Predictors of Hospital Length of Stay in Patients with Exacerbations of COPD: A Cohort Study

Abstract

Background: Various studies have tried to delimit the predictors of hospital length of stay (LOS) for patients with exacerbated chronic obstructive pulmonary disease (eCOPD), but have been disadvantaged by certain limiting factors. Objective: Our goal was to prospectively identify predictors of LOS in these patients and to validate our results. Design: This was a prospective cohort study. Participants: Subjects were patients with eCOPD who visited 16 hospital emergency departments (EDs) and who were admitted to the hospital. Main Measures: Data were recorded on possible predictor variables at the ED visit, on admission and 24 hours later, during hospitalization, and on discharge. LOS and prolonged LOS (≥ 9 days, considering the 75th percentile of LOS in our sample) were the outcomes of interest. Multivariate multilevel linear and logistic regression models were employed. Results: A total of 1,453 patients were equally divided between derivation and validation samples. The hospital variable was the best predictor of LOS. Multivariate predictors of LOS, as log-transformed variables, were the hospital, baseline dyspnea and physical activity levels and fatigue at 24 hours, intensive care or intensive respiratory care unit admission, the need for antibiotics, and complications during hospitalization. Predictors of prolonged LOS were also the hospital, baseline dyspnea and fatigue at 24 hours, ICU or IRCU admission, and complications during hospitalization (AUC: 0.77). Models were validated in the validation sample (AUC: 0.75). Conclusions: We identified a number of modifiable factors, including baseline dyspnea, physical activity level, and hospital variability, that influenced the LOS of patients with eCOPD who were admitted to the hospital.