Background Excess mortality in patients admitted to hospital at weekends has been reported in many healthcare systems, influencing healthcare policy. The limitations of existing data are well described; there is a need for condition-specific research in well-described populations with adjustment for baseline mortality risk.¹ Acute exacerbations of COPD (AECOPD) are one of the commonest reasons for hospital admission, with high rates of inpatient mortality. We aimed to establish if inpatient mortality is associated with day of admission or death amongst patients admitted with an AECOPD.

Methods Consecutive admissions from six UK hospitals were identified from the DECAF derivation and validation studies.² All patients (n = 2,645) had definite COPD (including spirometric confirmation) and the primary reason for admission was AECOPD. DECAF indices (dypsnoea, eosinopenia, consolidation, acidaemia and atrial fibrillation) and age were collected.

We captured the number of inpatient deaths per day of admission (compared to the total number of admissions on each day) and per day of death (compared to the total number of bed days for each day). Proportions were compared using Fisher’s exact test. The association between period of admission (weekday/weekend) and mortality was assessed in a binary logistic regression model, including the DECAF indices and age.

Results Inpatient mortality was 9.3% (63/676) for those admitted on weekends, compared to 8.4% (165/1969) on weekdays (p = 0.47). For day of death, no clear difference in mortality was seen between weekdays and weekends although fewer deaths were seen on Fridays. Exacerbation severity was similar between weekday and weekend admissions (median DECAF score 2 vs. 2, p = 0.83). Following adjustment for baseline mortality risk, there was no association between weekend admission and inpatient death; OR 1.11 (0.79 to 1.56), p = 0.55.

Discussion In a well-described population with an AECOPD, there is no relationship between inpatient mortality and day of admission or day of death, even after adjusting for baseline mortality risk.

References

1. Becker DJ. Expert Rev Pharmacoecon Outcomes Res 2008.

2. Echevarria C, et al. Thorax 2016.