2018 and identified eligible randomized controlled trials. The following data were extracted from each study: first author, year of publication, sample size, patient characteristics, study design, intervention drug, control drug, follow‐up time, and incident bone fracture events. A meta‐analysis was conducted using Review Manager 5.3 software to calculate the odds ratio (OR) and 95% confidence intervals (CI) for dichotomous variables. Results A total of 38 studies with 39 795 patients with T2DM were included. There were 241 incident bone fracture cases (107 in the GLP‐1 RAs group and 134 in the control group). Compared with patients who received placebo and other anti‐diabetic drugs, those who received GLP‐1 RAs treatment showed a pooled OR of 0.71 (95% CI, 0.56‐0.91) for bone fracture. Subgroup analysis showed that treatments with liraglutide and lixisenatide were associated with significantly reduced risk of bone fractures (ORs, 0.56; 95% CI, 0.38‐0.81 and 0.55; 95% CI, 0.31‐0.97, respectively). However, other GLP‐1 RAs did not show superiority to placebo or other anti‐diabetic drugs. Moreover, these beneficial effects were dependent on the duration of GLP‐1 RAs treatment, only a GLP‐1 RAs treatment period of more than 52 weeks could significantly lower the risk of bone fracture in patients with T2DM (OR, 0.71; 95% CI, 0.56‐0.91). Conclusions Compared with placebo and other anti‐diabetic drugs, liraglutide and lixisenatide were associated with a significant reduction in the risk of bone fractures, and the beneficial effects were dependent on the duration of treatment.