Suicide risk in prostate cancer patients: epidemiological trends and a predictive modeling

TL;DR

Prostate cancer patients have a higher suicide risk in the first six months after diagnosis, and a new model helps predict this risk for better intervention.

Contribution

A nomogram model for predicting suicide risk in prostate cancer patients using LASSO and Cox regression was developed and validated.

Findings

Suicide risk is highest in the first six months after diagnosis (SMR = 26.007).

A nomogram model with nine risk factors showed strong predictive performance (C-index 0.714).

Abstract

Prostate cancer (PCa) is a common malignancy among men. Although survival rates have steadily improved, some patients still face an increased risk of suicide following diagnosis. Understanding the temporal patterns and key risk factors associated with suicide after a PCa diagnosis is crucial for developing effective intervention strategies. However, comprehensive models for individualized suicide risk prediction remain limited.

This study utilized data from the SEER database covering the period from 2010 to 2021. First, the suicide mortality trend among PCa patients was assessed by calculating the standardized mortality ratio (SMR). Subsequently, variable selection was performed using the least absolute shrinkage and selection operator (LASSO) regression, combined with Cox proportional hazards regression to further identify independent risk factors associated with suicide risk. A total of 404,120 eligible patients were included for model construction and randomly divided into training and validation cohorts at a 7:3 ratio. Based on the results of multivariable Cox regression, a nomogram prediction model was developed. The model’s performance was evaluated and validated using the concordance index (C-index), receiver operating characteristic (ROC) curves, and calibration curves.

Among 475,139 PCa patients identified, the suicide risk was highest during the first six months after diagnosis (SMR = 26.007, 95% CI: 19.303–34.287) and declined significantly after five years. Nine independent risk factors, including age, race, and Gleason score, were incorporated into the nomogram model. The model showed strong predictive performance, with C-indices of 0.714 and 0.685 in the training and validation cohorts, respectively. ROC curve analysis demonstrated good discrimination ability, and calibration curves indicated excellent agreement between predicted and observed outcomes.

This study highlights the heightened risk of suicide among PCa patients, particularly within the first six months following diagnosis, underscoring the urgent need for early psychological intervention. The developed nomogram provides an effective tool for personalized suicide risk assessment, enabling clinicians to identify high-risk patients and implement targeted preventive strategies. Integrating mental health care into routine oncology practice may significantly improve overall patient outcomes.

Not applicable.

The online version contains supplementary material available at 10.1186/s12888-026-07806-7.

Figures