Development of Prediction Models for Mortality Risk After Colorectal Cancer Surgery

Researchers developed prediction models using preoperative data to assess short-term mortality risk following colorectal cancer surgery. The models showed good discrimination and calibration, with an area under the receiver operating characteristic curve of 0.88, 0.878, and 0.861 for 30-day, 90-day, and 1-year mortality, respectively. Subgroup analysis did not improve discrimination or calibration. Combining all operated patients resulted in better performing models than separate subgroups, with an overall mortality incidence of 4.48%, 6.64%, and 12.8% for 30-day, 90-day, and 1-year mortality, respectively.