OBJECTIVES: There have been controversies on the efficacy and safety of human insulin compared to animal insulin. The aim of this study was to compare the efficacy and safety between human and animal insulin among Chinese patients. METHODS: A systematic literature search with key terms for identifying studies on human and animal insulin among Chinese population was performed using MEDLINE, China National Knowledge Infrastructure, Chinese Scientific Journals Database, Wan Fang database and Chinese Biological Medical Database. For each clinical outcome, meta-analysis was conducted when enough number of studies (≥3) meet inclusion criteria. Mean difference (MD) and risk ratio (RR) were pooled for continuous and count measurements, respectively. RESULTS: A total of 597 publications were retrieved and 21 studies were identified for inclusion, including 7 randomized controlled trials (RCT), 5 non-randomized controlled trials (NRCT), and 9 studies where patients switched from animal insulin to human insulin (SW). Compared with animal insulin, human insulin was associated with significantly less daily dose, with MD(U/d) [CI] of -9.67 [-12.35,-6.99] for RCT, -9.19 [-10.01,-8.36] for NRCT, and -10.06 [-14.79,-5.33] for SW, as well as lower incidence of hypoglycemia, with RR of 0.29 [0.15,0.56] for RCT, 0.66 [0.51,0.86] for NRCT and 0.27 [0.11,0.67] for SW. In the analyses of RCTs, human insulin also had lower of local swelling and induration (RR: 0.09 [0.02, 0.39]) while the incidence difference of allergy was not statistically significant (RR: 0.19 [0.03, 1.09]). In addition, patients had significantly larger decrease of HbA1c when switching from animal insulin to human insulin with MD of -2.42% [-3.83, -1]. CONCLUSIONS: Results from this meta-analysis suggest that human insulin may show more benefit on efficacy and safety among Chinese patients with diabetes compared to animal insulin. More detailed prospective studies are warranted to further explore this comparison.