Risk Factors Affecting Death from Hospital-Acquired Infections in Trauma Patients: Association Rule Mining

Introduction: Trauma patients are potentially at high risk of acquiring infections in hospitals, which is the main cause of in-hospital mortality. The aim of this study was to identify the risk factors contributing to death from hospital-acquired infections in trauma patients by data mining techniques. Methods: This is a cohort study. A total of 549 trauma patients with nosocomial infection who were admitted to Shiraz trauma hospital between 2017 and 2018 were studied. Sex, age, mechanism of injury, body region injured, injury severity score, length of stay, type of intervention, infection day after admission, microorganism cause of infections, and the outcomes were collected. Association rule mining techniques were applied to extract knowledge from the data set. The IBM SPSS Modeler data mining software version 18.0 was used as a tool for data mining of the trauma patients with hospital queried infections database. Results: The age older than 65, surgical site infection skin, bloodstream infection, mechanism injury of car accident, invasive intervention of tracheal intubation, injury severity score higher than 16, and multiple injuries with higher than 71 percent confidence level were associated with in-hospital mortality. The relationship between those predicators and death among hospital-acquired infection was strong (Lift value 1). Conclusion: Factors such as increasing age, tracheal intubation, mechanical ventilator, surgical site infection skin, upper respiratory infection are associated with death from hospital-acquired infections in trauma patients by data mining. Keywords: Mortality, Hospital-acquired infections, Trauma, Association rule mining, Data mining.