Objective To investigate the association between diabetes and pancreatic cancer to provide the basis for the early di‐agnosis and prevention of pancreatic cancer .Methods By formulating the retrieval stategy ,the databases of PubMed ,EMbase ,Ov‐id 、EBSCO ,CNKI and CBMdisc ,VIP ,Wanfang/database etc .were retrieve by computer from their establishment to July of 2013 .At the same time ,the relevant literature references were also retrieved .All case‐control studies(CCS) on the association between diabe‐tes and pancreatic cancer were included and their quality was assessed .The meta analysis was conducted by using Review Manager 5 .0 software .Results A total of 21 CSC were included ,finally 18 case‐control studies met the inclusion criteria for conducting meta analysis ,including 6 184 cases in the pancreatic cancer group and 10 296 cases in the control group .The meta analysis on the pan‐creatic cancer group and the control group showed that the result of heterogeneity test from all research data was I2 = 74% ,χ2 =65 .38 ,P< 0 .01 ,the random effects model was adopted ,the pooled odds ratio(OR) was 3 .60 and 95% confidence interval(95% CI) was 3 .28 - 3 .94 ,the difference showed the statistic significance(P< 0 .05) ,which indicating that the risk suffering diabetes in the patients with pancreatic cancer was 3 .6 times of the normal population .Conclusion Diabetes is associated with pancreatic cancer and diabetes may be an early clinical manifestation of pancreatic cancer .