OBJECTIVES: As previous studies have reported contradictory findings regarding the relationship between schizophrenia and cancer, we evaluated the association between schizophrenia and cancer diagnoses.

METHODS: In this retrospective cohort study, the IQVIA Disease Analyzer database was utilized to examine the incidence of cancer among patients aged over 18 years diagnosed with schizophrenia in German general practices from 2005 to 2022. Patients with schizophrenia were compared with those without the condition, with adjustments made for age, sex, index year of diagnosis, average annual practitioners visit frequency, and comorbidity. Kaplan-Meier curves were used to analyze the 10-year cumulative incidence of schizophrenia and cancer in total amongst patients with and without schizophrenia. Univariable Cox regression analysis was performed to calculate Hazard Ratios (HR) of cancer risk and their 95% confidence intervals (CI) of cancer in total and of specific cancer types.

RESULTS: Patients with schizophrenia (N=13.711) had a lower risk of cancer diagnosis compared to those without (N=68.555). Specifically, 10.4% of patients with schizophrenia and 12.5% of patients without the condition were diagnosed with cancer (p<0.001). Cox regression analysis showed a significant association between schizophrenia and subsequent cancer in the total population (HR: 0.82; 95% CI: 0.76-0.90), and among men (HR: 0.70; 95% CI: 0.61-0.80), but not among women (HR: 0.94, 95% CI: 0.84-1.04). Analyses stratified by cancer type and sex revealed a strong and significant association between schizophrenia and a decreased risk of prostate cancer in men (HR: 0.38; 95% CI: 0.24-0.61). Furthermore, there was also a negative association between schizophrenia and colorectal cancer risk, but statistical significance was not reached (HR: 0.58; 95% CI: 0.37-0.93).

CONCLUSIONS: This study demonstrates negative associations between schizophrenia and subsequent cancer, and more specifically in men for prostate and colorectal cancer. However, further research is required to explore the underlying reasons for these associations.