Objective To explore the influencing factors on the clinical outcome of pre-hospital pressure injury (PPI) in orthopedic patients and to construct and validate a predictive model.
Methods A retrospective analysis was conducted on the clinical data of 546 orthopedic inpatients. Univariate analysis was used to screen independent variables related to the outcome of orthopedic PPI. Aftervariable screening, a multivariate predictive model was constructed based on the generalized estimating equation (GEE) framework (with patient ID as the clustering variable to correct for the non-independence of data from multiple woundsin the same patient). The area under the curve (AUC) of the receiver operating characteristic (ROC) curve was used to evaluate the model's discrimination ability, and the calibration curve and Hosmer-Lemeshow test were used to evaluate the model's calibration ability.
Results Multivariate GEE model analysis showed that a history of diabetes (OR=0.458, 95%CI, 0.247 to 0.859), friction and shear force score (OR=0.737, 95%CI, 0.614 to 0.922), whether orthopedic external fixation devices were used before admission (OR=0.314, 95%CI, 0.172 to 0.533), whether the wound was infected (OR=0.401, 95%CI, 0.238 to 0.669), and the NRS 2002 nutritional risk screening score (OR=0.514, 95%CI, 0.311 to 0.856) were independent influencing factors for the outcome of orthopedic PPI. The AUC of the predictive model was 0.858 when applied to the training set and 0.732 when applied to the validation set. The Hosmer-Lemeshow test showed χ2=5.217, P=0.513, and the calibration curve indicated good model fit.
Conclusion The outcome of PPI in orthopedic patients is influenced by various factors such as a history of diabetes, friction and shear force, and wound infection. The risk predictive model constructed based on these influencing factors has high sensitivity and specificity and can provide support for community screening and early warning as well as precise hospital interventions.
DownLoad: