PHP100 - The Influencing of Hospitalization Costs Through Length of Stay Among Inpatients with Positive Antimicrobial Susceptibility Based on the Method of Structural Equation Model

Abstract

This multicenter and retrospective study aims to investigate the relationship between patient characteristics and total hospitalization costs (THC) while taking into account length of stay (LOS) as an independent or intermediate variable in the structural equation model (SEM) analysis. In addition, we compared the differing influences on THC of multi-drug resistance (MDR), non-MDR and susceptible cases. We collected information from inpatients with positive antimicrobial susceptibility in three hospitals in China from 2013 to 2015, which can be divided into three groups: MDR, non-MDR and susceptible. Possible variables influencing THC were investigated using SEM with LOS as an intermediate variable. The direct, indirect, and total effects of the independent variables on THC in this model were calculated. Those included in this study totaled 2587 inpatients, 65.4% of whom had detectable infections of MDR organisms. Compared to drug susceptible cases, there were more complex relationships between patient characteristics and THC among antimicrobial resistance (AMR) cases, especially MDR cases. LOS, as a major driver of hospitalization costs, had important effects on THC whether through a LOS path or when bypassing the LOS path. This study highlights the excessive burden imposed by AMR, especially MDR, on current Chinese healthcare systems, and supports better decision making on allocating limited healthcare resources.