Risk of diabetes in viral hepatitis B or C patients compared to that in noninfected individuals in Korea, 2002-2013: A population-based cohort study.

While the association between hepatitis C virus (HCV) infection and diabetes has been established, the relationship between hepatitis B virus (HBV) infection and diabetes remains unclear. Therefore, we compared the association between diabetes development in HBV, HCV and co-infected (HBV/HCV) patients to that in noninfected participants using population-based cohort data. We used the National Health Insurance Service-National Sample Cohort, which consists of 514 791 randomly selected persons among those who underwent health check-ups from 2002 to 2003 aged 40-79 years. Adults found to have HBV or HCV infection from 2002 to 2003, without a prior history of diabetes, were selected as subjects. Competing risk regression models were used to estimate cumulative incidence and hazards ratios (HRs) of diabetes development. The cumulative incidences, incidence densities and HRs of diabetes were highest in the co-infected group, followed by those in the HCV-, HBV- and noninfected groups. The 12-year cumulative incidences were as follows: 42.0% in HBV/HCV-, 32.9% in HCV-, 23.9% in HBV- and 18.3% in the noninfected groups. The incidence density per 1000 person-years was 55.0, 51.5, 38.2 and 28.2 for the HBV/HCV-, HCV-, HBV- and noninfected groups, respectively. The adjusted HRs for diabetes were 1.90, 1.68 and 1.41 for the HBV/HCV-, HCV- and HBV-infected groups, respectively. Our findings suggest that both HCV and HBV infections are associated with the development of diabetes; therefore, prevention of, screening for, and treatment of both may reduce the risk of diabetes in these patients.