Email updates

Keep up to date with the latest news and content from Critical Care and BioMed Central.

Open Access Highly Accessed Open Badges Research

Effect of non-clinical inter-hospital critical care unit to unit transfer of critically ill patients: a propensity-matched cohort analysis

Helen Barratt1*, David A Harrison2, Kathryn M Rowan2 and Rosalind Raine1

Author Affiliations

1 Department of Applied Health Research, University College London, 1-19 Torrington Place, London WC1E 6BT, UK

2 Intensive Care National Audit & Research Centre, Napier House, 24 High Holborn, London WC1V 6AZ, UK

For all author emails, please log on.

Critical Care 2012, 16:R179  doi:10.1186/cc11662

See related letter by Droogh et al.,

Published: 3 October 2012



No matter how well resourced, individual hospitals cannot expect to meet all peaks in demand for adult general critical care. However, previous analyses suggest that patients transferred for non-clinical reasons have worse outcomes than those who are not transferred, but these studies were underpowered and hampered by residual case-mix differences. The aim of this study was to evaluate the effect of transferring adult general critical care patients to other hospitals for non-clinical reasons.


We carried out a propensity-matched cohort analysis comparing critical care patients who underwent a non-clinical critical care unit to unit transfer to another hospital with those who were not transferred. The primary outcome measure was mortality at ultimate discharge from acute hospital. Secondary outcomes were mortality at ultimate discharge from critical care, plus length of stay in both critical care and acute hospital.


A total of 308,323 patients were admitted to one of 198 adult general critical care units in England and Wales between January 2008 and September 2011. This included 759 patients who underwent a non-clinical transfer within 48 hours of admission to the unit and 1,518 propensity-matched patients who were not transferred. The relative risk of ultimate acute hospital mortality was 1.01 (95% confidence interval = 0.87 to 1.16) for the non-clinical transfer group, compared with patients who were not transferred but had a similar propensity for transfer. There was no statistically significant difference in ultimate critical care unit mortality. Transferred patients received on average three additional days of critical care (P < 0.001) but the difference in length of acute hospital stay was of only borderline significance (P = 0.05).


In our analysis the difference in mortality between non-clinical transferred and nontransferred patients was not statistically significant. Nevertheless, non-clinical transfers received, on average, an additional 3 days of critical care. This has potential ramifications in terms of distress, inconvenience and cost for patients, their families, and the National Health Service. We therefore need further evidence, including qualitative data from family members and cost-effective analyses, to better understand the broader effects of non-clinical transfer.