Risk prediction and domiciliary care in acute exacerbations of chronic obstructive pulmonary disease

Abstract

Background Acute exacerbations of COPD (AECOPD) requiring admission have high mortality, readmission rates and healthcare costs. The DECAF score (Dyspnoea, Eosinopenia, chest radiograph Consolidation, Acidaemia and atrial Fibrillation) risk stratifies for acute mortality in these patients, but no validation or implementation study has been performed. Low risk patients may be suitable for Hospital at Home treatment. No risk stratification score to predict 90 day readmission/ death without readmission has been developed and validated in this patient group. Methods Consecutive patients with AECOPD were admitted to one of six hospitals. Predictors of mortality and readmission were collected. Data were combined with the DECAF derivation study to create a score to predict 90 day readmission/ death without readmission. Discrimination was assessed with the area under the receiver operator curve characteristic (AUROC). A non-inferiority, randomised controlled trial was performed to compare usual care to Hospital at Home (HAH) with patient selection by low risk DECAF score (0 or 1). The primary outcome was cost, with a cost-utility analysis as a secondary outcome. Results In 1,725 patients, the DECAF AUROC curve for inhospital mortality was 0.82 (95% CI 0.79 to 0.85), with a mortality risk of 1.0% in the DECAF 0 or 1 group and an overall mortality risk of 7.7%. In those that survived to discharge (n=2417), the strongest predictors of readmission/ death without readmission in the final model were: Previous admissions, eMRCD score, Age, Right-sided heart failure and Left sided heart failure (PEARL). The PEARL AUROC was 0.70 (95% CI 0.66 to 0.72). In 118 patients in the RCT, mean 90-day costs were £1,016 lower in HAH than usual care, but the one sided 95% cost interval crossed the non-inferiority limit (CI -2343 to iv 312). A sensitivity analysis assuming an extra days’ stay in usual care met the inferiority limit: cost difference £-1262, (CI -2590 to 66). HAH had a 90% chance of cost-effectiveness at a threshold of £30,000 per quality adjusted life year. Discussion The DECAF and PEARL score are simple tools that can be used at the bed side to risk stratify patients with AECOPD for inpatient mortality and readmission/ death without readmission respectively. Patient selection for Hospital at Home services using DECAF is cost effective.