Diabetes, insulin use, and gastric cancer: a population-based analysis of the Taiwanese.

BACKGROUND

The association between diabetes and gastric cancer is rarely studied, and the risk associated with insulin use is not known.

MATERIALS AND METHODS

Gastric cancer prevalence rates in 2005 were calculated in 329,198 insurants aged 45 years or older and without type 1 diabetes from a random sample of 1,000,000 insurants of the National Health Insurance in Taiwan. Logistic regression evaluated whether diabetes or insulin use could be an independent risk factor after adjustment for age, sex, occupation, and living region. Sensitivity analyses were performed by excluding patients with diabetes duration <5 years and by excluding patients with cancers other than gastric cancer.

RESULTS

A total of 1464 cases of gastric cancer were identified. The prevalence of gastric cancer in diabetic and nondiabetic subjects were 595.0 and 387.7 per 100,000 population (P<0.0001), respectively. In diabetic patients, the respective prevalence of gastric cancer in insulin users and nonusers were 656.0 and 589.5 (P=0.4743). Adjusted odds ratio for diabetic versus nondiabetic subjects was 1.139 (1.022 to 1.270), and for insulin users versus nonusers in the diabetic patients was 1.002 (0.745 to 1.346). Age and male sex were also associated with significantly higher risk of gastric cancer, but the association with occupation and living region were not consistent. Results from sensitivity analyses were similar.

CONCLUSIONS

Diabetic patients may have a 14% higher risk of gastric cancer. However, insulin use is not related to the increased risk.