Suicide Risk Prediction in Cancer Patients: a Retrospective Cohort Study

Cancer is a serious public health concern, with almost 10 million people dying from cancer in 2020. Previous studies have reported that cancer patients are more likely to die by suicide than the general public, especially in the six months to one year following cancer diagnosis. Since suicide is a result of the interaction of various factors such as socioeconomic factors, clinical characteristics, and environmental factors, it is necessary to construct a multivariate prediction model to predict the suicide risk in cancer patients.

A retrospective cohort of cancer patients based on the Surveillance, Epidemiology, and End Results (SEER) program database was used to assess the association of socioeconomic factors, clinical characteristics and meteorological factors with cancer patients' suicide, and to establish prediction model with multiple predictors for cancer patients. Another retrospective cohort conducted from Shandong Multi-Center Healthcare Big Data Platform (SMCHBDP) was used to verify the predictive ability and generalization ability of the prediction model.