"The lack of surgeons and consultants at the weekends is thought to cost around 3,000 lives every year."—Sky News, 18 May 2015
"4,400 extra deaths… are attributed to sub-standard care at weekends."—Sunday Times, 26 July 2015
"Around 6,000 people lose their lives every year because we do not have a proper 7-day service in hospitals."—Jeremy Hunt, Health Secretary, 16 July 2015
Does a lack of staff or poor care on NHS wards at the weekend cost thousands of lives per year? It's impossible to say.
Researchers can estimate the difference between the number of weekend patients who die and the number who would have died had they had the same chance of mortality as weekday patients. That gives an idea of the number of "excess deaths" related to going into hospital at the weekend—but not why they happen.
Some of the increased mortality might be due down to the type of patient who is admitted at the weekend, and while the various research projects have tried to account for that, critics have argued it's not possible to control for everything.
There's evidence behind a variety of excess death figures—estimates vary between 3,000, 4,400, close to 6,000, and 11,000. The variety is explainable: the research in question looks at deaths in different years, different types of hospital admissions, and uses different methods. The 11,000 estimate includes deaths after admission on a Friday and Monday, on the basis that some services wind down on Friday evening and have to deal with a weekend backlog on the following Monday.
There are thousands of 'excess deaths' for weekend patients
In 2010, researchers from Imperial College London found that in 2005/06 there were 3,400 'excess deaths' associated with emergency admission to NHS hospitals in England at the weekend ("weekend patients"), rather than on a weekday. These were emergency admissions to hospitals rather than planned procedures, and the deaths counted happened in hospital. The figure takes into account factors like age, deprivation, and the type of condition diagnosed.
Weekend patients were 10% more likely to die than weekday patients—mortality was 4.9% for weekday emergency patients and 5.2% for weekend patients.
The researchers said the differences "may reflect differences in quality of care".
In 2012 another group of researchers looked at data from 2009/10 and found a 16% greater risk of death for weekend patients for both emergency and non-emergency admissions to hospital. In this case researchers looked at all deaths that happened within 30 days of admission.
This research was later used by the Department of Health to calculate a figure of 5,700 excess deaths for weekend patients in that year. The department said that this supports Jeremy Hunt's reference to "around 6,000 people".
It certainly supports his figure for excess deaths, but not necessarily the diagnosis of what causes them. The researchers concluded only that seven-day access to services "may" improve outcomes for some patients.
In May 2015 researchers from the University of Manchester found there were either 4,400 and 5,400 excess deaths in 2010/11, depending on the methods used to account for the characteristics of patients. This too looked at deaths that occurred within 30 days of admission, for emergency admissions only. The researchers said there was as yet "no clear evidence" that seven-day working would reduce the weekend death rate.
An estimate of 4,400 has also been put forward by NHS England—this was the source used by the Sunday Times.
Most recently, an update to the 2012 research which used data from 2013/14 found 11,000 excess deaths associated with hospital admission on Friday, Saturday, Sunday or Monday, compared with the other three days of the week, saying:
"It is not possible to ascertain the extent to which these excess deaths may be preventable; to assume that they are avoidable would be rash and misleading".
Excess deaths don't necessarily represent what a 7-day NHS could prevent
At the moment any claim about how many lives would be saved by a full "7-day NHS" service are not backed up by research. It would be fairer to say that the evidence suggests that up to a certain number of deaths a year are associated with weekend admissions, and that some or all of this might be caused by poor care.
And it's an extra step to connect any figure to the number of lives that could be saved by the "7-day service" being suggested recently—one that isn't supported by this evidence.
Update 18 August 2015
The article originally said we didn't know where the Health Secretary's figure of 6,000 excess deaths came from. That was true at the time. We've updated to include the source after it was published by the Department of Health.
Update 11 September 2015
We have updated the article again to include details of an estimate of 11,000 excess deaths, published recently.