Sadly, it is not news that poorer areas of England have worse health outcomes. Research from Imperial College London suggests women in poor areas typically die eight years earlier than their wealthier counterparts, and men 10 years earlier. These outcomes are driven by a huge range of factors – from bad housing, to smoking rates, to the simple fact that people whose illnesses stop them working get pushed by house prices into poorer areas.
One important factor in balancing all this out should be access to a universal health care system committed to providing equitable care to all: the NHS. But we were asked by the Financial Times to look at the interesting question of whether the health service too is leaving certain areas behind.
As a measure of poverty I used the Office for National Statistics’ 2015 figures for multiple deprivation in 33,000 communities, and then aggregated these up to get the average deprivation of a person in each of the 207 clinical commissioning groups that oversaw the NHS in England last year. Each of these effectively represents an area of the country with an average of just below 300,000 inhabitants. To measure the standards of NHS care, I looked at waiting times, staffing, activity, and some outcomes of health service interventions like mental health recovery rates.
According to need
In many respects, the results show that the NHS does quite well in delivering equal services to rich and poor. There is no obvious significant link between the deprivation of a CCG’s residents and the proportion left waiting more than 18 weeks. The same is true for the four-hour A&E target, counted in March 2017.
Deprived areas actually have very slightly shorter median waiting times for planned care, with the average person waiting 48 days rather than 50. However, this is a fairly tiny gap and the relationship hovers on the edge of statistical significance. The turnover of staff – a measure of how stretched the workforce in trusts is – appears if anything slightly lower in more deprived areas, giving a higher ‘stability index’.
There is a faint link between deprivation and the proportion of people waiting more than two months for cancer treatment. People in more deprived areas are slightly less likely to recover after a course of psychotherapy, although that may partly reflect factors beyond the doors of the health service.
The chart below shows results for a selection of these indicators for the most deprived fifth of CCGs, relative to the least deprived fifth.
These measures largely reflect ‘secondary care’ – hospitals and other services that you might access on referral from a GP. What could explain how the NHS manages to provide such an equitable service here? One relevant fact may be that money for these services is distributed through a sophisticated formula, which takes account of trends in many different diseases at a local level. That might mean more deprived areas tend to get enough to keep up with the higher needs they face.
Where health follows wealth
A somewhat different picture emerges when you look at general practice. I used the many questions asked in the GP Patient Survey, filled out by more than 800,000 people between January and March 2017, to get a sense of whether people in poorer areas had longer waits or a worse experience.
Across many measures, a gap emerges between the most and least deprived areas. The difference is typically not all that large – for instance, people in deprived areas on average saw the GP they wanted to see 52% of the time, compared to 59% in wealthy areas. However, it is consistent, running across many measures of both how easy it is to get an appointment, and how satisfied people are when they do get one.
Four examples are shown below. For each of these, statistical analysis of all CCGs suggests there would be less than one chance in a hundred of the apparent relationship happening by coincidence.
None of this means that poor people are necessarily getting a poor service. Especially when it comes to how GPs behaved towards them, people in deprived areas still report a good experience. But their wealthier counterparts report a slightly better one. Overall, an average of 83% of people in the most deprived fifth of CCGs have a good experience of visiting their GP, but this is 87% in the least deprived fifth.
These findings echo long-standing worries about poorer areas losing out in their access to general practice in particular. Our report on continuity in general practice earlier this year found a similar link for much smaller postcode-sized areas between poverty and not being able to see the doctor you prefer. And it was in the 1970s that the Welsh GP Julian Tudor-Hart first famously proposed an “inverse care law”. As originally stated, this says that “the availability of good medical care tends to vary inversely with the need for it in the population served”.
Strikingly, another measure I looked at suggests this remains literally true with respect to deprivation in English general practice. Even though you would expect them to have higher needs, practices that reported data in more deprived areas have fewer GPs per person on their list than those in less deprived areas. In September 2017, on average there were 1,869 patients on GP lists for each doctor in the least deprived fifth, compared to 2,125 in the most deprived. This relationship is still present even when the greater number of people over 65 in wealthier areas is added to the equation. The same pattern does not exist for nurses in general practice.
This leaves some local areas with a striking mismatch between apparent need and staffing. Thanet in Kent is not only well within the most deprived fifth of CCGs, but has a proportion of people over 65 far above the national average, at 23% compared to 18%. Yet at 2,270 patients for each doctor, GPs are in even shorter supply than the average in poorer areas.
Research using a tailor-made survey in Scotland recently found very similar issues. The authors, from Glasgow University, also concluded that appointments in the poorest areas were shorter than those in wealthy areas, while GPs answering the service reported higher levels of stress.
What could explain why GPs do not seem to be distributed in a way that allows them to provide an equally good service to rich and poor? Money for general practice is given out using a simpler allocation formula. While it does actually give some money out based on Index of Multiple Deprivation deciles, age plays a larger role. There is no specific model of how particular social factors and particular diagnoses actually determine the need for care, as exists for secondary care. Experts have warned that the formula is not capturing some major drivers of workload.
It is also possible that doctors cannot be recruited as easily in more deprived areas, meaning actual capacity fails even to keep up with the money available. The prolonged and large staffing shortage in general practice means there are not enough GPs to go around, so gaps will show up where they are less keen on working – which sadly may tend to be the most deprived communities. The Royal College of GPs warned recently that they think “under doctored areas” do often tend to be more deprived, and have called for more to be done to attract more GPs.
In addition, given how long this problem has been discussed, we may simply be dealing with a legacy of history. General practice relies on small businesses, with partners sinking their capital and careers into local premises. It is not easy to simply redistribute resources neatly like counters on a map, particularly to where the investment opportunities may look less than appealing.
An urgent issue
Something also seems to be different between the most and least deprived areas in terms of how people get the most intensive levels of treatment. Looking at admissions to hospital across all CCGs, we can see that in wealthier areas you are consistently more likely to be admitted as a planned patient, rather than as an emergency. As you move to more deprived areas, the rate of emergency admissions climbs rapidly. However, the rate of planned care changes much less.
In fact, there doesn’t appear to be a statistically significant link between planned care and deprivation. In some of the most deprived areas, you are more likely to be admitted in an emergency than as a planned patient.
Perhaps the most striking example is Bradford City, the most deprived CCG in England. Here, waiting list and planned admissions are actually lower than average – while emergency admissions are much higher. 199 people per thousand were admitted as an emergency in Bradford in 2017, and only 149 as a planned patient. The national figures were 109 per thousand for emergency admissions, and 155 for planned and waiting list admissions.
We would expect people in more deprived areas to have generally higher health needs, so one possibility is that while they get extra emergency care to match, planned care is somehow held back. Could patients in more deprived areas, with a lower expectation of how healthy they should be, be less likely to go to a doctor and ask for help for problems that planned surgery could help with? Are hospitals in poorer areas struggling to fit in as many planned treatments because the higher rate of emergency admissions is using up more beds?
The answer to our question seems to be that there are certain aspects of care where more deprived areas do not receive as good a service as their less deprived counterparts. Exactly why this might be is well worth looking into – and we cannot rule out that some third factor not yet considered may drive both.
But it is also possible to see these findings in a more positive way. If the NHS is capable of delivering the same level of service to rich and poor when it comes to so many areas, it should be possible to replicate that everywhere, and fully live up to the service’s founding principles of equity.
Dayan M (2018) “Are parts of England 'left behind' by the NHS?” Nuffield Trust comment. https://www.nuffieldtrust.org.uk/news-item/are-parts-of-england-left-behind-by-the-nhs