Keep going!: Embedding guidelines on cancer diagnostics into IT systems, NICE guidance and mouse mats

For ten years, Willie Hamilton was a jobbing GP. Like many of his colleagues, he found that current cancer guidelines, which had been developed from the perspective of acute settings, didn’t work well in primary care. Many early cancers that could have been successfully treated were going undetected. In the early 2000’s, he moved into academia to specialise in cancer diagnostics in primary care.

This was fortuitous timing. Politicians from all parties plus the Department of Health and Social Care recognised there was a problem. International comparisons showed that the UK was lagging behind other countries in the detection and treatment of cancer. This led to a national Cancer Plan, including local and national cancer action teams. The national cancer action team lead was dynamic and passionate about improving the early detection of cancer; she also had access to funding. Local cancer action teams had been set up. The leading research charity, Cancer Research UK, were keen to make a difference. The National Awareness and Early Diagnosis Initiative (NAEDI), a charity-public sector partnership, had initiated some projects. Everyone wanted a solution and knew that this required more than publishing scientific papers. But they needed the right tools.

After developing relationships with these stakeholders and carrying out smaller studies into cancer symptoms and diagnostics, Willie and his colleagues were awarded a £2 million NIHR Programme Grants for Applied Research award for the Discovery programme, which included as a major part the development of risk assessment tools in many different cancers from 2010-2016. Although many scientific papers were published, including Royal College of General Practitioners paper of the year in Lancet Oncology, to make change happen the whole team and stakeholders knew that they needed to do more.

Knowledge mobilisation is about effectively sharing knowledge between research users and research producers to make change happen. Throughout the knowledge mobilisation phase, Willie and others kept the end user – the jobbing GP – in the forefront of their minds. They carried out several initiatives including:

1. Mouse mats and desktop easels with the risk assessment tools were distributed to all English practices by local cancer action teams who had the networks and resources. This was funded by the National Cancer Team. Total production and distribution costs were about £60,000. An evaluation in 175 practices found that use of the mouse mats and easels led to increased diagnostic activity and cancers being diagnosed earlier. Another evaluation found that the training and support package, including visits by local cancer action teams were crucial.
2. The National Cancer Director set up a committee with the Royal Colleges to raise awareness of cancer. As the ‘go to’ GP who had published widely in cancer-related areas, Willie was invited to be a member. Subsequently, the National Cancer Director persuaded the charity Macmillan Cancer Support and the Department of Health and Social Care to fund electronic tools to embed the algorithms into GP IT systems. To help, Willie was videoed, but the project was driven by Macmillan, who had good engagement with local practices and managed to embed the tools into the GP patient record software Vision IT system. This software development cost several tens of thousands of pounds. Subsequently the risk assessment tools were added into SystmOne and Microtest via a research trial, thus making them available in three of the four most common GP IT systems.
3. Another project in the Discovery Programme involved developing new diagnostic pathways for cancer in two areas. In these localities, two ‘reference groups’ of local clinicians, commissioners, public health professionals and others were convened by research team members to try and incorporate the research knowledge, including the risk assessment tools into clinical pathways for lung, colorectal and pancreatic cancers. An evaluation found that one group successfully developed pathways, while the other did not, but it is not known whether the pathways were fully embedded in either site. The cost of this was about £75,000.
4. Willie has done unpaid work for influential cancer charities including Bowel Cancer UK and Cancer Research UK, such as sit on some of their policy committees. Developing these relationships is important, as these charities are powerful and have excellent links with Parliamentarians. Consequently, Wille has given evidence for the cancer All-Party Parliamentary Group three times, which he believes has fed into their decision-making particularly for pancreatic cancer, one of the cancers most difficult to diagnose.
5. Willie and his work were well known to policymakers, and the risk assessment tools fed into Department of Health and Social Care policy guidance Improving Outcomes: a national strategy for cancer (2011) and Direct Access to diagnostic tests for Cancer (2012).
6. In 2012, Willie was recruited as lead clinician of the NICE committee to revise guidance on cancer diagnostics. By now, several papers were emerging from the Discovery programme, including one that found that patients would be happier if they were investigated at a much lower risk threshold than the current 5%. Given the contentious nature of this change, Willie’s presence at the committee was essential to persuade others to take on board the findings of high-quality research evidence.

What was the impact?

Although it is difficult to disentangle the contributions that the Discovery programme made in tandem with and separate to many other initiatives to improving cancer detection, the following changes have been noted over the past few years:

• Two-week wait referrals have increased to almost 2 million annually. There is strong research evidence linking increased two-week referrals with better cancer survival.
• From 2010 to 2018 the proportion of cancers detected within the two-week-referral care pathway has increased from 42.4% to 51.3%.
• Time between first symptom presentation of cancer to primary care and diagnosis (the diagnostic interval) has decreased by a week. The best estimate is that survival worsens by 1% for each week of diagnostic delay.
• Decreased proportion of cancers diagnosed as an emergency with 6,000 fewer emergency presentations per year.
• Detection of cancer earlier equating to over 10,000 more potentially curable patients annually.

The cumulative benefit of all the above means better cancer survival for more patients.

Although this story has focused on Willie, he did not act in isolation. From early on, he developed relationships with key influential people (lead National Cancer Action team, National Cancer Director) and created partnerships with charities, the Department of Health and Social Care, and cancer networks. These served him well by providing contacts, funding and resources to reach target audiences. He also had an excellent research team and many GP colleagues working with him.

With increased awareness of the importance of cancer diagnostics at national policy level, the timing was excellent. Through a combination of his contacts and publishing widely, Willie became well-known. Although initially, he had to work to make connections, with time he found that policy-makers, parliamentarians and other decision-makers came to him.

Many different methods were used to share the research findings - mouse mats, electronic software, local pathway development, NICE guidelines and attendance at parliamentary committees. Willie and his colleagues did not rely on just one.

The Discovery programme is one of many studies, building a lifetime of work. For example, recently Willie and his colleagues won £2.75 million from a philanthropist to test the impact of electronic risk assessment tools in a cluster randomised trial. Throughout, Willie and his colleagues have combined producing high-quality research evidence and pragmatic application with on-going evaluation.

Advice from Willie

1. Always keep your end user in mind. Think from the beginning – how will this make a difference? – and hold onto that.
2. Work with third parties and charities who are interested in changing things. Build and maintain a dialogue because the results could be contentious.
3. Keep grounded in what you are doing by regularly going back to PPIE reps and clinical GPs (without academic roles) who will challenge ‘what are you going to do with this?’
4. Take the risk of trying something new.
5. Be prepared to put a lot of time into it.
6. Keep going!

Study details

Lead: Professor William Hamilton, Professor of Primary Care Diagnostics, University of Exeter

Research project: Optimising diagnosis of symptomatic cancer: the DISCOVERY programme

Funding stream: NIHR Programme Grants for Applied Research 

Dates of award: 2010-2015

Aim of study: To optimise symptomatic cancer diagnosis, with a specific focus on colorectal, lung and pancreatic cancers.

Methods: This programme grant was made up of several sub-studies including:

• A matched case-control studies using 12 million records from the General Practice Database (GPRD) to develop risk assessment for colorectal, lung and pancreatic cancers (CAPER)
• Electronic survey using vignettes to identify the levels of risk that the population would find acceptable for investigation. Previous NICE guidance had set at 5% i.e. if someone was at 5% risk of developing cancer, then that warranted investigation. (PIVOT)
• A mixed methods study using questionnaires and interviews to identify factors affecting symptom appraisal (SYMPTOM)
• An implementation study whereby groups of clinicians, public health professionals, and others worked together to develop local pathways based on the risk assessment tools (CDAPT)
• An economic modelling study to identify the most cost-effective pathways for patients presenting to GPs with symptoms that could trigger further investigations for suspected cancer.