Multicentre prospective study to establish a risk prediction model on pressure injury in the neonatal intensive and intermediate care units

Abstract

Background The association between pressure injuries (PIs), risk factors, and preventive measures in hospitalised infants has not been extensively studied. Objective The aim of this study was to explore the incidence of PIs in hospitalised infants, the risk factors and preventive measures associated with them and construct a risk prediction model. Methods A multicentre, prospective, cohort study was conducted in infants hospitalised in two level III neonatal intensive care units and two neonatal intermediate care units in Italy. Sociodemographic and clinical information, the Italian-Neonatal Skin Risk Assessment Scale (i-NSRAS) scale score, presence of medical devices, risk factors, preventive measures, and PI characteristics were collected and evaluated using univariate and multivariate analyses. In addition, the decision tree technique was applied to analyse variables that best explained the occurrence of PIs in the neonatal population. Results A sample of 209 infants were included. A total of 62 PIs occurred in 40 infants; the cumulative incidence was 19.1%. In intensive care units, the cumulative incidence was 26.4%, whilst in intermediate care units, it was 8.3%. The PIs were categorised as stage I, 38.7%; stage II, 53.2%; and stage III, 8.1%. The most frequent location was the nose, 46.8%. Multivariate analysis identified the following risk factors: i-NSRAS score (relative risk [RR]: 0.81; 95% confidence interval [CI]: 0.69–0.94; p = 0.007); sedation (RR: 7.35; 95% CI: 1.67–32.40; p = 0.008), local pressure relief devices (RR: 4.41; 95% CI: 1.3 5–14.38; p = 0.014), and fasting (RR: 5.04; 95% CI: 1.37–18.47; p = 0.015). The decision tree detected that an i-NSRAS score ≤17 and the local pressure relief devices were the variables that best explain the appearance of PIs in infants. Conclusion The incidence of PIs in infants is high, especially in critically ill ones, and is mainly due to the presence of medical devices. The constructed clinical tree model can easily predict the risk of PI in hospitalised infants and, consequently, apply effective preventive strategies. To assign preventive measures based on the risk assessed according to objective criteria, we suggest the application of predictive models as part of a strategic PI prevention plan.