Increasing evidence suggests that a history of diabetes mellitus (DM) may be associated with an increased risk of colorectal cancer (CRC). To provide a quantitative assessment of the association between DM and risk of CRC, We evaluated the relation between DM and incidence and mortality of CRC in a systematic review of cohort studies. Full publications of cohort studies were identified in MEDLINE, EMBASE and Science Citation Index Expanded, through February 28, 2011. Summary relative risks (SRRs) with 95% confidence intervals (CIs) were summarized using a random-effects model. Between-study heterogeneity was assessed using the Cochran's Q and I2 statistics. A total of 41 cohort studies (35 articles) were included in this systematic review. Combining 30 cohort studies which presented results on diabetes and CRC incidence, diabetes was associated with an increased incidence of CRC (SRRs 1.27, 95% CI: 1.21-1.34), with evident heterogeneity among studies (P=0.002, I2=48.4%). Subgroup analysis and meta-regression analysis by controlling the confounders showed that the increased incidence of CRC was independent of geographic locations, sex, family history of colorectal cancer, smoking, physical activity and body mass index. Diabetes was also positively associated with CRC mortality (SRR 1.20, 95% CI: 1.03-1.40), with evidence of heterogeneity between studies (P<0.001, I2=81.4%). Results from this systematic review support that compared to non-diabetic individuals, diabetic individuals have an increased risk of CRC.