The public are the most dissatisfied with the NHS in Labour-run Wales.
There’s no statistically significant difference between the figures for Wales, England and Scotland. So although the figure from the Welsh sample was lowest, it’s possible that a survey of Wales as a whole would show it was more satisfied than Scotland or England.
There is a 60% dissatisfaction rate in patients in Wales.
This is the estimated rate for the Welsh public, not for patients specifically.
The last Labour government achieved the highest patient satisfaction in the history of the NHS.
This appears to be a reference to satisfaction among the public, not patients specifically. If so, it’s essentially true going back to 1983, but we don’t know about the history of the NHS before that. Technically, this level was also recorded just after Labour left office in 2010.
The public are the most dissatisfied [with the NHS] in Labour-run Wales.
The last Labour government delivered the highest patient satisfaction in NHS history. When Labour left office in 2010, it was at 70%.
The last Labour government achieved the highest patient satisfaction in the history of the NHS.
Following the release of new data on public satisfaction with the NHS in Great Britain, both the health secretary, Victoria Atkins, and the Labour party made comments about it that aren’t quite as simple as they seem.
Neither claim is completely wrong, but both need more context to understand the uncertainty around them.
Politicians should take care to describe data about public services correctly, so people can base their vote on an accurate picture of the world.
Honesty in public debate matters
You can help us take action – and get our regular free email
In a series of posts on X (formerly Twitter), Ms Atkins said: “This is a UK wide survey, including the NHS in Labour-run Wales and SNP-run Scotland. It shows that the public are the most dissatisfied in Labour-run Wales.” (Health policy is devolved so that the UK government controls the NHS in England, whereas the devolved governments control it in the other nations.)
The survey in question, called the British Social Attitudes survey (BSA), involved 3,374 people, and was carried out by the National Centre for Social Research across Great Britain—so not quite the whole of the UK, since it didn’t include Northern Ireland.
To support her point, Ms Atkins shared a chart taken from a report based on the survey data by the King’s Fund and the Nuffield Trust, two health think tanks. The chart shows the percentage of respondents saying they were either quite or very satisfied with the NHS, and it does show a lower central value for Wales than for Scotland or England.
That isn’t quite the end of the story, however, because the chart also shows what’s called a “95% confidence interval” around each point, which essentially shows a range of reasonable uncertainty around it.
Confidence intervals are important, because surveys are never perfect. When you pick a sample of people from a wider population, it’s possible that the people you choose aren’t quite typical of the population as a whole.
That’s a bigger problem when the sample is small, because it’s easier for smaller groups to be more unusual. In this case, the data on Wales came from just 174 people, meaning it should be “viewed with caution” in the words of a note at the bottom of the chart.
You can see how much more uncertain the Welsh figure is, because the confidence interval around it is very wide compared with most of the others, suggesting a wide range of reasonable uncertainty. Indeed this uncertainty means we can’t rule out many possible values that would be higher than England’s.
As a result, this data doesn’t rule out the possibility that satisfaction with the NHS in Wales, or Scotland, is higher than it is in England. This means the difference between them is not considered “statistically significant”—a point that both the Nuffield Trust and the King’s Fund emphasised in later posts on social media. Page 10 of the report also says: “There were no statistically significant differences in satisfaction between groups.”
So is satisfaction with the NHS lowest in Wales? It’s possible, but the truth is we can’t be sure, making Ms Atkins’s comment more like an informed guess than a statement of fact.
As the report puts it: “If a change or difference is statistically significant, this means we can be 95% confident that the survey result reflects a real change or difference in public views, rather than being down to chance. Where a change or difference is not statistically significant, we cannot be confident that it reflects a real change or difference in public views.”
A separate post from the Conservatives on X and Facebook on 27 March seemed to cite the same research by claiming there was a “60% dissatisfaction rate in patients” in Wales.
This figure shows the other half of the picture, so people who said they were very or quite dissatisfied with the NHS. But again it has a wide confidence interval, which means the true rate in Wales could be quite different.
More importantly, the Conservative post is wrong to say that it represents dissatisfaction among patients. The survey responses came from samples of the general public, so we don’t know how many of the people answering the question had been NHS patients that year.
The report itself says: “The British Social Attitudes survey is a survey of public satisfaction rather than patient experience of health and care services. Many factors are likely to influence reported satisfaction, including national debate in the media around the NHS and social care.”
Wes Streeting and Labour have also made this mistake on several occasions, as we will see.
A Labour Party post on X, shared by many Labour MPs, said that “the last Labour government delivered the highest patient satisfaction in NHS history”. This echoed earlier comments from the shadow health secretary Wes Streeting and the shadow health minister Andrew Gwynne.
In each case, the precise wording is more important than it might appear at first. As we said in a fact check about similar comments by Mr Streeting last year, it seems likely that these claims are based on the same BSA data, which allows comparisons back to 1983. No other source provides a consistent measure with which to compare different periods of government.
If so—and Labour has not confirmed its source with us—then as we’ve already seen, it is a measure of public satisfaction, not patient satisfaction.
With that in mind, it’s true that the highest level of satisfaction in the series comes in 2010—although we haven’t checked whether the difference between this level and others was statistically significant.
Another caveat we’ve mentioned before is that the 2010 data was actually collected just after Labour lost the 2010 general election, meaning that it technically reflects public satisfaction under the Conservative-led coalition government. It is unlikely that the new government had significantly affected the quality of NHS services by this point, however.
The main issue with Mr Streeting’s and Mr Gwynne’s comments, however, is that this data only tells us how satisfied people were each year back to 1983, not “in the history of the NHS”, which was established in 1948.
Image courtesy of Richard Townshend and David Woolfall
Full Fact fights for good, reliable information in the media, online, and in politics.
Bad information ruins lives. It promotes hate, damages people’s health, and hurts democracy. You deserve better.
