Achieving the world’s best cancer survival

UK cancer survival has doubled in the last 40 years. But it still lags behind the best in the world. Why?

To find out, we need to delve into data, and compare things like how quickly NHS services are operating, the stage at which patients are being diagnosed, and how many people are surviving the disease. Only by doing this can we begin to understand where there is room for improvement.

And we know it’s a complex picture. On the one hand, there are glowing tales of cancers diagnosed quickly, compassionate caring staff and world-class treatment. On the other, there are too many troubling and frustrating stories of missed opportunities for diagnosis and difficulties getting the right treatment.

We urgently need to understand how these experiences combine to cause the UK’s cancer survival to lag behind other similar countries.

The good news is that in the UK we now have a world leading network of experts dedicated towards doing just that.

And at the National Cancer Intelligence Network (NCIN) ‘Cancer Outcomes’ conference in Birmingham last month, we saw the start of some answers. We thought we’d look back to some of the data presented at the conference, and how this sits in context with previous studies.  In doing so we can explore some of the issues surrounding the UK’s lagging cancer survival rates.

How far behind ‘the best’ is the UK?

Comparing international survival rates for different types of cancer is no mean feat.

One of the first projects to do this was EUROCARE, a Europe-wide project launched in 1989. Its results were among the first to raise concerns about UK cancer survival.

At the conference, Dr Milena Sant, from Milan’s Istituto Tumori, presented some of the latest EUROCARE results which look at patients diagnosed up to 2007. They found that, despite improvements in all countries, cancer survival in the UK and Ireland (combined) remained disappointingly low.

Despite some improvements, across four cancer types – lung, kidney, ovarian and stomach – our survival is the lowest among the regions of Europe (although there is considerable variation within those regions).

For example looking at lung cancer survival between  2005-2007 15.4 per cent of patients in Central Europe survived at least five years after diagnosis, compared with just 9.5 per cent in the UK and Ireland:

Now, while these results offer some very useful comparisons they do come with a large caveat. The UK’s cancer data covers the entire population. But that’s not the case in many other countries, including wealthy nations such as France and Germany, where their ‘national’ data only include about a quarter of their population.

This means that we can’t be certain what the survival in those countries really is – you’re only looking at a small portion of the data. In other words, comparing the UK with these countries isn’t comparing like with like.

So, to try and understand these differences, we have been a key player in the International Cancer Benchmarking Partnership (ICBP), which compares countries with similar data and health care systems: the UK, Australia, Norway, Canada and Denmark, Norway and Sweden.

Unfortunately, even with this level of rigour, the UK still comes bottom of the class – as you can see from the graphic below:

At the NCIN conference, Sara Hiom, our director of patient engagement and early diagnosis, summarised the ICBP’s results so far and the impact they’ve had on practice and policy. We’ve blogged about these before – here are detailed discussions of their results on breast, lung and ovarian cancer survival. Sara also covered the results on bowel cancer, which paint a similar picture.

So, why is UK cancer survival lagging behind?

These findings are obviously of great concern to us all as more than 1 in 3 people are expected to be diagnosed with cancer in their lifetime. So where is the UK going wrong and how can it  improve?

This is a question that many researchers have been looking into, and their research falls into two broad areas; early diagnosis and access to the best treatment.

1 – Improving early diagnosis

You are more likely to survive cancer if it’s spotted early.

But the ICBP studies showed that for lung cancer and, to a lesser extent, bowel cancer, UK patients are often diagnosed at later stages compared to other similar countries. This could help explain the lower survival we see for those cancers.

Further highlighting the importance of early diagnosis, the ICBP has shown that the UK’s cancer survival looks much more like that of other similar countries if you only look at the proportion of UK patients who survive the first year after diagnosis, and who then go on to survive at least five more years.

In other words, this analysis excludes the UK patients who died within the first year – who are most often those who were diagnosed late. This phenomenon – of the UK’s overall five-year survival figures being ‘dragged down’ by the people diagnosed late, seems to occur for all the cancers studied except lung cancer.

A study by the NCIN that we’ve discussed before – called ‘Routes to diagnoses’ – supports these concerns. For patients with lung cancer, the study found that between 2006 and 2008, four in ten (39 per cent) were diagnosed via an emergency hospital admission. Patients with brain and pancreatic cancer are even more likely to be diagnosed via this route, six (62 per cent) and five (50 per cent) in ten respectively.

This is generally bad news. Survival falls for patients diagnosed via an emergency admission and this is thought to be due to them being diagnosed later.

Although the treatment patients receive will also be important, these studies give us clear evidence that diagnosing cancer earlier will boost UK survival.

Public attitudes and symptom awareness

Although the ICBP has shown that people in the UK are as aware of cancer symptoms as their international counterparts, differences emerge when we look at attitudes to using GP services.

People in the UK were more likely to mention barriers to seeing their GP. In Sweden, just one in 10 people said they would be worried about wasting the doctor’s time – in the UK it’s three in 10 (there’s more on this in our blog post from last year).

So how do we break the pattern? Avoiding stereotypical British behaviour – like keeping a stiff upper lip and worrying about bothering our doctor – would be a good start. If you spot a persistent change to your body that’s out of the ordinary then your GP needs to know – the sooner the better.

At the conference Lucy Ironmonger, from Cancer Research UK, presented results of the 2012 Be Clear on Cancer campaign for lung cancer.

The campaign appeared on TV, radio and on posters across England, urging people to go to the GP if they had a cough lasting longer than three weeks. This resulted in around 700 more people being diagnosed, 300 of whom were diagnosed early enough to go on to receive surgery, which is vital for improving their chance of survival.

Thanks to a very thorough evaluation, the team showed that the impact came from improvements across the board – from public symptom awareness to the number of people GPs sent for chest x-rays.

Referring patients to be diagnosed

People often raise concerns over whether GPs refer enough people for tests and, whether these tests are carried out quickly enough, something a number of talks at the conference examined.

It’s important to note that diagnosing cancer early is a challenge for GPs and there are many factors at play when GPs make decisions to refer or not. On average they see fewer than eight new cancer cases each year, despite thousands of patients arriving with potential cancer symptoms.

To support GPs in diagnosing cancer earlier we need to examine issues such as how easily they can refer people on for diagnostic tests, and look at the referral process to help GPs act when symptoms are less clear cut.

Sean Duffy, National Clinical Director for Cancer, presented data from the Diagnostic Imaging Dataset on the time it takes for diagnostic tests using imaging such as ultrasound and CT scans to be carried out for patients in hospital compared to those referred by a GP.

In hospital, with the high costs of having a patient staying in a bed, tests were carried out in just a few days. But when GPs requested tests it took longer for almost all the types of test. Women referred by their GP for an ultrasound test can end up waiting almost 50 days for them to be carried out, and there were additional delays in getting the results back to the GP.

This is clearly too long, prompting important questions about whether the system is placing too many restrictions on GPs. The ICBP has also been looking at how to support GPs in diagnosing patients early by using online surveys, to see whether GP attitudes or behaviour contribute to later referrals or tests.

The results are not yet published but will examine key questions including whether GPs in England, Northern Ireland and Wales may be less likely to send a patient for tests or to refer them at their first consultation than those in the other ICBP countries.

2 – Access to treatment

But early diagnosis alone does not explain the UK’s lower survival; access to the best treatment has increasingly been shown to be a problem. And this is particularly true of treatment rates in older people. Both the EUROCARE and ICBP studies have shown the survival for older patients in the UK is lower than in comparable countries.

For example, in 2010 the ICBP has shown that one-year survival for older bowel cancer patients (people over 70) who were diagnosed between 2005 and 2007 was 62 per cent in England, compared with 78 per cent in Sweden – potentially because they’re less likely to receive treatment that aims to cure them.

And in 2011, analysis from NCIN showed that in the UK, for a range of common cancers, there is a dramatic decrease in surgery rates as people get older – as this graph shows:

Of course, we can’t rule out the possibility that a patients’ treatment choices, or the stage of their disease, are contributing to these variations. But studies are now showing that the variation is not simply due to older patients being ‘too sick’ for surgery.

For example, research carried out by NCIN & Macmillan Cancer Support showed that patients aged 25-64 were around six times more likely to receive surgery than those aged 75+. There were also large differences between adjacent age brackets: the surgery rate for patients aged 70-74 was more than double that of patients aged 75+.

We’ve also seen increasing evidence that across all ages more patients should be benefiting from surgery. Victoria Coupland, from Public Health England, presented a study of pancreatic cancer death rates in relation to the proportion of patients receiving surgery. Hospitals were divided into five groups (known as quintiles) ranging from highest to lowest according to their surgery rates between 2005 and 2009 in England. They found that hospitals in the highest quintile for surgery had 12 per cent lower death rates than those in the lowest surgery quintile.

We must use these figures to learn from the best performing areas, in order to help clinicians make sure that every patient that could benefit is offered life saving treatments like surgery.

But as well as understanding more about surgery rates, data on radiotherapy and chemotherapy are steadily improving in quality, which will make it much easier to ask the same questions for these types of treatment. Reports on the radiotherapy data have already highlighted similar issues with older patients as well as points on providing the most effective and advanced forms of this type of treatment. As a key part of the treatment of many patients that are cured, it’s vital that all patients who could benefit from it have access to it, and that is the ambition we set out in our Vision for Radiotherapy.

And for chemotherapy, the Systematic Anti-Cancer Therapy dataset, which – as we heard at the conference – is nearing completion, will soon also allow us to examine the use of chemotherapy treatments across the NHS. These data should reveal prescribing patterns (i.e. the types and number of treatments prescribed) all the way down to the level of individual oncologists.

This will undoubtedly provide very exciting opportunities to examine the use of chemotherapy across the NHS, indentifying key areas for improvement.

Patient data is key

The UK’s cancer intelligence system has undergone vast improvements in recent years and continues to do so.

What really rang true at this year’s conference was that smart use of patient data allows us to focus our efforts in areas that can turn these stats around.

As the NHS emerges from enormous change, and budgets are squeezed more than ever, it’s vital to maintain and improve the NHS’s ability to always put  patients first. This will only be achieved if we have data that everyone – from commissioners to clinicians, charities to researchers, GPs and, crucially, patients – can see and use. (And we’ve recently launched our local cancer statistics website to help make that happen).

But all this could be stopped in its tracks by potential European laws – going through the EU Parliament – that would make studies like these illegal. Naturally, we’re very concerned about this.

We recently announced our ambition that in the next 20 years we want to increase the number of patients surviving cancer for at least 10 years from two in four to three in four. But without access to the data we won’t even be able to know if we are making progress, let alone how or why.

In the words of our chief clinician, Professor Peter Johnson, in a session at end of the conference: “Data is a precious and life-saving resource”. That is only true if we make it accessible, and act on what it tells us.

Matt – Matthew Wickenden is a statistical information manager at Cancer Research UK

• Image via Wikimedia Commons