Investigation of the risk of surgical infections at the “Federico II” University Hospital by regression analysis using the Firth method

Abstract

Surgical infections (SSIs) are among the most common type of healthcare associated infections (HAIs) and a major cause of morbidity among surgical patients, increase of hospitalization days and of healthcare expenditure In this work, we present a logistic regression model to study the impact that different clinical, demographic and organizational factors have on the risk of occurrence of HAIs in a surgery department. The proposed model regression model is based on the Firth's penalized maximum likelihood logistic regression, a well-suited methodology for the analysis of unbalanced datasets, such as those related to events with a low occurrence rate, which is often the case of hospital infections. The model proved to be able to identify the factors most influencing the risk of SSIs and offers a promising tool for the systematic study of SSIs.