Development and validation of an intensive care unit acquired weakness prediction model: A cohort study

Abstract

BackgroundAt present, intensive care unit acquired weakness (ICU-AW) has become an important health care issue. The aim of this study was to develop and validate an ICU-AW prediction model for adult patients in intensive care unit (ICU) to provide a practical tool for early clinical diagnosis.MethodsAn observational cohort study was conducted including 400 adult patients admitted from September 2021 to June 2022 at an ICU with four ward at a medical university affiliated hospital in China. The Medical Research Council (MRC) scale was used to assess bedside muscle strength in ICU patients as a diagnostic basis for ICUAW. Patients were divided into the ICU-AW group and the no ICU-AW group and the clinical data of the two groups were statistically analyzed. A risk prediction model was then developed using binary logistic regression. Sensitivity, specificity, and the area under the curve (AUC) were used to evaluate the predictive ability of the model. The Hosmer-Lemeshow test was used to assess the model fit. The bootstrap method was used for internal verification of the model. In addition, the data of 120 patients in the validation group were selected for external validation of the model.ResultsThe prediction model contained five risk factors: gender (OR: 4.31, 95% CI: 1.682–11.042), shock (OR: 3.473, 95% CI: 1.191–10.122), mechanical ventilation time (OR: 1.592, 95% CI: 1.317–1.925), length of ICU stay (OR: 1.085, 95% CI: 1.018–1.156) and age (OR: 1.075, 95% CI: 1.036–1.115). The AUC of this model was 0.904 (95% CI: 0.847–0.961), with sensitivity of 87.5%, specificity of 85.8%, and Youden index of 0.733. The AUC of the model after resampling is 0.889. The model verification results showed that the sensitivity, specificity and accuracy were 71.4, 92.9, and 92.9%, respectively.ConclusionAn accurate, and readily implementable, risk prediction model for ICU-AW has been developed. This model uses readily obtained variables to predict patient ICU-AW risk. This model provides a tool for early clinical screening for ICU-AW.