To investigate the correlation between diabetes and hip fracture, and to systematically review and analyze all available data to evaluate whether diabetes mellitus increases the risk of hip fracture.

According to the requirements of Meta analysis, we have a wide range of products and services, including Vip, Wanfang, Springer, Science Direct, CBM, PubMed, Embase, Cochrane, Library, and Web of Chinese and English databases such as CBM, CNKI, EMBASE, China Statistical and Analytical Network of Scientific and Technological Papers, Patent Database, etc. The literatures of controlled experimental studies on hip fracture in diabetic patients and the general population were strictly evaluated and valid relevant data were included. The retrieval period was from January 1,2018 to the present. NOS literature quality assessment criteria were used to evaluate the quality of literatures that met the inclusion and exclusion criteria. Q test and I² statistic test were used to study the heterogeneity of the included literatures and sensitivity analysis of leave one out. Egger's test and Begg's test were used to evaluate the publication bias of the included literatures.

A total of 29 studies were included, including 8,453 193,subjects. There were 6,430,654 subjects from the general population, 110,752 patients with hip fracture, 2,022,539 patients with diabetes, and 33,954 patients with hip fracture. Diabetic patients had a higher risk of hip fracture than non-diabetic patients (OR=1.507, 95% CI: 1.238, 1.834, Z=4.09, P<0.1).

Meta-analysis shows that diabetes may increase the risk of hip fracture.