Abstract
AIMS: Decades of research show that people with schizophrenia have an increased risk of death from cancer; however, the relationship between schizophrenia and cancer incidence remains less clear. This population-based study investigates the incidence of seven common types of cancer among people with a hospital diagnosis of schizophrenia and accounting for the effects of age, sex and calendar time. METHODS: This population-based study used 1990-2013 data from three nationwide Swedish registries to calculate the incidence (in total, by age group and by sex) of any cancer and of lung, oesophageal, pancreatic, stomach, colon, (in men) prostate and (in women) breast cancer in 111 306 people with a hospital diagnosis of schizophrenia. The incidence in people with diagnosed schizophrenia was compared with the incidence in the general population. Risk estimates accounted for the effects of calendar time. RESULTS: In 1 424 829 person-years of follow-up, schizophrenia did not confer an overall higher cancer risk (IRR 1.02, 95% CI 0.91-1.13) but was associated with a higher risk for female breast (IRR 1.19, 95% CI 1.12-1.26), lung (IRR 1.42, 95% CI 1.28-1.58), oesophageal (IRR 1.25, 95% CI 1.07-1.46) and pancreatic (IRR 1.10, 95% CI 1.01-1.21) and a lower risk of prostate (IRR 0.66, 95% CI 0.55-0.79) cancer. Some age- and sex-specific differences in risk were observed. CONCLUSIONS: People with schizophrenia do not have a higher overall incidence of cancer than people in the general population. However, there are significant differences in the risk of specific cancer types overall and by sex calling for efforts to develop disease-specific prevention programmes. In people with schizophrenia, higher risk generally occurs in those <75 years.