Research News

Risk of dying is higher in patients admitted to intensive care at weekends, finds Austrian study

BMJ 2017; 358 doi: (Published 07 September 2017) Cite this as: BMJ 2017;358:j4185
  1. Susan Mayor
  1. London

Patients admitted to intensive care units (ICUs) at weekends are at higher risk of death than those admitted on a weekday, even after adjusting for severity of illness, an Austrian registry study has found.1 But patients already on ICUs had lower mortality at weekends than weekdays.

“This effect requires thoughtful considerations both by clinicians and policy makers, because in the 21st century the day of admission should not influence patient outcomes,” said Paul Zajic from the Medical University Graz, Austria, who was lead author of the study.

The study analysed anonymised data for all 167 425 adult patients admitted to 119 ICUs in Austria between 2012 and 2015. Just under one in five (25 838, 17.1%) of these patients were admitted on a Saturday or Sunday.

Results, reported in Critical Care, showed that the risk of death in patients admitted to ICUs on a Saturday was 15% higher than for those admitted on a Wednesday (hazard ratio 1.15, 95% confidence interval 1.08 to 1.23); the risk was 11% higher for patients admitted to ICU on a Sunday (1.11, 1.03 to 1.18).

By contrast, the risk of dying in patients already in ICUs was significantly lower at the weekend than on weekdays. The risk of death on Saturdays was 7% lower than on Wednesdays (0.93, 0.87 to 1.00), and the risk on Sundays was 15% lower (0.85, 0.80 to 0.91).

“Contrary to what one might expect, weekend admissions do not alter patient outcomes immediately but are felt further down the line, potentially explaining why risk of death in the week was actually higher than the weekend,” said Zajic.

The higher risk of death in patients admitted to ICUs at weekends was not explained by the case mix. They were more likely to be admitted after emergency surgery, but this was not associated with increased mortality.

The complexity of providing critical care and the retrospective nature of the study made it difficult to identify factors associated with the weekend effect. But the authors found that patients admitted at weekends were less likely to receive several key treatments on the day of admission, including central venous catheters, peripheral arterial lines, and treatments to improve lung function, even though they received a higher rate of interventions overall.

Commenting on the study, Amy Mason from the University of Cambridge, UK, said, “This large, statistically robust study from Austria shows an apparent contradiction—that admission on the weekend has a higher death rate than admission on a weekday, while simply being in ICUs at the weekend has a lower death rate than during the week. This ‘contradiction’ is common in weekend effect studies and appears in other papers, including the well known Freemantle study on emergency admission mortality rates in UK hospitals.”

She added, “The most recent study of NHS ICU admissions did not find any increase in death rates due to admission at the weekend, and these new results do not generalise to the NHS. However, this new study fits with a body of evidence that the weekend effect is observed in many countries and that it is unclear whether lower quality of care or patients being more ill at the weekend is the dominant cause of each effect.”


View Abstract

Sign in

Log in through your institution