Predictors of hospital admission two months after emergency department evaluation of COPD exacerbation

Abstract

Background: Limited information is available regarding the factors related to short-term hospital admission following an exacerbation of chronic obstructive pulmonary disease (eCOPD). Objectives: The aim of this study was to identify variables related to short-term admission in patients with an eCOPD. Methods: This was a prospective cohort study of patients with an eCOPD who attended an emergency department (ED) at 1 of 16 hospitals. Information on possible predictor variables was recorded during the ED stay, 24 h after admission to the hospital or after ED discharge home, and at hospital discharge or 1 week later if discharged home from the ED. An admission after an eCOPD within 2 months was the outcome of interest. Multivariate models were employed for patients admitted to the hospital or discharged home from the ED. Results: For patients discharged home from the ED, eCOPD-related hospital admissions in the previous year [odds ratio (OR) 1.98 and 2.33], pCO2 at ED admission (ORs2.02 and 2.90), the number of ED visits within 1 week of the index ED visit (OR 5.14) and dyspnea level 1 week after the index ED visit (ORs 2.66 and 1.40) were predictors of shortterm admission [area under the curve (AUC) 0.82]. For patients admitted to the hospital during the index ED visit, baseline FEV 1 % (ORs 1.32 and 1.88), eCOPD-related hospital admissions in the previous year (ORs 1.28 and 2.51), severe baseline dyspnea (OR 2.57) and dyspnea level 1 week after the index ED visit (ORs 2.15 and 1.74) were predictors of short-term readmission (AUC 0.73). Conclusions: Just a few easily recorded parameters may allow clinicians to identify patients at a higher risk of short-term readmission and establish preventive strategies.