The “weekend effect”BMJ 2016; 353 doi: http://dx.doi.org/10.1136/bmj.i2801 (Published 19 May 2016) Cite this as: BMJ 2016;353:i2801
“Almost nothing is clear in this tangled tale,” says Martin McKee in his editorial on the so called weekend effect (doi:10.1136/bmj.i2750). So this week we try to help make sense of what we know and what we don’t know about the apparent association between weekend admission to hospital and risk of death (doi:10.1136/bmj.i2781).
Recent weeks have seen a flurry of new analyses attempting to enlighten this increasingly heated and politicised debate. Two papers in particular concluded that data artefacts or case mix (or both) were to blame for the apparent effect. The first, published in the Journal of Health Services Research and Policy (doi:10.1177/1355819616649630), found a higher threshold for admitting patients at weekends, which could explain why those who get into hospital at weekends are sicker and more likely to die (doi:10.1136/bmj.i2598). The second, published in The BMJ this week (doi:10.1136/bmj.i2648), finds that some patients recovering from a previous stroke were coded as having had a new stroke. These patients were more likely to be admitted on weekdays and less likely to die. An analysis limited to patients verified to have had new strokes found no weekend effect.
As an aside, both papers have instructive prepublication stories. The first had been rejected by The BMJ, a fact I can share because the authors told the media, along with the identity of a peer reviewer. This led to criticism of The BMJ and, more troublingly, of the reviewer. In answer we have published, with permission from all parties, the four peer reviewers’ comments and an explanation of our decision to reject (blogs.bmj.com/bmj, 16 May). The second paper was under fast track consideration by this journal when the authors released the headline results to the media last week (doi:10.1136/bmj.i2667). We were disappointed by this behaviour, as it left the public unable to assess the validity of the authors’ claims (bmj.com/about-bmj/resources-authors/media-releases). But given that we had provisionally accepted the paper, and the data had not been released, we agreed to continue with publication.
From these and other studies McKee concludes that at least part of the weekend effect is data artefact and that any remaining association does not seem to be due to medical staffing. To the extent that a weekend effect exists, he says, the evidence based response would be to provide more primary care and more nurses. Some increased funding has now come through for primary care (doi:10.1136/bmj.i2357), but as for nurse staffing all we have is empty rhetoric, says David Oliver (doi:10.1136/bmj.i2665).
What’s needed now is more research into seven day NHS services and less political interference. When the health secretary unaccountably alighted on the junior doctors as his first target he picked the wrong fight for the wrong reasons.
thebmj.com Articles and blogs on the “weekend effect” from The BMJ and BMJ Careers are collated at bmj.com/weekend.