Academic calls for MPs to investigate “weekend effect” in view of conflicting evidence

The House of Commons health select committee is expected to consider holding an investigation into whether patients admitted to hospital at weekends face a greater risk of death than those admitted during the week, after an academic called for a review earlier this week.

The government has argued that evidence of such a “weekend effect” exists and has used this as the basis of its argument for a “seven day” NHS, one of the key issues in its protracted dispute with junior doctors in England over their new contract. In particular, the health secretary for England, Jeremy Hunt, has cited a paper published in The BMJ to push for changes. The paper, by Nick Freemantle of University College London and colleagues, found that 11 000 more people died each year within 30 days of admission to hospital if they were admitted on a Friday, Saturday, Sunday, or Monday than if they were admitted on other days of the week.1 However, this analysis did not apportion any cause for that excess mortality, and there is a growing body of evidence disputing that a weekend effect exists.

Last week a paper published in the Journal of Health Services Research and Policy concluded that the death rate after admission to hospital at the weekend was a statistical artefact and was higher only because the number of patients admitted at the weekend was lower and these patients tended to be sicker than those admitted during the week.2 It said that previous studies did not include patients attending emergency departments who were not admitted.

On Monday 9 May Peter Rothwell, professor of neurology at Oxford University’s Nuffield Department of Clinical Neurosciences decided to talk to the media3 about the findings of a research paper that is due to be published in The BMJ and urged the House of Commons Health Committee to investigate whether a weekend effect actually existed. Rothwell’s research indicates that inaccurate administrative coding can lead to some patients admitted during the week for routine care being miscoded as emergency cases and that this would falsely inflate weekend death rates.

Rothwell, who said that he did not support the junior doctors’ strikes, said that he had decided to make his research findings public before journal publication because the Department of Health and the BMA had agreed to a week of last ditch talks to try to resolve the junior doctor contract dispute. He told The BMJ that he thought that his findings might be helpful and that, if he waited, “events would have taken their course.” He likened the standoff over the seven day NHS to a “Shakespearian tragedy” and urged the Health Committee to investigate the data properly.

Politicians were doing the right thing in their minds but had been badly advised, Rothwell said. “They are genuinely being told that there is a big problem. And at the same time junior doctors realise that it is just not like that.”

The Health Committee has indicated that it is concerned about how the mortality data were being interpreted. On 9 May, while the committee was taking evidence from Hunt on the effects of the government’s comprehensive spending review on health and social care, the committee member Philippa Whitford, the Scottish National Party member for Central Ayrshire, pointed out that there was now an almost 2:1 ratio in numbers of papers disputing a weekend effect versus those supporting one.

This was the day before the Lancet published two further studies contesting a weekend effect. The first involved 74 307 patients admitted over a year to 199 hospitals in England and Wales for acute stroke.4 It found no difference in 30 day survival between patients admitted during the day at the weekend and those admitted during weekdays. The second found no association between risk of death and senior doctor staffing levels.5

Whitford asked Hunt, “Do you not think that the first thing that is required is to actually gather proper evidence as to whether it [weekend effect] exists and what the cause of it might be? Is it not beholden on the health secretary to actually know what the problem is before you spend billions fixing it?”

Hunt claimed that hospital staff applying a more stringent admission threshold at weekends was evidence of a weekend effect because “we don’t offer the same standard of care at weekends as we do in the week.” He added, “I’m not an academic, but I think the mistake for a health secretary is to look at the overwhelming amount of evidence there is of a weekend effect and decide to get off the hook by disputing the methodology.”

Whitford asked whether, if the NHS offered the same services every day of the week, “Is there not the danger that in actual fact we will admit more people, [that] the ratio will look better but exactly the same number of people will have died?”