A study of healthcare associated infections in the Intensive Care Unit of “Federico II” University Hospital through Logistic Regression

Abstract

The prevention of healthcare–associated infections (HAIs) is one of the most important parameters to evaluate healthcare service quality. In this work, we report on the application of the Firth's penalized maximum likelihood logistic regression to find some patients characteristics that can be related to HAIs and used as predictor factors. Data of 344 patients who have been hospitalized in the Adult Intensive Care of the “Federico II” University Hospital of Naples who underwent a wide range of surgical procedures between January 2018 and December 2019 were acquired using the departmental information system. This procedure allowed the identification of variables that influenced the risk of HAIs. Data distributions were evaluated to demonstrate their non-normality and then statistical analyses were performed such as Firth's penalized maximum likelihood logistic regression. Results show a correlation among the vascular catheterization days and the possibility to contract HAIs. This information, together with other tools for reducing the risk of infection such as surveillance, epidemiological guidelines, and training of healthcare personnel, could be of great help to re-design the healthcare processes and improve the quality of the health care system.