Overdiagnosis: how ministers and the NHS are fuelling a rising epidemic
GPs may have lost the war against the increased medicalisation of patients, reports Emma Wilkinson
The past decade has seen overdiagnosis reach epidemic status. Unofficial screening has become more prevalent as UK guidelines have pushed down thresholds for testing, diagnosing or treating hypertension, chronic kidney disease (CKD), diabetes – and prediabetes – to name just a few.
GPs have had little choice but to go along with this advice, driven by fear of the potential consequences of deviating from guidelines, or by funding linked to specific agendas – such as the controversial incentive to identify cases of dementia in 2015.
There has been pushback from GPs, with moderate successes, such as NICE agreeing to review its contentious guidance on prescribing statins for primary prevention in over-75s (1) as part of an update of evidence for cardiovascular disease treatment and prevention.
But the problem remains that politicians and policymakers love to talk about prevention and screening, and spotting signs of a disease at an earlier stage – regardless of the evidence. And now the NHS long-term plan and the new GP contract have enshrined greater screening and case-finding into GPs’ clinical practice. So have GPs lost the war against medicalisation?
Dr Dermot Ryan, board member of the Respiratory Effectiveness Group and retired Loughborough GP, says the contract and long-term plan may tie GPs’ hands further. He says: ‘What the long-term plan is trying to do is funnel people down channels from which there is no escape – and not treating them as individuals.
‘You have a person with a condition and you have to go through certain procedures with them even if the patient doesn’t necessarily want it. Or even if they don’t have a condition – everybody over a certain age must have a statin to get their cholesterol below a certain level, even though there is no evidence in real life to support that approach.’
And he is concerned GPs may lose their autonomy along the way: ‘My worry is doctors will start treating people because they are required to do it by contract – not because it’s something that the patient needs.’
My worry is doctors will start treating people because they are required to do it by contract
Dr Dermot Ryan
Under the new GP contract in England, specifications for new networks of practices serving between 30,000 and 50,000 patients will be introduced from 2020 in line with the priorities of the NHS long-term plan, including increased screening and earlier detection of cancer and case-finding in CVD.
The long-term plan also doubles funding for the NHS Diabetes Prevention Programme over the next five years. It will build on the much-maligned NHS Health Checks, with more opportunities for the public to check if they have a high-risk condition through blood pressure checks and other tests in non-practice locations such as pharmacies and workplaces. It even plans to offer whole-genome sequencing to healthy patients who are willing to pay.
Cancer care will see the biggest changes. The thresholds for GP referral will be lowered, and the NHS is looking to maximise the number of cancers detected through screening. There are also plans to allow patients who suspect they have cancer symptoms to self-refer for rapid diagnostic testing. Lung cancer screening through breath tests and CT scans will be extended – for instance via mobile scanners in supermarket car parks.
How the GP contract and NHS long-term plan could lead to more overdiagnosis
• Practices will have ‘a key role in helping ensure high and timely uptake of screening and case finding opportunities within their neighbourhoods’. This will include bowel cancer and primary HPV screening.
• QOF quality improvement modules will be updated each year, offering opportunities for non-evidenced initiatives to be introduced.
• Two of the seven national service specifications – which networks must deliver to receive funding for the additional staff – are supporting early cancer diagnosis and CVD prevention and diagnosis.
There will also be changes on blood pressure control, and the review of the heart failure domain, plus ‘a new CVD national prevention audit for primary care’ that will be supported through a benchmarking tool. NHS England will test ‘hitherto undiagnosed patients, including through local pharmacies’.
NHS long-term plan
• A national ambition that by 2028, the proportion of cancers diagnosed at stage 1 and stage 2 will rise from about half now to three-quarters of cancer patients.
• GPs will have lower thresholds for cancer referral as well as access to rapid diagnostic centres that will be rolled out from 2019.
• The bowel cancer screening age will be reduced from 60 to 50.
• Testing for familial hypercholesterolaemia will be extended to identify at least a quarter of patients with the condition over the next five years.
• There will be direct access to cancer diagnostics for people with red-flag symptoms.
These plans were welcomed by patients – and headline writers. But it is a balancing act. In reality, the changes are likely to see more patients being referred, more screening done and more people unnecessarily put on medication – all of which increases the potential for unintended harm to patients.
Glasgow GP Dr Margaret McCartney, founder member of the RCGP overdiagnosis group, says lung cancer screening, which is being trialled in Manchester and was cited in the long-term plan, is a prime example.
‘If you screen people for lung cancer you will find lung cancer,’ she says. ‘The question is whether you are doing that in a useful way. I’m really sad we’re not taking a step back and critically and independently evaluating it.’
If you screen people for lung cancer you will find lung cancer
Dr Margaret McCartney
The National Screening Committee (NSC) is tasked with balancing competing demands – but the GP contract and NHS long-term plan are indicative of the increasing number of screening programmes that now bypass the committee.
This is often through labelling that avoids the word ‘screening’ to position them as routine activities that patients should expect. Atrial fibrillation is a case in point – the NSC ruled in 2014 that a national screening programme would not be beneficial.
Yet the GP contract and the long-term plan both specifically mention more detection of AF through new schemes.
Meanwhile, as the NHS ploughs on with new screening projects not endorsed by the NSC, the National Audit Office has recently warned that the health service is not getting the basics right in running the current evidence-based bowel, breast and cervical cancer screening programmes.
Dr Julian Treadwell, a GP and researcher in primary care at the University of Oxford, says in England – as seen in the NHS plan – policymakers are attracted by narratives of early detection and intervention, despite the absence of evidence.
Dr Treadwell says: ‘It seems to me this is a mixture of policymakers not understanding the harms and problems this causes, or understanding them but assuming the benefits must outweigh the harms – or vested interests from industry deeply embedded in the system who want more medicine generally.’
Guildford GP Dr Martin Brunet adds: ‘Often screening programmes come through the back door, like QOF or other Government incentives, and no one’s rigorously looking at the evidence.’
Often screening programmes come through the back door, like QOF or other Government incentives
Dr Martin Brunet
But resources are not infinite, he says: ‘Whenever you take a healthy person and look for disease, you’re doing so from limited resources – obviously most of our resources go to help people with symptoms. It is not an unlimited pot.’
Dr Samuel Finnikin, a GP in Sutton Coldfield and a researcher at the University of Birmingham, underlines that screening has sizeable opportunity costs and the potential for harmful overdiagnosis. It is irresponsible, he says, to use scarce NHS resources on policies that are not evidence based just because they will be popular with the public.
He adds: ‘The big problem in the whole of medicine is that we do things based on population data but the vast majority are not going to notice any benefit. You can never identify which patients have been helped or harmed.’
Clinical guideline changes contributing to growing medicalisation
Chronic kidney disease
Population-based monitoring of CKD is now routine. NICE guidance in 2008 recommended testing for patients with risk factors and changed the staging criteria. In 2014, the recommendations were updated to include additional tests to prevent overdiagnosis at the earliest stages (4)
In 2014, NICE lowered treatment thresholds to 10% – from 20% – for people with a 10-year cardiovascular risk, which made millions of new patients eligible for statins. (1,3) NICE is now in the process of updating this guidance, in response to a ‘growing body of evidence’ on the use of statins in CVD
In 2017, US guidelines lowered the threshold for stage one hypertension from an average systolic blood pressure of 140 to 130mmHg, and from 160 to 140mmHg for stage two. (5) NICE is considering the evidence from the SPRINT trial (6) that prompted those changes as part of its hypertension guidance update
Diabetes and prediabetes
An estimated one in three UK adults has blood glucose levels higher than normal but below the diabetes threshold. (7) NICE guidance on prevention in high-risk people encourages individual risk assessment.(8) For prediabetes, lifestyle change – with annual testing of blood, weight and BMI – is advised, even medication in some cases.
What is definitely known, however, is that more screening and lower thresholds risks medicalising a greater number of patients.
When it published its CKD prevalence model in 2014, Public Health England said it expected some 2.6 million over-16s in the UK to be diagnosed, including 32.7% of all over-75s.(2)
Guidance from NICE in 2014 that set new CVD treatment thresholds – stating those with a 10% or greater 10-year risk of developing CVD were eligible for drugs, instead of those with a 20% risk – created 4.5 million new candidates for statins.(1)
A study led by Harvard University researchers in 2017 estimated that if the guidance was followed to the letter, 12 million people in the UK would have been on the drugs by that point.(3)
GPs have been aware of this problem for years. The backlash ramped up in 2014 when the RCGP raised an official warning flag about medicalisation creep and set up its standing group on overdiagnosis. Then in 2016, the Choosing Wisely UK campaign, supported by the medical colleges, was set up to counteract the growing culture of medical intervention.
We are seeing phrases such as case-finding used to avoid the strict requirements of screening
Dr Andrew Green
The message had started to trickle through. In 2016, NICE published its multimorbidity guidance calling on GPs to take a tailored approach to patient care even if it meant stopping treatment it had recommended.
At one point it did seem the tide might be turning against medicalisation. But there is a belief that the wider political focus on prevention will mean this ends up being a losing battle.
BMA GP Committee prescribing lead Dr Andrew Green warns this focus is linked to the creeping use of case-finding – actively searching for groups suspected to be at risk of a disease, without them having presented with symptoms.
He says: ‘We are seeing phrases such as case-finding used to avoid the strict requirements of screening, and there is a danger when this happens of exposing patients to unnecessary risks, as well as the opportunity costs that happen when activity is introduced into health services without regard for resourcing.’
But he believes that changes in the GP contract could actually help to reduce the related problem of overtreatment.
‘The change from exception reporting to a personalised care adjustment will enable GPs to tailor care to the individual without appearing to “blame” the patient or facing accusations of gaming.’
But for Dr Brunet, there needs to be a culture shift. He says the way to counter the risks of the political focus on prevention is to empower the NSC.
He says: ‘Health secretary Matt Hancock regularly says prevention is better than cure as if that’s indisputable. Sometimes it is true – prevention by stopping smoking, for instance, is always better than cure – but finding a cancer that was never going to give you any trouble and chopping it out may not be.’
Dr Brunet says the NSC must be given authority by the Government to be the only route for screening programmes: ‘It could empower them and say “you’re the experts on what preventive medicine is good, we will be bound by what you say and won’t be allowed to come up with a screening plan you haven’t approved”.'
(1) NICE CG181. Cardiovascular disease: risk assessment and reduction, including lipid modification. London: NICE; 2014.
(2) Public Health England. Chronic kidney disease prevalence model. London: Crown Copyright, 2014.
(3) Ueda P et al. Application of the 2014 NICE cholesterol guidelines in the English population: a cross-sectional analysis. BJGP 2017; 67 (662): e598-e608
(4) NICE CG182. Chronic kidney disease in adults: assessment and management. London: NICE; 2014.
(5) American College of Cardiology. 2017 guideline for high blood pressure in adults. Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol 2018;71:e127-e248
(6) SPRINT Research Group. A Randomized Trial of Intensive versus Standard Blood-Pressure Control. NEJM 2015; 26;373(22):2103-16
(7) Mainous A et al. Prevalence of prediabetes in England from 2003 to 2011: population-based, cross-sectional study. BMJ Open 2014;4:e005002
(8) NICE PH38. Type 2 diabetes: prevention in people at high risk. London: NICE; 2012.