Risk of suicide in cancer patients: nationwide cohort study

BACKGROUND: Diagnosis of cancer can be a stressful and life-threatening event that is associated with suicide risk. AIMS: To investigate how suicide risk changes over time after cancer diagnosis, and, specifically, when it becomes similar to that of matched controls. METHOD: Using a nationwide population-based database, we identified a total of 171 474 individuals aged ≥20 years newly diagnosed with cancer between 2009 and 2017 and 1:5 age- and sex-matched controls. We calculated adjusted hazard ratios (aHRs) with 95% confidence intervals of suicide in cancer patients for the full period and with a 1- to 5-year lag period. RESULTS: During a mean follow-up of 6.7 years, 0.3% of cancer patients (491 of 171 474) died by suicide, with incidence rates of 0.4 per 1000 person-years. Cancer patients had higher risk of suicide (aHR 1.64, 95% CI 1.48-1.81) compared with matched controls. Suicide risk remained higher than that of matched controls with a 1- or 2-year lag period (aHR 1.38, 95% CI 1.23-1.55 and aHR 1.32, 95% CI 1.16-1.50, respectively), but there was no significant difference with a 5-year lag period (aHR 1.13, 95% CI 0.93-1.38). However, those with haematologic cancers were at higher suicide risk than matched controls even 5 years after diagnosis (e.g. aHR 9.26, 95% CI 1.30-65.87 for Hodgkin lymphoma). CONCLUSIONS: In cancer patients, suicide risk remained elevated for several years after diagnosis, but decreased over time and became similar to that of matched controls after 5 years. However, the temporal pattern varied by cancer type, and suicide risk remained high for patients with haematological cancers. Suicide risk screening is necessary from the time of cancer diagnosis, even in long-term survivors.