A predictive nomogram for in-ICU deterioration of stage 1 pressure injuries: a retrospective study

Background Preventing Stage 1 pressure injuries (PIs) from worsening in the ICU is a key clinical challenge. Early prediction of high-risk patients enables targeted prevention. We aimed to develop a model for this progression using admission data. Methods In this retrospective cohort study, eligible ICU patients with Stage 1 pressure injuries were randomly allocated into training (70%) and validation (30%) sets. Predictors were selected using LASSO regression. A multivariable logistic regression model was constructed and visualized as a nomogram. Model performance was evaluated by discrimination (AUC), calibration, and clinical utility (decision curve analysis). Results A total of 278 patients were randomly divided into training ( n = 195) and validation ( n = 83) sets. LASSO regression identified three independent predictors: diabetes (OR: 2.451; 95% CI: 1.139–5.274), maximum norepinephrine dose (OR: 2.051; 95% CI: 1.322–3.182), and albumin level at ICU admission (OR: 0.834 per unit increase; 95% CI: 0.776–0.897). The nomogram demonstrated excellent discrimination, with an AUC of 0.800 (95% CI: 0.736–0.865) in the training set and 0.785 (95% CI: 0.675–0.895) in the validation set. Good calibration and clinical utility were confirmed. Conclusions A nomogram incorporating three readily available factors at ICU admission effectively predicts the risk of Stage 1 PI progression. This tool is designed to aid early risk stratification and could help guide preventive measures.