Back

Development of a Nomogram for Predicting the Risk of Hospital-Acquired Pressure Injuries in Patients in the Cardiovascular Intensive Care Unit

Li, H.

2023-11-14 nursing
10.1101/2023.11.14.23298510 medRxiv
Show abstract

PurposeHospital-acquired pressure injuries (HAPIs) increase the medical burden of patients in the cardiovascular intensive care unit (CCU). Thus, identification of CCU patients with a risk for HAPIs is important. To establish a nomogram model for predicting the occurrence of HAPIs in patients in the CCU. MethodsThis was a retrospective cohort study in patients in the CCU at our hospital who developed HAPIs between January 2023 and June 2023. Patient data were extracted from the hospitals information management system. Risk factors for HAPIs were identified using univariate and multivariate logistic regression analyses and the screened indicators were integrated into a nomogram. The effectiveness of the nomogram was evaluated and verified using receiver operating characteristic curve and decision curve analysis (DCA). ResultsIn this study, a total of 161 patients were enrolled. Univariate logistic regression analysis showed that vasopressor use (OR=2.51, 95%CI=1.24-5.09, p=.010) and NT-proBNP (OR=2.92, 95%CI=1.45-5.92, p=.003), lactic acid (OR=9.43, 95%CI=4.15-21.41, p <.000), procalcitonin (OR=1.37, 95%CI=1.07-1.75, p=.012), D-dimer (OR=1.16, 95%CI=1.05-1.28, p= .004), and albumin (OR=0.88, 95%CI=0.81-0.95, p=.002) levels were independent risk factors for HAPIs. Multivariate logistic regression analysis showed that vasopressor use (OR=3.05, 95%CI=1.20-7.73, p=.019) and lactic acid (OR=12.05, 95%Cl=4.13-35.21, p <.000), procalcitonin (OR=1.30, 95%Cl=1.01-1.69, p=.043) and albumin (OR=0.82, 95%Cl=0.73-0.93, p=.002) levels were independent risk factors for HAPIs. The nomogram was well-calibrated and showed good discriminative ability (AUC=.87). The DCA showed a better net benefit, and the results were confirmed within the validation cohort. ConclusionThe nomogram model developed in this study showed good predictability and can identify patients at risk of developing HAPIs and aid the formulation of targeted interventions.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.