Developing a model for predicting suicide risk among prostate cancer survivors DOI Creative Commons
Jie Yang, Haiming Liu,

Xiang Qu

et al.

Frontiers in Medicine, Journal Year: 2025, Volume and Issue: 12

Published: April 10, 2025

Given the significantly higher suicide risk among cancer survivors compared to general population, and considering that prostate make up largest group of survivors, it is imperative develop a model for predicting survivors. Clinical data patients were extracted from surveillance, epidemiology, end results (SEER) database randomly divided into training cohort validation in 7:3 ratio. Initial variable selection was performed using univariate Cox regression, Best Subset Regression (BSR), Least Absolute Shrinkage Selection Operator (LASSO). Variables be included final selected backward stepwise regression. Model performance evaluated Concordance Index (C-index), Receiver Operating Characteristic (ROC) curves, calibration curves. Data 238,534 obtained SEER database, which 370 (0.16%) died by suicide. Seven variables including age, race, marital status, household income, PSA levels, M stage, surgical status model. The demonstrated good discriminative ability both cohorts, with C-indices 0.702 0.688, respectively. ROC values at 3, 5, 10 years 0.727/0.644, 0.700/0.698, 0.735/0.708, Calibration curves indicated high degree consistency between predictions actual outcomes. High-risk had 3.5 times than low-risk (0.007 vs. 0.002, P < 0.001), finding supported entire cohort. A reliable predictive successfully established based on seven readily obtainable clinical predictors. This can effectively aid healthcare professionals quickly identifying high-risk timely implementation preventive interventions.

Language: Английский

Developing a model for predicting suicide risk among prostate cancer survivors DOI Creative Commons
Jie Yang, Haiming Liu,

Xiang Qu

et al.

Frontiers in Medicine, Journal Year: 2025, Volume and Issue: 12

Published: April 10, 2025

Given the significantly higher suicide risk among cancer survivors compared to general population, and considering that prostate make up largest group of survivors, it is imperative develop a model for predicting survivors. Clinical data patients were extracted from surveillance, epidemiology, end results (SEER) database randomly divided into training cohort validation in 7:3 ratio. Initial variable selection was performed using univariate Cox regression, Best Subset Regression (BSR), Least Absolute Shrinkage Selection Operator (LASSO). Variables be included final selected backward stepwise regression. Model performance evaluated Concordance Index (C-index), Receiver Operating Characteristic (ROC) curves, calibration curves. Data 238,534 obtained SEER database, which 370 (0.16%) died by suicide. Seven variables including age, race, marital status, household income, PSA levels, M stage, surgical status model. The demonstrated good discriminative ability both cohorts, with C-indices 0.702 0.688, respectively. ROC values at 3, 5, 10 years 0.727/0.644, 0.700/0.698, 0.735/0.708, Calibration curves indicated high degree consistency between predictions actual outcomes. High-risk had 3.5 times than low-risk (0.007 vs. 0.002, P < 0.001), finding supported entire cohort. A reliable predictive successfully established based on seven readily obtainable clinical predictors. This can effectively aid healthcare professionals quickly identifying high-risk timely implementation preventive interventions.

Language: Английский

Citations

0