Oral evidence: National Health Screening, HC 244Wednesday 11 June 2014

Oral evidence, 11 June 2014.

Witnesses

Síle Lane, Director of Campaigns, Sense About Science, Dr Margaret McCartney, Glasgow GP, and Dr John Middleton, Vice President for Health Policy, UK Faculty of Public Health, gave evidence.

Chair

Can I welcome you here this morning? Thank you very much for coming. Perhaps you would be kind enough to introduce yourselves for the record.

Dr McCartney

I am Margaret McCartney. I am a GP from Glasgow.

Síle Lane

I am Síle Lane, director of campaigns at Sense About Science, which is a charity that equips people to make sense of science and evidence. We work with groups across society, including community groups, women’s groups, patient groups, policy makers and the media to help people make sense of science and evidence. In 2009 we worked with those groups to produce “Making Sense of Screening” in response to many of them telling us about the absence of good quality information about screening programmes.

Dr Middleton

I am John Middleton. I am vice president of the UK Faculty of Public Health, which is the body for professional standards in public health in the UK. We have about 3,500 members, including people in specialist and training grades.

Chair

Thank you very much for coming. We started the earlier discussion talking about public understanding of some of the messages that go out about screening. Can I ask the three of you where you think the responsibility lies for communicating risks and benefits to the public participating in screening programmes?

Dr McCartney

From my point of view, it should be the person who is inviting the citizen for a screening test. One of the problems we have is that invitations for screening are sent from a centralised body. They often appear to come from the general practitioner, but they actually come from the screening service.

Chair

They come from both, don’t they?

Dr McCartney

In Scotland, where I work, invitations for cervical screening, for example, will appear to come from me, but I have nothing to do with the invitation letter. It has come from a centralised organisation, the screening office within the Scottish health directorate.

Chair

In England, for example, bowel screening for the over-60s just appears through the post with no reference to your GP practice.

Dr McCartney

Some screening programmes do; the letter appears to come from the general practitioner, but it comes from the screening office.

Síle Lane

I agree with Margaret. The National Screening Committee has a very evidence-based approach in looking at the benefits, harms and costs of screening programmes, but sometimes those considerations are not clearly communicated in the information they produce, so they need to lay out that information in the leaflets they produce. Ministers and civil servants need to be better briefed on this as well. When we produced this document in 2009 the Committee might remember that it was a time when there was a lot of discussion about screening among the public, led by celebrity stories, I suppose. Kylie Minogue was diagnosed with breast cancer, and the TV personality Jade Goody was diagnosed with and died from cervical cancer around then. Politicians were hearing a lot of calls and clamouring for more and wider screening programmes. They need to be able to respond to those calls as well, and that means clearly setting out the considerations that go into making decisions about screening programmes.

Dr Middleton

The question you asked is right at the heart of the whole screening issue. I agree with Margaret that the person or organisation that is leading, running and implementing a screening programme has to be responsible for the quality of the information that is provided to the person who is likely to be receiving the screening. Where we have formally agreed and implemented national screening programmes, there is a national responsibility to get the information right. We heard earlier today about the breast screening materials. There has been a very strong and conscious effort to improve the quality and content of that patient leaflet, and not hide the harms. The notion that many of my colleagues may have had, including myself when running screening programmes as a director of public health in a primary care trust, was the complicit idea that, if you tell people the whole truth, getting them into the screening programme will somehow be jeopardised. We need a much more sophisticated, honest and open set of information to give people, to enable them to make informed choices.

Chair

Clearly, while multiple skills are needed in devising effective methods of communicating, the science and accuracy of the information has to be driven by clinicians. What would you do to improve the way this kind of topic is covered in the undergraduate medical syllabus?

Dr McCartney

The very first thing I would do is try to separate the targets that in UK screening are set to get people to go for screening tests from the aim of the information.

Chair

I was looking specifically at the people. I want to make sure that the people have the right skills to do the job you want them to do.

Dr McCartney

Sure, but you have to make sure that the practitioner is there for that individual and can help them make a good, informed choice, not that the person is working to a target or trying to get that individual to do something, or not do something, on the basis of the structures within which they work, but that they are able to advocate for that person’s choices. Until you do that, it is almost irrelevant how you teach people, because they cannot put into practice what they learned as undergraduates.

Chair

I beg to differ on that. It seems to me you have to have the right skills before you can undertake the task.

Dr McCartney

True.

Chair

If you were back in medical school now, what would you change in terms of the process you went through to help you do the task that you clearly have to do as a GP?

Dr McCartney

When I was at medical school we learned about screening, and we learned about the Wilson and Jungner criteria written by the World Health Organisation to try to tell us what a good screening test is and is not. We were given that education, but it is difficult to put into practice. The knowledge is there, but we do not get a chance to use it properly because of the structures in which we work. Medical students are taught about screening. I teach screening to second-year students in Glasgow. We have a whole session devoted to the statistics, practice, ethics and difficulties of screening tests. I know you think I am not answering the question, but the problem is that it is very hard to put into practice when I do not get a chance to discuss with my patient whether or not she would like to go for breast screening, because the invitation comes centrally.

Chair

In your case, there was something in the syllabus not just about screening but about communicating risk associated with screening.

Dr McCartney

Absolutely. It is difficult. It is hard to do, but it was there.

Chair

Anybody else?

Síle Lane

It is difficult, because screening is not like any other kind of medical intervention. Talking about screening is not just setting out a list of instructions; it is not the same kind of material as telling somebody when the results of their blood test might come back. Screening comes along with a unique narrative and unique factors that need to be cut through. I am thinking of things like the direct campaigning pressure that is being exerted about screening, which is often not well thought out and not evidence-based, and does not take into account harms at all. The expectations people have about screening are not matched by what screening programmes can deliver, because our mental model of screening is diagnosis. We see the story lines in soap operas and read the personal stories in magazines and newspapers about people turning up for screening and being given a yes or no answer as to whether or not they have cancer, but screening tests are not diagnostic tests. Screening cannot do that. It aims to pick up people who are at higher risk of getting the disease so that they can go on to get more investigation and diagnostic tests and the disease can be caught earlier and treatment can be offered. Because screening programmes look at populations of healthy people, most of whom will never get the disease in question, they do not help the majority of people who turn up for them. In that context, what we ask of information about screening has to be higher; it has to cut through that narrative.

Dr Middleton

The question of undergraduate training is highly pertinent, but we need the same discussion about a higher level of knowledge in the general population. When you look at the 22 screening criteria of the National Screening Committee they seem self-evident; they seem to be things that people would expect. At the same time, looking critically at each one to see how few screening programmes meet particular criteria is crucial. In the Faculty of Public Health we are reviewing our curriculum generally for specialist training, but we have also entered into a new memorandum of understanding with medical schools about undergraduate education in public health, and for us getting this into the undergraduate programme is very important.

Chair

I think there is a slight difference between the two of you. Dr McCartney seems to suggest that when she was at medical school she got some good training. My own experience—I expressed this to the last panel—is that, on the whole, doctors are not very good at explaining risk to members of the public.

Dr McCartney

I would concur with that.

Chair

What is missing in the training?

Dr McCartney

Training is a continuous process; it does not suddenly stop when you leave medical school.

Chair

I accept that.

Dr McCartney

You leave medical school, get postgraduate education and train to be a GP, and then you realise that you don’t know anything properly, so you have to go back to the beginning and start it all again. It is a continuous, difficult process. That is matched by the fact that, when you look at the work statisticians have done about how well the public understand even basic statistics, it is pretty poor. You are starting from a difficult position. I have lots of information that I want to try to give my patients, but that information is quite difficult. It is not particularly expected information. The picture a lot of people have is that screening is easy and straightforward. It is sensible and logical; why would you not want to have it? When you come to a patient and say, “Actually, there are pros and cons, pluses and minuses,” you are opening up a big box that perhaps people did not expect to be there, and it takes a lot of time and energy when there are lots of competing priorities in the consulting room. It is very difficult. We are not very good at it. There is evidence to show that doctors do not understand screening statistics very well. That is shared by midwives, nurses and members of the public, so we have our work cut out to try to explain screening properly. I can speak only for the medical school I teach in; I certainly do not know about curriculums elsewhere, but I think it is a known problem, as opposed to an unknown one.

Stephen Metcalfe

Following up on that, in your book “The Patient Paradox” you call not for more information but for better information. We heard from the previous panel about the wide discrepancy in the quality of the information provided. How do we produce better information consistently across the wide range of screening programmes there are? How do we physically get to that point?

Dr McCartney

It is a massive project. One of the problems just now is that it is being done by lots of well-intentioned people probably working quite separately from one another. To me, the overriding principle is that we should not talk about targets for screening in terms of how many people go for screening tests; we should talk about targets for how many people make good choices about screening. Once you start with that, you have a completely different sheet of paper and you can work out how best to do it. It is immensely difficult fully to consent someone for many screening tests through a bit of paper. That is a really difficult thing to do, because screening is one of the most complex interventions in medicine. If someone is going for surgery, we sit down and explain the pros and cons; we check our patient’s understanding; we see if there are any questions; and then we go ahead with it. Screening tests do not work like that. For dementia screening—we are not allowed to call it that; we have to call it case finding—under the English GP contract, if you are in an at-risk group, and if you have high blood pressure, for example, and you see your doctor for an unrelated matter, you can be asked questions about your memory to screen you for dementia, but where is the consent given by the patient first? When did the patient get a chance to say whether or not they want this? Do they want to have a label that they have been diagnosed, for example, with mild cognitive impairment that might never lead to dementia? They could be given a label, so I do not think consent is being taken seriously enough. Once we do that, it then opens the door to say, “What information do patients need?” What level of information do patients need? Some people might just want a small amount of information—not very much at all; others might want to sit and work through an hour-long shared decision-making model on a computer. We should be able to cope with all kinds of people’s needs for information in that scenario.

Stephen Metcalfe

How do we start that piece of work? Where should it start?

Dr McCartney

I think it should start with the UK National Screening Committee, who I respect highly as an evidence-based organisation. They make very thorough, careful judgments about things. Every few years they review their screening policy; they do a thorough review and present the pros and cons and try to make clear where the uncertainties are. Starting from that model, what do we need to translate for patients? My big concern is that sometimes when people write information leaflets their big fear is that they might put people off having screening tests, but we are talking about informed choice. That means some people will choose one particular set of risks compared with another. That is called autonomy, and we have to respect it. Until we get in place the principle that people must be allowed to make a free choice, based on good information and with support from professionals, we will not get it right.

Stephen Metcalfe

You have confidence that the National Screening Committee can do that.

Dr McCartney

I have always been impressed by them. What impresses me about them is that they are unafraid to express uncertainty. Very often when screening leaflets come through it seems quite certain and obvious what should happen, whereas the UK National Screening Committee usually produce reports that say, “We are not sure about this or that.” That makes me have more confidence in them, because once we start talking about where the uncertainties are we can start to reduce them, and at least acknowledge them with our patients.

Stephen Metcalfe

Are there any advantages or disadvantages of information being produced about screening programmes outside the NHS?

Dr McCartney

There are huge problems. One of the really big ones is that we are screening but we are not calling it screening, and we are not calling it screening because it does not meet the scientific criteria for screening tests. But it is capable of doing harm. One of two big areas I would mention is dementia screening as part of the GP contract; it is called case finding, but Wilson and Jungner make no distinction between case finding and screening. I think it should have the same scientific basis and it does not. The UK National Screening Committee have made it clear that there is no evidence base for population dementia screening. The other area is health checks, which are now mandated in England. They have not been shown to be beneficial. The interim 1999 trial was published yesterday in the BMJ showing that a behavioural and risk reduction programme for cardiovascular disease does not produce a mortality reduction at 10 years. I would challenge the spending of money by Public Health England for these now mandated health checks when they cannot be sure they are going to do more good than harm. I really do not think so.

Stephen Metcalfe

What would the harm be?

Dr McCartney

There are several levels of harm. My biggest concern is that you are spending money on something that does not work and missing the opportunity to spend it on something that does. You have the healthy attender effect, where the most fit and the most healthy in society—the people who are already destined to live longer—turn up for these tests rather than the people who are not, in which case you are widening health inequalities. The first thing is the direction of public health travel. The next thing is over-diagnosis: giving people lots of diagnoses that they do not necessarily benefit from. You end up treating someone with high blood pressure or high cholesterol, which on paper might look as though it is going to reduce risk, but in real life they have a small chance of benefit. They end up taking pills for life that they think will help them, but the real improvements in health lie elsewhere, outside medication.

Stephen Metcalfe

My final question is about where the National Screening Committee is located within Public Health England. Is that the right place for it?

Dr McCartney

I am quite confused about who answers to whom, particularly in England. Scotland, England, Northern Ireland and Wales take advice from the National Screening Committee. I do not know. I would like to feel that they are as independent as possible; that can only be to the population’s advantage.

Dr Middleton

The National Screening Committee is clearly working for all four nations. It reports to a committee of the deputy chief medical officer that makes recommendations to the four chief medical officers. That much is independent and professional. The recommendations to the devolved nations about their country are clear enough. In England, NHS England is responsible for the money, the budgets and the implementation, but they are using Public Health England specialists seconded to them for the purpose and it all gets a lot messier. I doubt whether NHS England could implement a new screening programme without reference to the Secretary of State, but, strictly speaking, it is NHS England’s responsibility.

Graham Stringer

I think everybody is agreed that we want high- quality information going to people in order that they can make a decision whether to participate in a screening programme, but what might be very high-quality information to you as a trained clinician might not be high-quality information to somebody going into it. Is the information in leaflets and in other forms regularly evaluated to see if it is effective? If it is not, should it be regularly evaluated to see that it is effective in delivering information about risks and benefits?

Dr McCartney

I am not sure that it is. It would be possible to do a randomised controlled trial of information.

Graham Stringer

It would be possible.

Dr McCartney

Absolutely. It has been done before. Several American trials have looked at the effect of information, usually in the form of shared decision models. Often, they find that people will decide not to go for a screening test, but are far better informed about it. There have been trials, particularly in the States, and it would be absolutely possible to do the same in the UK.

Graham Stringer

Decisions are made. It is whether the decision is made on the basis of really understanding the information, so it is effective communication, isn’t it?

Dr McCartney

Absolutely.

Graham Stringer

That is what needs to be evaluated.

Síle Lane

That is right. It is not just the numbers, although whole numbers, real numbers and understandable numbers are vital in this information; it is also the other factors that help people understand what screening is, and what it can and cannot do. To underline what I said earlier, people have expectations of screening programmes that screening programmes just cannot deliver, so there are insights about what screening is—that it can identify some of the people who are going to get the disease, but cannot prevent it; that it can cause harms; that some false positives and false negatives are an unavoidable cost of screening populations of healthy people; and that it rarely benefits all sections of the population, so it needs to be targeted at those most likely to benefit. We found in our work with public health professionals, GPs and others that those are the insights that are most useful to the people who have to talk to other people about screening.

Dr Middleton

The pace of clinical innovation is far faster than any of the information resources in place. It has always been the case that the information and materials available to patients have not been seen as terribly high priority. We used to have a health education authority; we then had Health Education England producing stuff from a national level, but now we have a very ad hoc set of arrangements where some enthusiasts might do a one-off piece of work on bowel cancer or anything else, but sophistication about decision making—the social marketing part that makes information relevant to particular communities—is very under-resourced and unsupported.

Graham Stringer

Should the information focus on survival rates, reductions in mortality, quality of life or something else? If so, why?

Dr Middleton

That goes back to the ethical questions. Why are we doing screening? Presumably, we want to save lives and reduce disability. In some instances with some of the newborn screening it is preparing a child and a parent; for sickle cell thalassemia, it is more about preparing them for what to expect with bouts of the disease. But in each case we are obligated to do some good. We are also obligated first to do no harm. There has to be a big objective in relation to reducing disability, disease or death.

Graham Stringer

Does anybody else want to comment on whether the information should use survival rates or reduction in mortality?

Dr McCartney

One of the big problems is lead-time bias, where it can appear that a screening test is doing a lot of good but in effect is not. It has been described as two people who die of cancer in, say, 2015: one was diagnosed in 2013 through symptoms; one was diagnosed in 2010 because of screening, but they both die at the same time. It appears that one person survived for far longer through screening, and one person did not. That kind of lead-time bias can be very deceptive in the screening statistics and can often make screening look far better than it actually is. We need profound care around the kind of statistics we choose to use, because it is very easy to persuade people that screening is good rather than not. Another problem—sometimes called the popularity paradox—is when a bad test produces lots of false positives. People go on to have diagnostic tests that show them not to have the disease. Instead of feeling angry about a false positive diagnosis they are relieved that they did not have the disease, so a bad screening test becomes even more popular, because people get reassurance that they did not need in the first place.

Graham Stringer

Cancer Research tells us that identifying reductions in all-cause mortality would be the ideal way of evaluating, but it is very difficult.

Dr McCartney

Absolutely.

Graham Stringer

Why is it so difficult?

Dr McCartney

It is difficult because you need big trials, but it is very important because you can harm someone because of the treatment as an outcome from a screening test for a diagnosis that you did not need and would never benefit from. You end up doing someone harm, but saying there’s a reduction in mortality because of the treatment intervention for a positive screening test that would otherwise have gone undetected and would be unknown—so-called over-diagnosis and over-treatment. You have to anticipate that your screening intervention could do as much harm as not screening. Because of that you need all-cause mortality. You cannot just use cancer deaths as your outcome; it must encompass all possible outcomes, including the possibility that you have harmed people more through screening and interventions than through not doing that.

Graham Stringer

What is the difference in using disease-specific mortality? It is easier.

Dr McCartney

It is easier, and you can use smaller trials. That is the advantage. The disadvantage is that it is not going to pick up harms accumulated through the screening and treatment process.

Dr Middleton

The argument that with most medical conditions when we do trials we judge them against the outcome of that particular disease is more difficult with screening; saying, for instance, “We’ve stopped you dying of aortic aneurism but we haven’t prolonged your life in any way.” You have brought people into a screening programme when they did not know they had an aortic aneurism, and they make all of the choices that they are then obliged to make. It is not the same as somebody who knows they have a condition and then has a treatment, or no treatment. It feels to me that it is even more incumbent on people who are recommending a screening programme that those overall mortality results are looked at.

Mr Heath

As an aside, thinking back to my undergraduate career, such as it was, albeit in a pre-clinical degree, I do not think I was ever taught anything that related to a conscious, sentient patient, but there we are. Perhaps I missed those days. To continue the theme about evidence, what worries me about a lot of the screening programmes is that the test is in the efficacy. Is it doing more good than harm? It seems to me that in a lot of cases the evidence as to what the outcomes are is very limited. At least in the cancers we have the cancer registries; we have something that gives us evidence. We do not have that in some of the other conditions. Should that sort of approach be extended? I do not know who I am asking. Dr Middleton, would you like to pick that up?

Dr Middleton

Neonatal screening has quite sophisticated data returns from the labs that run that. As the new national screening programmes have come in there has been quite elaborate quality assurance, which has covered everything from the final outcomes through the process measures of how many people are required to go to second-stage assessments and so on—

Mr Heath

Can I interrupt you for a second? I do not know whether you were here during the previous evidence. Professor Baum directly questioned whether the statistics that accompanied the tests were accurate.

Dr Middleton

On the figures he quoted for the numbers needed to treat on screening for breast cancer, I am certainly familiar with the 2,000. Until more recently, that was the figure in my mind and in my professional practice. The figure in the breast cancer screening leaflets is that 200 women need to be screened to prevent one breast cancer death. I have seen only 84 as the number in an American journal. The range that Professor Baum described of 1,000 to 2,000 is even bigger. It means that for both monitoring and patient information it makes it extremely difficult.

Mr Heath

Dr McCartney, do you say the same?

Dr McCartney

A huge amount of quality assurance goes on within the screening programmes validated by the UK National Screening Committee. A huge amount of work goes on to ensure that it does what it says on the tin. What I am concerned about are the programmes that are screening but are not being looked after by the UK NSC, so you do not get the same quality assurance. Perhaps the audit going on is looking at the minutiae of what is going on day to day, which is incredibly important, but is perhaps not always looking at the bigger picture. Does this programme still do more good than harm? Are we still informing people properly? Did people who are now at the end stage of treatment know what they were getting into at the very beginning? Those kinds of questions are perhaps not being asked in the same way.

Mr Heath

Can I turn to what you just said? You think the monitoring is very carefully done, but if you haven’t even got accurate figures on mortality, let alone morbidity, I cannot reconcile it.

Dr McCartney

I agree. For example, if someone has an interval breast cancer in the screening programme, all the radiographs are pulled up and looked at very carefully by other people. People are trying very hard to do the best they can within the programmes, but it is the bigger picture. You are right about mortality and morbidity. Did people know what they were consenting to at the beginning? Is it still doing more good than harm? Has our population changed? The breast cancer statistics we are now using are 20 years old, and treatments have changed hugely. I suppose the bigger picture I am concerned about is not fulfilling what we should be doing, rather than the people, who I think are working genuinely hard within the programmes themselves.

Síle Lane

I would like to add support to what Margaret said. Good evidence for screening programmes depends on them being continually or at least regularly monitored, and continually weighing up that they are doing what they set out to do, or doing what they aim to do. As soon as any part of the screening programme changes, whether it is the test, or a change in treatment, or a change in society, which means different kinds of people are attending for screening, that will change the balance of benefits and harms in the programme and change its effectiveness. That needs to be continually fed back in, monitored and weighed up, as well as continually weighing up other things that are not screening that might make a difference to the detection and treatment of disease, whether it is putting more money into educating GPs about symptoms, for example, or putting into place systems for speedier referral to specialists, which might be a more cost-effective route.

Mr Heath

Is somebody doing that work? Is somebody identifying the gaps?

Síle Lane

That is the National Screening Committee’s role for the programmes they oversee. As Margaret said, there are other programmes that are not overseen by the NSC.

Dr McCartney

They do a very good job, but there are bigger questions. Who is holding them to account in terms of the bigger questions? I know they interact a lot with policymakers. I suppose it is a matter of always asking the question: are you doing more harm than good with this programme? Is it time to stop doing it? Is it time to change it radically? I know that decisions like that have been made before—for example, raising the age of cervical screening based on evidence.

Dr Middleton

Looking at cervical screening, that is an example where modification and change to a programme has happened based on good evidence—raising the age and frequency of screening because there is no benefit in older women having it three-yearly. There is a constant process of refinement of screening programmes. The commonest screening programme we all consent to is dental screening. The expectation that we have a six-month check is no longer the case. I certainly did not apply the screening criteria to dental screening, but there are holes in your teeth that can get better on their own. If 100 dentists look at your mouth, all of them will come up with a different range of work that needs doing. When we look back at the criteria for screening, the notion that we are doing good by finding it early is not always the case. Other kinds of screening are now being advocated—around domestic violence, for instance. A recent review says that universally asking women about domestic violence will not help, because we simply do not know enough about the natural history of it. Will you cause harm by meddling or investigating at that point? The criteria of screening really need to be probed in detail.

Mr Heath

I feel that I should leap to the defence of the dental profession, but I shan’t.

Chair

Clearly, screening programmes can pick up something like a cancer earlier, but it does not necessarily change the point at which the person is likely to die. It might as medicine improves. In those circumstances does it make the screening a valueless proposition, or are there other benefits the patient gets from knowing they are being properly cared for?

Dr Middleton

That is where it is no longer a clinical scientific professional judgment; it is very much in the public domain, and we need that more sophisticated discussion with patients. I think it is a major burden to be informing people of something that you are not going to treat, or giving somebody knowledge that does not help. With things like aortic aneurism screening, for instance, there is an operative mortality. By definition, the outcome or the harm of a screening programme could be people dying. The people who are given the result that they have a small aneurism that does not need operating on are in effect being told about something that they may have until the end of their life and will not be the cause of their death. These are things that need to be rehearsed more with patients and the public.

Síle Lane

There is a tension, isn’t there, between screening programmes, which are population-wide initiatives and are designed and evaluated against doing good across the population, and the fact that it is an individual who turns up for them? The programme may or may not do anything for that individual on that day; it may have no impact on their lives at all. That is a real tension.

Chair

Dr Middleton’s example plays havoc with things like insurance.

Síle Lane

Yes. It can lead to things like increased anxiety if you are told, “There may be a risk factor, so we are going to send you for further interventions.” Something that we have not yet discussed is the kind of false reassurance people can be given when they get a negative result. Another piece of information that I suggest needs to be included in any information about screening is clarity on what getting an all-clear at a screening test means. It does not mean you are never going to get that disease.

Chair

That leads me to questions about the ethical issues around screening. Should the National Screening Committee provide more explicit statements about what to consider when scrutinising the ethical acceptability of screening programmes? Is there a case for some sort of standing body to be looking at the ethics of screening programmes?

Síle Lane

I don’t know.

Dr McCartney

You have to have ethical oversight. You have to have clear medical ethics embedded in invitations, such that it is crystal clear to people that there is a risk in coming and a risk in not coming, and what those different risks are. How best that would fit in, I do not know. The UK NSC talk about ethics in their reports; they talk about the ethical issues surrounding screening in different circumstances, but I am not sure how that is done currently. It has certainly not been tested. As Mr Stringer said, there are no RCTs behind the vast majority of leaflets that I can see, so what criteria are they being held to? There does not seem to be a set of criteria that says you must ensure x, y and z, and make that clear in your invitation leaflets.

Dr Middleton

Certainly, there is scope for a more formal ethical consideration through the National Screening Committee. That would involve some professional ethicist, but I think there should be wider public involvement as well. We tend to involve stakeholders about specific screening programmes. You might have an interest in prostate cancer and therefore you get asked about it. There is Healthwatch, for instance; shouldn’t everybody who is a member of a local healthwatch understand something about the detail of screening programmes? On several occasions I did exercises on screening for elected members in the scrutiny committee. Raising the level of understanding and debate that is possible with elected members and other community interests that are not specific to a particular screening programme is important.

Síle Lane

You have to give people full and honest information. I do not know whether there is a need for another standing body to be set up to ensure that is happening. That is perhaps another discussion, but you need to give people full and honest information; otherwise, you are coercing them. You have to tell people about all the risks and all the benefits, with clear and full information about coming to screening programmes. If you cannot give someone an argument to attend a screening programme that is based on the truth, perhaps you need to look again at the screening programme.

Dr McCartney

You absolutely have to have citizen involvement wherever you go with this. The only thing I would caution about is that most people do not know if they are being harmed by screening tests. How can you represent that group of people? In the past, I worked closely with a group of women who felt they had been over-diagnosed in the breast cancer screening programme, and that they were not given proper information. It is difficult to find such people. It tends not to be vociferous members of charities speaking out in those terms, so you really have to make sure you have adequate citizen representation, not just from people who believe they are being helped by screening but also people who accept that they may have been harmed by it.

Chair

Just to reinforce the point I made earlier, part of that message needs to be about some of the consequential effects, because, taking the minor condition that is identified, it has consequential effects on the size of your wallet; insurance costs go up and things like that. Insurance companies as yet are just useless at doing anything other than tick boxes on matters like this. You have a really key responsibility, haven’t you, in communicating that risk to the public?

Dr McCartney

Absolutely. Quite often anxiety is talked about as though it is nothing. Anxiety is a horrible thing. I have met many people—there is evidence of it as well—that people who have a false positive prostate-specific antigen test are anxious even a couple of years after, compared with people who did not want to have the screening test. We really have not examined enough the psychological impacts of screening on a population. As a GP, it is the kind of thing you see all the time. Every few weeks someone comes in saying they’ve had a letter saying that their smear test is abnormal. They have been writing their will and contacting charities, and are absolutely terrified. It seems a real shame that it is only quite far down the process that someone says to them, “False positive tests are extremely common.” You can almost expect to get one as a woman if you participate in the screening programme.

Chair

My final point, developing it further, is that, as we move towards a more personalised medical system, is mass screening on a population basis compatible with that general move?

Dr McCartney

No.

Chair

What is going to happen? Do you envisage that as we get to know more about the individual patient, mass screening will slide away into the distance?

Dr McCartney

I do not know, but certainly a vast amount of private screening companies would like you to believe that that is how it should be. You can go down Harley Street today and pay £5,000 for someone to scan every bit of your body in the belief that it is the ultimate check-up. Numerous companies are trying to get in on this. We have not made it clear enough to the population that screening is a matter of pros and cons; it comes with harms and it is not the be-all and end-all. Because of that, lots of private screening companies have tagged themselves on, with the idea that more information is always better information, and are offering it to private citizens, who often pay hundreds of thousands of pounds in the private sector, which is effectively an unregulated market, and the NHS has to mop up. The problem is that people come back from their BUPA health assessment, their Harley Street check-up, or one of the Life Line Screening clinics with vast amounts of information, none of which is terribly relevant or really helps the patient. What are they telling them? To reduce their cardiovascular risk is not to be overweight, to exercise regularly, to eat a good diet, not to smoke and not to drink too much: exactly the same advice as I would have given them had they never had the screening test, but now they are 150 quid poorer for it.

Síle Lane

That is an important point. Those private screening tests are not a screening programme; it is not being run and overseen by the National Screening Committee, and it is not inviting specific people in a specific population to come along for a specific test. It is inviting everyone to come along, and I suspect that it is not people with symptoms, but mostly the worried well and healthy people who turn up for those programmes. They are offering direct consumer tests. There are lots of problems with that, as Margaret set out. Most tests were not designed to be used on well people with no symptoms. As Margaret said, tests are not regulated, so there is no need for the people offering them to have any evidence that a test will show what they say it will show. Anybody can set themselves up with a lab and start testing people. Should it be regulated?

Dr McCartney

Yes.

Stephen Mosley

Our Chairman talked about social and ethical issues. I am afraid I want to turn to money now. NICE suggests that a health intervention should cost less than about £20,000 to £30,000 per quality adjusted life year to be considered cost-effective. Should screening programmes be subject to the same threshold?

Dr Middleton

Yes. My last reading of breast cancer screening was that it certainly came within that range; I think cervical cancer is less expensive. The cost-effectiveness of screening programmes is less well evaluated, and it takes you back to the question of randomised trials. The only way you can tell that a screening programme is better than nothing, or better than the other best alternative, and to cost it, is by doing randomised trials. The answer to the question is that screening programmes should be capable of being measured on cost.

Dr McCartney

There is no reason why we should not have the same standards for cost-effectiveness of screening as for any other medical intervention, and I do not think it is being done thoroughly enough in the NHS at the moment. I am particularly concerned about programmes that are screening but claim not to be, such as dementia screening in health checks. I think they should be subject to rigorous cost-effectiveness analysis; otherwise, we are just going to keep pouring good money after bad.

Dr Middleton

To take you back to the criteria and an important health condition, syphilis in pregnancy is very rare, but the test is very cheap. Preventing syphilis in pregnancy and treating it saves a whole lifetime of disability. Those kinds of economic evaluations are not necessarily about things that are very common or that will kill you. Sometimes the rare conditions can be cost-effective, so that economic evaluation needs to be applied.

Stephen Mosley

You are saying it should be applied.

Dr Middleton

Yes.

Stephen Mosley

It needs to be applied. Can I ask you what weighting is currently given to economic factors by the UK National Screening Committee when they review evidence for a screening programme?

Dr Middleton

I cannot tell you the detail of how much economic evaluation they look at, but one of the criteria is a simple, cheap and acceptable test. Another criterion is about assessing randomised trial evidence, and there is also assessing the effectiveness of treatment. There are various points in the criteria at which cost-effectiveness can be judged. I would not say that I could tell you precisely what they do.

Stephen Mosley

Is it your impression that they are doing that, or do you think greater transparency needs to be in place in order to judge that?

Dr Middleton

I believe that they operate in a very rigorous way, but I cannot say I have seen it transparently on their website.

Stephen Mosley

I think we can deduce something from that, can’t we?

Stephen Metcalfe

I would like to come back for a few moments to the issue of private health screening. I think you said there is very little regulation.

Dr McCartney

Absolutely. I have been trying to write about these concerns over the last decade, pretty unsuccessfully it would seem. Basically, lots of full-page adverts go out, often in the quality and tabloid press, saying, “Come along for these tests at your local church hall or scout hall. We can save your life.” There is no evidence at all that that is the case. They are offering ultrasound scans, usually for the abdominal aorta and the carotid arteries in the neck; an ECG; and blood tests for cholesterol and glucose, which you could have got, had you needed them, from your GP. Therefore, the tests are available and could be done on the NHS, and the tests done are not those recommended by the UK National Screening Committee, or any other organisation. They are very poor screening tests that lead to lots of false positives. Then it flips over to the NHS, because patients come in and say, “I’ve had a leg scan done and it is a bit abnormal.” They want to see a vascular surgeon. We set up a website called privatehealthscreen.org where there is testimony from vascular surgeons saying that whenever Life Line Screening is in town they get lots and lots of appointments. They are clogged up with people who have been for an unnecessary test, and want reassurance because of it. There is a big impact on the NHS from the private health screening sector. That is ridiculous. Why should the NHS have to pick up the pieces from unwarranted private sector testing that did not need to be done in the first place?

Chair

That sounds like a strong case for regulation.

Dr McCartney

I have been to the Advertising Standards Authority who have done what they can. In two judgments, from midnight last night companies are no longer allowed to say they have saved lives, because they have no evidence of doing so. Life Line Screening say that they have screened over 250,000 people in the UK and Northern Ireland in four years. That is a huge amount of testing. If there are any abnormalities, people are told to go back to their GP. I have been to Trading Standards, who said the companies are doing what they have said they will do, so there is nothing they can do about it. I have been to the General Medical Council, because I believe that the doctors who run these clinics have been complicit in allowing misinformation and poor advertising to perpetuate. The GMC have not acted, so I would be really pleased if you would be interested in doing something about this.

Stephen Metcalfe

Regulate it or ban it?

Dr McCartney

Well—

Stephen Metcalfe

Be honest.

Dr McCartney

I think if people are going to have these things done they should know that they are paying for something that is not recommended by any reasonable evidence-based authority, and they should be responsible for paying for all follow-up tests that they wish to have as a consequence, and the NHS should be spared from having to do the private sector’s leftover dirty work for them. That is one thing. The second thing is that any advertising that these companies wish to do should be cleared by an independent group of evidence-based clinicians or researchers. I have been playing catch-up for the last decade. Every time they put up a new advert I have to make another complaint to the NSC, who usually uphold most if not all of the points I make. Then they produce a new advert and I have to go back again. They are even allowed to say that something “might” or “could” save your life when there is no evidence to suggest that is the case. We are pinning all these hopes on tests that really do not need to be done, and the big issues are the basic ones: smoking, drinking alcohol, being in work, and not being poor. All those kinds of things have a much bigger impact on your health than unnecessary screening tests.

Síle Lane

I do not know whether or not regulation is the answer, but it would perhaps help us move the information being given about these kinds of tests closer to where the National Screening Committee are moving with their evidence now, i.e. clearly laying out the benefits, harms and misconceptions about direct consumer health tests for well people. Selling tests to well people is usually not a good idea as most tests are not designed to be used on well people. They come along with all the same kinds of misconceptions as in screening programmes—that a test can give you a yes/no diagnosis immediately, whereas in reality tests are usually part of a much longer diagnostic pathway. They can give false reassurance from an all-clear. At the bottom of it is the fact that most tests were not designed to be used on well people, but these are the kind of people who are showing up for these health scans; in fact, these are the people they are being promoted to, with the message, “Wouldn’t it be better to know?”

Stephen Metcalfe

I suppose the problem is that some who want to have these tests probably think they are doing the NHS a favour. They do not want to trouble you; they have a concern and they will get a test. How do you square that particular circle?

Dr McCartney

One of the things I picked up Life Line Screening on is the fact that they had a couple of anecdotes from people who said they had been hugely helped by it, but both those people had symptoms, and symptoms need sorting out by doctors. It could be even just making it clear that screening tests are meant to be for people who are asymptomatic. The second thing is to hold them to account. Every time they put out a new advert they should be required to evidence every single statement they make in it. I did some paid work for Which? We looked at private screening companies and what information they were offering. We found that almost none—I think one out of six or seven companies—was prepared to say that their screening tests could do harm when we asked them that specifically on the telephone. There are similar groups out there who are aware of the need for balanced information, but they struggle to get their voices heard, because we are up against big glossy adverts like, “Why don’t you save your life? Come with us,” and the evidence is left as a very small voice in the corner.

Stephen Metcalfe

Presumably, when these tests are conducted either in the NHS or privately, data are collected. How is that stored and shared?

Dr McCartney

I am sorry?

Stephen Metcalfe

Let’s talk about breast cancer screening, where perhaps we end up with the figure of one in 100. There must be a result from all the screening programmes. Presumably, those data are held somewhere and then used to decide whether or not it is a valuable programme.

Dr McCartney

Yes. I think that is about quality assurance and the need for an internal audit to be consistently taking place, and it is one of the bigger questions not being offered. In the private screening sector there is none of that. They put claims on their website. One was, “We’ve saved thousands of lives.” You have to go back and challenge them and say, “Where’s your evidence?” Of course, they have no evidence because they have not followed up people in the long term. In the private sector the data disappear, as far as I can see, unless they are storing them. I have been told that BUPA store some information, but they have yet to make that available to me.

Stephen Metcalfe

Where data are collected, either privately or within NHS screening programmes, are patients made aware that the data will be collected and stored, and are they aware of how it might be used in the future?

Dr McCartney

I have no idea about the private screening sector. Within the NHS I presume it is just the same as any other information that is gathered as part of your medical record.

Stephen Metcalfe

Is there is a tiny box somewhere at the bottom that says, “We might store and use the data,” or is it made clearer than that?

Dr McCartney

I do not know. In the private screening sector, in terms of data I presume it is all under their own code, which you submit to, but in the NHS it is presumed that your data will be used to investigate how well the programme is working.

Graham Stringer

Who, if anybody, gives information about risk and benefits on NHS Health Check?

Dr McCartney

The NHS Health Check website has recently been revamped. There are lots of things on it that they say can benefit you. There are benefits to be gained if you exercise more, if you stop smoking and drink sensibly, and if you are not overweight, but I cannot find any shared decision aids from the NHS Choices website. I cannot find any choice-based information that would help people decide whether or not to attend for a check.

Graham Stringer

Do you think these tests should have oversight by a body that can independently scrutinise their effectiveness?

Dr McCartney

Absolutely.

Síle Lane

Yes.

Graham Stringer

And that is not there at the present time.

Síle Lane

No, it is not. I saw a report published last year by the charity Diabetes UK. They looked at the effect of the NHS Health Check programme four years after it was rolled out. They discovered that four years after it was rolled out there were some PCTs—at least they were PCTs in 2009 when it launched—that had never offered anybody a diabetes check as part of the NHS Health Check programme. They found that diabetes checks were being offered sporadically, illogically and randomly. Their conclusion was that because of that not enough people were being offered the diabetes check, and the subsequent intervention and lifestyle advice that comes from it, to make any difference at all to the prevalence of diabetes.

Dr Middleton

The report Margaret referred to earlier, which is the Danish study of health checks, comes out virtually as close to one as you could be—there is no difference between the tested and the not tested. It says that elements within the general health check that work should be looked at and tested. Things like quit smoking advice are perfectly sound and evidence-based interventions, but whether you need a whole national mandated programme for somebody to get quit smoking advice is another question. The health checks crept in; it was not a National Screening Committee programme. It was implemented through the reforms that Lord Darzi put through when he was a Health Minister, but it is enshrined in legislation and is part of the NHS constitution, so that will need some change in law. The Faculty of Public Health recognise that health checks do not comply with the criteria for screening programmes, but many of our service people have regard to a risk stratification approach using GP lists and trying to call up people at most risk of a heart attack in the next 10 years—people in Stoke, Haringey, Nottingham, Sandwell and Cumbria. That more targeted approach does seem to offer a real opportunity to get to the people who are at most risk, but it is not a screening programme; it is using GP practice lists, identifying people directly and case finding.

Graham Stringer

Finally, Dr McCartney, we have touched specifically on some problems with NHS Health Check, and it has come up once or twice in the questions. Can you just finish by telling us generally what your concerns are about these checks?

Dr McCartney

My concerns are that they are not proven to work and they will attract only the most healthy who have the least to gain, and we widen health inequalities; and also there is the presumption that this will make a difference to public health, so we do not concentrate on the other things that we know work—for example, fair laws on smoking or tobacco, fair laws to do with children getting to school, not needing to be driven in cars but walking instead, social inequalities and social deprivation. These are the far bigger problems that are much more likely to help reduce health inequalities. One of the big issues is that we should stop doing things that we know do not work. For example, we know that population screening for diabetes does not work; it was said in The Lancet in 2012, but it is now an integral part of the health checks. Rather than keep on doing something that does not work and trying to make it work better, just accept that we are not going to solve the problems with our health inequalities in the UK based on doing more of that. We have to do something else instead.

Graham Stringer

It is very useful to get that on the record.

Chair

To go back to Stephen Metcalfe’s question about data, in the context of screening, to whom do the data belong: the individual or NHS?

Dr McCartney

Pass. I would not be able to answer that.

Chair

I know what the answer should be.

Síle Lane

I do not know. I assume it is the same as for any NHS appointment you attend when data are collected.

Dr Middleton

You are entitled to have your data. The simple answer is that it is within the NHS screening programme. There is a bigger issue about data sharing and linkage. In relation to breast cancer, for instance, we know from the big studies that the screening programme does have an impact on mortality. We know that treatment has had a spectacular effect on reducing mortality, but we cannot tell you which of the lives saved from breast cancer have been due to the screening programme or to the treatment. Part of that is tied up with the lack of availability of routine data links between hospital and GP. You will be aware of the problems over care data, but if we are to get a more sophisticated way of measuring outcomes we need linked data between GP and hospital.

Mr Heath

I have a very quick question on diabetic screening. Has any comparison been done of the effectiveness of population screening for diabetes as such and the simple use of the eye test and retinopathy as an effective means of screening out diabetes?

Dr McCartney

Retinopathy as?

Mr Heath

Diabetic retinopathy being something picked up as an ancillary result of a routine eye test, rather than asking people to go and have a diabetic test.

Dr McCartney

I am not aware of that.

Dr Middleton

We had that kind of situation before formal diabetic eye screening came in, and we regard the diabetic eye screening programme as one of the more successful screening programmes, and certainly much more successful than ad hoc eye tests where opticians might just pick it up.

Mr Heath

If they do not pick it up, they are seriously deficient in their testing.

Dr Middleton

Yes.

Dr McCartney

If someone has a diagnosis of diabetes they automatically enter the diabetic retinopathy screening programme. If you have a diagnosis of diabetes you go in for that, but in terms of opticians picking up diabetes when doing examination of eyes, I do not think there is any data on that. I do not think that would be a useful screening test, based on a quick thought about the Wilson and Jungner criteria.

Chair

Thank you very much for your attendance this morning. It has been extremely helpful. Oral evidence: National Health Screening, HC 244 32