Comparative Diagnostic Accuracy of the Braden and Cubbin/Jackson Scales for Predicting Pressure Injury Development in an Adult Intensive Care Unit: A Prospective Cohort Study

This study evaluated and compared the diagnostic accuracy of the Braden and Cubbin/Jackson Pressure Injury Risk Scales for predicting pressure injury development in adult ICU patients. In this prospective cohort study, 153 patients admitted to an adult ICU between September and November 2025 were assessed within 24 h using both scales and followed during their ICU stay. Diagnostic accuracy indices were calculated across cut-off points and optimal cut-offs were selected using the Youden J index. ROC curves were compared using the DeLong test. Precision–recall analysis and adjusted logistic regression analyses were also performed. Pressure injury developed in 37 patients (24.2%). The optimal Braden cut-off was ≤ 12, with sensitivity 43.2%, specificity 67.2%, PPV 29.6%, NPV 78.8%, LR+ 1.32, LR− 0.84 and Youden J 0.105. The optimal Cubbin/Jackson cut-off was ≤ 35, with sensitivity 73.0%, specificity 50.9%, PPV 32.1%, NPV 85.5%, LR+ 1.49, LR− 0.53 and Youden J 0.238. AUCs were 0.546 (95% CI: 0.439–0.653) and 0.605 (95% CI: 0.503–0.707), respectively; the difference was not statistically significant (p = 0.187). Precision-recall analysis showed limited predictive performance and adjusted logistic regression analyses indicated that neither cut-off was a statistically significant independent predictor of pressure injury development (p > 0.05). Both scales showed limited discriminatory accuracy. Cubbin/Jackson performed numerically better, but not significantly so; therefore, both scales should support, rather than replace, comprehensive clinical judgement in intensive care nursing.