Associated factors analysis and prediction model development for self-perceived discrimination (SPD) in colorectal cancer inpatients: a cross-sectional study

Self-perceived discrimination (SPD) is prevalent among colorectal cancer (CRC) inpatients. This study aimed to investigate SPD prevalence, identify associated factors, and develop a prognostic nomogram. This cross-sectional study enrolled 464 CRC inpatients (January 2023–May 2025). Data were collected using standardized questionnaires (GSQ, FS, SCS, CCSPDS). Participants were dichotomized into SPD(+) and SPD(−) groups based on a median score >42. The cohort was divided into a training (n=326) and a validation set (n=138). The overall SPD prevalence was 51.51%. Multivariable logistic regression identified four independent predictors: colostomy bag presence, gender, financial burden, and psychological well-being. The developed nomogram demonstrated strong discrimination, with an area under the curve (AUC) of 0.839 (95% CI: 0.796–0.881) in the training set and 0.804 (95% CI: 0.729–0.878) in the validation set. Calibration and decision curve analysis (DCA) confirmed the model’s clinical utility. SPD is common among CRC patients. The proposed nomogram provides an accurate, practical tool for medical staff to identify high-risk patients and implement targeted psychological interventions.