NHS patients are more likely to die – it is right to ask why

Last week a think-tank report drew newspaper headlines with the claim that the NHS ranked second from bottom across "a series of major health outcomes" compared with other international systems. Does the NHS really perform so badly internationally, and is there a real problem with our health outcomes? Mark Dayan looks at the evidence in the round.

Blog post

Published: 06/05/2022

Four years ago, the BBC asked me and my colleagues across the Health Foundation, The King’s Fund and the Institute for Fiscal Studies to look into an unusually blunt question: how good is the NHS? The report aimed to answer by comparing every available source of data about it with 18 comparable countries across Europe, the G7 and the Anglosphere.  

Last week the think-tank Civitas used parts of our methodology to produce an alarming indictment of NHS care, placing the UK “second to bottom across a series of major health care outcomes”. A spate of news and comment articles reflected this, warning that the NHS was “the sick man of the world”.  

Is the problem real?  

Civitas’s report focuses at times on the most unflattering elements of the NHS’s record. They state that it achieves poor outcomes despite middling spending – but do not look at numbers in key staff groups where it is threadbare. The joint think-tank report found evidence that it has a relatively good record in preventing suicide among mental health patients, but this is not covered. The headline claim that “for treatable diseases the UK is second to bottom” is only true if the latest data for Greece and Portugal is ignored on the dubious grounds that these countries didn’t submit information every year in the past.  

The statements of alarm about specific findings may also put too much weight on the reliability of the source data. The OECD, who provide most of what Civitas use, do excellent work allowing us to compare across countries. But data collected differently across different systems always comes with a cloud of uncertainty.  

That doesn’t mean, though, that the overall conclusion is wrong. Data from every available source tend to show that patients treated by the NHS are more likely to die than they would be if they received treatment for heart disease, strokes, or cancer in most other developed countries. In 2014 survival rates in the UK were worse than average for all the types of cancer which cause the most death, and the worst of any comparable country for colon and pancreatic cancer.  

In 2016, we found that the UK lagged behind on both stroke and heart disease survival. Civitas are right to point out that, if anything, relative performance has become even worse. Although UK survival rates have improved, they have done so more slowly than others, and have now slipped behind Portugal and Finland as well as every other comparable country for which data exist.  

Why might the NHS have poor outcomes?  

The 2018 project also looked into the resources, the efficiency and the priorities of the NHS to understand why this might be the case. This is important, because understanding the apparent problem requires some sense of how other aspects of the NHS might differ from its counterparts. 

Again, the data are far from perfect. There is little sign of any particular inefficiency in terms of administrative spending. The tax-funded, universal structure of the NHS is also quite common, with countries such as Italy and Canada having health systems modelled on it. The UK does have relatively poor population health, with life expectancy lower and preventable deaths higher than France or Italy, though not Germany. This will not be primarily driven by health care, but by factors such as inequalities, diet and lifestyle. 

But there is a second unusual characteristic specific to the NHS that appears in several different sources of data: its lack of staff and equipment. The UK spends a low-to-medium amount on health care relative to its peers, generally less than its Northern European neighbours but more than Southern Europe. Yet unusually, it appears to have both significantly fewer doctors than average and significantly fewer nurses than average. As well as the OECD, Eurostat showed this too while we were EU members: by their count the UK had fewer doctors per 100,000 inhabitants than anywhere in Western or Central Europe.  

Meanwhile, the UK also has an astonishingly low number of MRI machines and CT scanners: fewer per person, according to the OECD, than any other developed country. This reflects a long history of low capital spending in the UK’s health service. Successive governments and the Treasury have cut back on long-term investment and planning, pushing short-term money out of the door in the hopes of winning the next election.   

Similar dynamics have arguably contributed to our poor history of workforce planning. England currently does not have an NHS workforce plan at all: the Treasury recently blocked a parliamentary amendment to require even a forecast of how many staff would be needed.  

Do we get what we ask for?  

Another factor we reflected on that is almost impossible to quantify was whether the UK has simply not prioritised clinical outcomes very highly.  

I feel that this hypothesis is still relevant. Given the UK’s poor performance in saving lives, it is remarkable how rarely this issue is raised and emphasised by government or even the opposition. The Secretary of State’s extensive recent speech on NHS reform barely mentioned the problem, and euphemistically described UK performance as “middling”.   

Civitas ascribe this to “uncritical worship of the NHS”. In fact, though, concerns about waiting times and access to GPs generate wall-to-wall negative coverage about the health service, and are widely held amongst the public, for whom satisfaction fell sharply in our latest British Social Attitudes survey with The King’s Fund.   

Yet international surveys by the Commonwealth Fund tend to suggest that the UK manages to achieve fairly respectable timely access to care. For example in 2021, 14% of older adults in the UK said they waited more than six days to see a doctor. This figure was lower than most countries, with over 20% of older adults in Norway, France, the United States, Sweden and Canada saying they waited this long.   

It is possible that clinical outcomes, which require specialised knowledge to interpret as good or bad, are harder for a voter or a minister to see than access or patient experience. A highly politicised health service run by national politicians may tend to overlook them relative to the tangible experiences of waiting for care and the easy fix of structural reorganisations, which convulse the NHS every few years. Even much of the British media commentary on the Civitas study has tended to put forward  structural reorganisation or changes to access to care – such as co-payments – as solutions.  

Uncertainty in data will remain a serious concern, and looking in more depth at comparing the UK to other countries for cancer, heart attacks and strokes should be a major priority. It is also very difficult to say how the UK might be performing in the crucial areas of health care which improve lives without necessarily saving them: dental treatment or hip replacements, for example. But Civitas are certainly right to point to cause for concern. We should hope that their report helps move us beyond the usual tired obsessions with privatisation and structures, onto the question of whether we could have a health service that saves more lives. 

Suggested citation

Dayan M (2022) "NHS patients are more likely to die – it is right to ask why". Nuffield Trust comment, 6 May.