Abstract
Background
Identifying colorectal cancer associated risks is important for conducting a program for the survey and prevention of colorectal cancer.
Aim
To investigate the association between use of insulin or metformin with colorectal cancer (CRC) in type 2 diabetes (T2DM).
Design
Population-based cohort study.
Methods
Through analysis of National Health Insurance (NHI) database between 1998 and 2010 in Taiwan, we identified 66 324 T2DM patients aged ≥ 20 years and selected subjects without diabetes by 1: 1 randomly matching with the study cohort based on age, sex and index date. We followed up the participants until 31 December 2011 or when they withdrew from the NHI program.
Results
Compared with non-diabetic subjects, the T2DM patients exhibited an increased risk of CRC [adjusted HR (aHR) = 1.56, 95% confidence interval (CI) = 1.39–1.75], after adjustment for age, sex, urbanization level, comorbidities and examinations of colonoscopy, sigmoidoscopy, or stool occult blood test. Among the T2DM patients, insulin usage increased the risk of CRC (aHR = 1.86, 95% CI = 1.58–0–2.19) after adjustment for age, sex, urbanization level, comorbidities, metformin usage and examinations; nevertheless, metformin decreased the risk of CRC (aHR = 0.65, 95% CI = 0.54–0.77) after adjustment for age, sex, urbanization level, comorbidities, insulin usage and examinations. Compared with the non-insulin cohort, the risk of CRC tended to increase with the incremental dosage of insulin exposure.
Conclusion
Our population-based cohort study demonstrated an association between T2DM and CRC. Among the T2DM patients, insulin use was associated with an increased risk of CRC and metformin use was associated with a decreased risk of CRC. Inability to obtain information on several potential confounding factors, such as lifestyle and dietary habits, is the major limitation of the study.