BMJ rejected study disproving weekend effect despite positive peer review
BMJ editors rejected the publication of a study that found the ‘weekend effect’ was a ‘statistical artefact’, despite being advised to publish by peer reviewers, it has emerged.
The study by researchers from the University of Manchester was eventually published in the Journal of Health Services Research and Policy, after being rejected by the BMJ after appeal.
The BMJ published its reasoning behind the rejection today in a blog, saying that there were ‘vagaries and limitations in the data’ and that the journal ‘must ensure that we bring light to this topic, rather than generate heat’.
But it also admitted that three of the four peer reviewers had said ‘they had hoped we would publish the paper and were surprised that we didn’t’.
The paper was particularly contentious because the so-called ‘weekend effect’ has been used by health secretary Jeremy Hunt to justify his move towards a seven-day service and the imposition of a new contract on junior doctors.
He has cited a paper published in the BMJ in October 2015 that concluded ‘patients admitted on Saturday or Sunday face an increased likelihood of death even when severity of illness is accounted for’.
Mr Hunt said that the paper had shown that 6,000 lives could be saved by better staffing at the weekend – although both the study authors and the BMJ editor said this was a misreading of the study.
Pulse revealed last week that one of the reviewers of the Manchester study was Professor Nick Freemantle, professor of clinical epidemiology and biostatistics at University College London, who led the October study.
However, the BMJ said that he - alongside two other reviewers - was in favour of publication, despite raising a number of issues in the writing and the data in the study.
The BMJ associate editor, Dr Rubin Minhas, who is also a GP in Kent, wrote to Rachel Meacock, the study lead, saying: that although it was ’a fascinating subject, and obviously topical… we must ensure that we bring light to this topic, rather than generate heat, although the findings slant away from previously published recent work in this area.
‘I am sorry to say, that after conferring with one of my fellow editors, we decided that we would let this go at this stage. The reviews would almost certainly not manage to persuade the full editorial committee that it should be approved for publication.’
In a blog, the BMJ editors wrote: ‘The decision to reject the Meacock paper was made by The BMJ editors and not by the four external peer reviewers who saw the original paper and a subsequent revision.
‘With permission from the authors and reviewers we have posted their signed comments, which can be viewed here along with editorial comments.
‘In agreeing to our request three of the reviewers, including Professor Freemantle, told us, unprompted, that they had hoped we would publish the paper and were surprised that we didn’t. We suspect that many people who read the reviewer comments may also think we made a mistake.’
The Meacock study looked at data involving 12,670,788 A&E attendances and 4,656,586 emergency admissions in greater detail than ever before.
It found that, although mortality rates of patients admitted at the weekend were higher than those admitted during the week, overall numbers of deaths were ‘significantly’ lower - 378 deaths within 30 days of patients who attended A&E on an average weekend day compared with 388 on an average week day.
They said this was because ‘hospitals apply a higher severity threshold’ at the weekend.
The row over the ‘weekend effect’
Health secretary Jeremy Hunt has imposed a contract on junior doctors removing extra pay for working Saturdays in a bid to increase the workforce at weekends in a cost-neutral way.
The Government justified the push for seven-day services by using a study in the BMJ led by Professor Nick Freemantle, chair of clinical epidemiology and biostatistics at University College London.
Mr Hunt claimed that the study revealed there were 6,000 preventable deaths due to fewer staff working at weekends – although the BMJ editor Fiona Godlee said he was ‘misrepresenting’ the data.
A University of Manchester study, published in the Journal of Health Services Research and Policy, that found that in fact fewer patients die after being admitted to hospital at the weekend.
Meanwhile, a new University of Oxford study of 92,728 stroke patients registered with nine practices in Oxfordshire - which is as yet unpublished - found that data for weekend admissions could be undermined by ‘inaccurate’ coding.
Two more studies published in the Lancet state that the weekend effect is a ‘major oversimplification’ of a complex pattern of weekly changes in quality of care which are unlikely to be addressed by just increasing the availability of hospital doctors on Saturdays and Sundays.