Are delays in diagnosis linked to the UK's poorer cancer survival?
GPs play a vital role in diagnosing cancer, but it’s not an easy job.
The average UK GP sees hundreds – if not thousands – of patients each year who have potential cancer symptoms. But fewer than eight of them will actually turn out to have cancer.
So it’s vital that GPs know what to look for, and are able to manage patients in the best way – whether by making a request for diagnostic tests, or referring a patient quickly if they do suspect cancer, or any other serious condition.
But are they? And could failings in this system explain why our national cancer survival figures fall behind other comparable countries?
New findings released today from the International Cancer Benchmarking Partnership (ICBP) – an ongoing research project involving experts from six different countries – begins to answer these questions.
Their study, published in the journal BMJ Open, had two key findings.
First, it found that the UK GPs who took part said they would be less likely to send a patient with potential cancer symptoms for tests, or to refer them to a specialist at their first appointment, compared with doctors from Australia, Canada, Denmark, Norway or Sweden.
This is the first time that differences in primary care systems have been linked with international variation in cancer survival.
And separately, it found evidence that certain tests for diagnosing cancer – CT and MRI scans – were less available to the UK’s GPs, and they had to wait longer for their results. This suggests substantial room for improvement in support for the UK’s primary care services.
So what do these results mean for the UK’s healthcare system?
Providing a benchmark
Before we look at the findings in depth – some background. The International Cancer Benchmarking Partnership (ICBP) was set up in 2009, and is co-ordinated by Cancer Research UK. It involves researchers, clinicians, data experts and policy makers from six countries: Australia, Canada, Denmark, Norway, Sweden and the UK.
As well as comparing international cancer survival – with the UK and Denmark typically lagging behind the other countries – the partnership is looking for reasons for these differences.
Over the last few years, they’ve shown that differences in public awareness and attitudes to cancer might play a role. And they’ve found that differences in the stage at which a cancer is diagnosed, and the treatment a patient receives, seems to be linked to survival differences in breast, lung, ovarian and bowel cancers.
But although these things are important, they don’t seem to paint the whole picture. So what else could be contributing? Could the answer lie with GPs? After all, GPs are a fundamental part of the health systems in each participating country.
So the partnership set out to look at whether differences in how GPs said they would manage patients with possible cancer symptoms were linked to a country’s survival statistics.
Testing the scenarios
The study revolved around a two-part online survey, and nearly 2,800 GPs took part from across the different countries (although for this study, the UK was only represented by GPs from England, Wales and Northern Ireland).
In the first part, each GP was randomly presented with two of five hypothetical scenarios, each of which described a patient with common symptoms that could be linked to cancer. GPs were then asked how they might act – for example, would they ask the ‘patient’ to come back later? Or prescribe them antibiotics? Or – as you might hope – refer them to a specialist, or for tests?
If at the first ‘visit’, the GP said they would do something other than refer the ‘patient’ for tests or to a specialist, the scenario would continue with the ‘patient’ coming back to the doctor with a change in their symptoms.
The five scenarios were devised to be clinically relevant – two involved patients with potential symptoms for lung cancer, two for bowel cancer and one for ovarian cancer.
The team was looking to see whether there were any international differences between the point at which GPs referred patients for further investigation.
Full details can be found in the paper itself, but the graphics below summarise the overall finding:
As the graphics show, in each of the patient scenarios, UK GPs were less likely to say they’d either send patients for tests, or refer them to a specialist at their first appointment, than GPs from other nations.
And crucially, in four out of five scenarios, this finding mirrored the differences between countries’ one-year cancer survival (thought to be an important indicator of whether patients are diagnosed early), and thus might partly explain international differences in cancer survival.
There was one exception: one of the lung scenarios, where the ‘patient’ had an existing long-term lung condition (chronic obstructive pulmonary disease – COPD). Patients with COPD have a greater risk of developing lung cancer, a fact which may prompt GPs to investigate or refer to a specialist.
This was the only scenario where GPs in the UK gave a similar response to colleagues in other countries.
What about Denmark?
Looking at the results, the eagle-eyed among you may have spotted that responses in Denmark buck the trend: although Denmark, like the UK has low cancer survival, its GPs seemed to be just as likely to take action as those in better performing countries.
Researchers think that this could be explained in part by significant health reforms introduced in Denmark in 2009, aimed at increasing use of diagnostic tests and speeding up the diagnostic process.
The ICBP survey took place in 2012-13 – after the reforms. But the international survival data used by the researchers, are from 2007.
So the differences seen in Denmark could suggest that the reforms are having an effect on GP behaviour, but survival data lags behind.
It will be interesting to update the analyses in the future, using more recent survival data, to see whether this pattern changes.
A little local difficulty?
So why might the UK’s GPs be less likely to refer their patients, at least initially?
This brings us to the second part of the survey, where the researchers asked GPs a series of questions about a range of issues:
- the availability of diagnostic tests and specialist cancer advice
- waiting times for test results
- systems within their practice for following up with patients
- their own attitudes and perceived role in diagnosing cancer
The researchers didn’t see any patterns in the answers to these questions that varied according to how GPs said they might act in the first part of the survey – so we can’t say for sure that these factors are behind the UK’s poorer outcomes.
However, the findings make for extremely interesting reading, and provide an insight into how the various tests and advice available in each country might influence what GPs do. And they could well contribute to a ‘wait and see’ approach in the UK, which is not seen in other countries.
For example, most UK GPs report having similar availability of blood tests, x-rays and ultrasounds as their counterparts in other countries. But only around one in five GPs in England reported having direct access to CT and MRI scans – which are more accurate tests used to diagnose cancer. In all other countries, a higher proportion of GPs reported having access to these tests than UK GPs.
And there were differences in waiting times for test results too: GPs in the UK said they had to wait far longer for the results of CT and ultrasound scans. Answers ranged from nearly five weeks in England to eight weeks in Northern Ireland. In the best performing region – New South Wales – the total waiting time is around one and a half weeks.
And – perhaps reflecting different philosophies about the role of GPs in their respective health services, GPs in the UK and Canada reported feeling more strongly about their role in protecting patients from too many tests, and in preventing specialists from being overloaded, than colleagues elsewhere.
Finally, GPs in the UK also said specialist advice was less available.
All of this suggests that there’s considerable scope for improvement in the system surrounding, and supporting, UK GPs in diagnosing cancer early.
But as is often the case with this type of research, these findings offer a starting point. Future studies will need to follow up the results, focusing on each of these areas, to work out the best ways to help GPs diagnose more patients earlier.
The study comes with caveats: response rates were lower in some countries than hoped, and the GPs who took part were not always representative of their local population. This too will need to be addressed in future research.
Understanding what’s behind the international differences in cancer survival between comparable, nations is a complex issue, but these results provide another piece of the puzzle.
So what next?
The Partnership will continue exploring these differences and identifying areas to focus policy and practice changes. Soon, we’ll see results from a study looking at the time between a patient first noticing symptoms, and subsequently being diagnosed and treated.
And another study is looking at the impact of a patient having other additional health problems (so-called ‘co-morbidities’) on short-term cancer survival. The results are eagerly anticipated, and we’ll be blogging about them over the course of 2015 and early 2016.
In the meantime, we’re continuing to work with people across the health service to ensure GPs and practice teams have the information, support and guidance they need to ensure they can confidently identify and diagnose cancer.
Through our work in GP surgeries, and our partnership with the Royal College of GPs, Cancer Research UK is aiming to do all we can to help GPs diagnose patients at an early stage.
Our programmes are providing GPs and practice teams with useful cancer data, tools to support clinical decision making, online and face to face training. And through the Accelerate, Coordinate and Evaluate (ACE) programme, Cancer Research UK is also helping to coordinate and evaluate new ways of diagnosing cancer.
The temptation would be to use these results to lay the blame at the door of the GP practice. But that’s not what the data really say.
Only by working together to make improvements across the primary care system can we close the international gap, and ensure that patients in the UK have the best chance of surviving cancer.
- Brad Groves is programme manager for the ICBP at Cancer Research UK
Rose, PW et al. Explaining variation in cancer survival between 11 jurisdictions in the International Cancer Benchmarking Partnership: a primary care vignette survey. BMJ Open (2015) DOI: 10.1136/bmjopen-2014-007212