To be published as HC 1054-i

House of COMMONS



Science and Technology Committee

Clinical Trials

Wednesday 13 March 2013

Professor Sir Michael Rawlins, Dr Keith Bragman and Dr Fiona Godlee

Evidence heard in Public Questions 1 - 36



This is an uncorrected transcript of evidence taken in public and reported to the House. The transcript has been placed on the internet on the authority of the Committee, and copies have been made available by the Vote Office for the use of Members and others.


Any public use of, or reference to, the contents should make clear that neither witnesses nor Members have had the opportunity to correct the record. The transcript is not yet an approved formal record of these proceedings.


Members who receive this for the purpose of correcting questions addressed by them to witnesses are asked to send corrections to the Committee Assistant.


Prospective witnesses may receive this in preparation for any written or oral evidence they may in due course give to the Committee.

Oral Evidence

Taken before the Science and Technology Committee

on Wednesday 13 March 2013

Members present:

Andrew Miller (Chair)

Stephen Metcalfe

David Morris

Stephen Mosley

Pamela Nash

Sarah Newton

Graham Stringer

David Tredinnick

Hywel Williams


Examination of Witnesses

Witnesses: Professor Sir Michael Rawlins, President of the Royal Society of Medicine, and Chair of the Academy of Medical Sciences Working Group on the Regulation and Governance of Health Research, Dr Keith Bragman, President of the Faculty of Pharmaceutical Medicine, and Dr Fiona Godlee, Editor in Chief of the British Medical Journal, gave evidence.

Q1 Chair: Thank you for coming this morning. I apologise for the delay, especially to you, Professor Rawlins, as I think you were rushing to get here on time and then we started late. We had a lively session, let us say, with a Minister on another subject just now. We will keep it as brief as possible because we realise that people have travel arrangements and so on.

For the record, may I ask the three of you to introduce yourselves?

Sir Michael Rawlins: I am Michael Rawlins. I am here representing the Academy of Medical Sciences, but I am also the chairman of NICE, and, if you want to throw questions at me that are more related to NICE, that is fine.

Dr Godlee: I am Fiona Godlee. I am editor in chief of the BMJ.

Dr Bragman: I am Keith Bragman. I am president of the Faculty of Pharmaceutical Medicine.

Q2 Chair: So that we understand the position that the three of you are in, what experience do you have of conducting or working with those who conduct clinical trials in the UK?

Sir Michael Rawlins: In my career, I have published the results of 19 clinical trials. I was chairman of the Committee on Safety of Medicines, so I have reviewed hundreds and hundreds of clinical trials, and it is the same at NICE.

Dr Godlee: As editor of the BMJ, I have reviewed many, many trials, and published some of them.

Dr Bragman: In my second career-this is my third one-I was working for a number of pharmaceutical companies where I have been responsible for the clinical development across a wide range of therapeutic areas, covering clinical cancer, immunology, virology and the rest of infectious diseases, inflammation and even some CNS disorders. That covers the best part of 20 or 25 years of work.

Q3 Chair: Aside from the regulatory landscape, what are currently the main barriers in conducting clinical trials in the UK?

Sir Michael Rawlins: Apart from the clinical trials directive, which is generally regarded as very flawed, hence the clinical trials regulations coming forward, there are two main barriers. This was from a review that I led on behalf of the Academy of Medical Sciences. First was the plethora of ethics approvals that sometimes needed to be garnered in order to embark on a clinical trial. The National Research Ethics Service, which has now been integrated into the Health Research Authority, was good, but, unfortunately, many other ethics arrangements had to be put in place.

By far the greatest impediment to doing clinical trials in Britain has been the fact that each single individual trust, if you are doing a multi-centre trial-and most trials are-takes on itself its own governance arrangements, with all of them looking at things like criminal records reviews and patient consent forms. The contracts go to their lawyers, and you do not send a contract to a lawyer without him commenting on it-and they all make different comments. We met one woman who had been a principal investigator in a study that involved 62 hospitals because it was a rare disease, and she had 62 CRB checks.

Q4 Chair: You are shaking your head for obvious reasons.

Sir Michael Rawlins: It is dreadful.

Dr Bragman: May I add to Professor Rawlins’ remarks? I fully endorse what he is saying. When I talk to the people who practise pharmaceutical medicine, members of the faculty who work not just in the pharmaceutical industry but also in regulatory agencies, academia, clinical practices and so on, they do not complain about problems, for example, in terms of getting ethics approval or getting approval to conduct a trial, for example, through the MHRA. The problems are very much focused on issues of governance, problems with getting contracts negotiated with many different area health authorities and with bureaucracy. These sorts of things are making clinical research increasingly unattractive here in the UK. It is very much governance-focused.

Q5 Stephen Metcalfe: You have just given us a flavour of some of the barriers. Would you explain how those relate to the clinical trials directive, or is that what you were doing? I did not quite follow whether that was so.

Sir Michael Rawlins: The clinical trials directive is the arrangements whereby the regulatory authority gives approval for a particular trial to be undertaken. The clinical trials directive-the one that is still in place-is over-burdensome, brings into its scope studies that probably should never be there and lacks proportionality. If I want, as an experiment, to give a patient a dose of paracetamol, which is pretty harmless stuff, I would still have to go through the whole business, and there would not be any distinction made between paracetamol and some rather more toxic agents.

To be fair, the MHRA has tried to take a proportional approach, but, broadly speaking, the clinical trials directive does not have a proportionality about it. That is one of the things that we criticise about it. It is one of the things that the clinical trials regulation is trying to address.

Q6 Stephen Metcalfe: On the over-burdensome aspect of the clinical trials directive, are you saying that proportionality is the biggest barrier?

Sir Michael Rawlins: It is a major barrier. There are some studies that probably should not be in it at all, and there are-

Q7 Stephen Metcalfe: I am sorry to interrupt, but could you give us an example of what should be in it and what should not?

Sir Michael Rawlins: Yes. A group of palliative care doctors wanted to see whether high-dose morphine, which is commonly used in end-stage cancer-late-stage cancer- affected cognitive thinking. That could be quite important, actually, if somebody wanted to remake a will or something like that. It was not a placebo trial for patients: they wanted to test their cognitive function. This fell within the scope of the European clinical trials directive, which meant that they had to get insurance. This was normal treatment. When they were told that the indemnity would have to be £6,000, they decided to give up and do something else. This is the sort of problem that we have.

Q8 Stephen Metcalfe: Do you think that the clinical trials regulation will help to address it?

Sir Michael Rawlins: It will help considerably, yes. There are problems; we sent in evidence about where they are, and I would not want to go over that, but it does help a lot, yes.

Q9 Stephen Metcalfe: Does anyone want to add to that?

Dr Bragman: Yes, very briefly. The quality of science and ethics should not be set at a different level irrespective of whether we are talking about a study being sponsored by a pharmaceutical company or one that is being sponsored by a university institution. Academic clinical research has greatly suffered from one-size-fits-all, and it came as a real shock to many clinical academics when they found the amount of bureaucracy that was being imposed upon them by the translation of the European clinical trials directive when it was transposed into national law. That brings us back to proportionality.

Stephen Metcalfe: There is a bit of gold plating going on.

Dr Bragman: Yes.

Q10 Graham Stringer: Sir Michael, you said a very interesting thing in answer to the Chair’s first question about having to get authority for 62 different health bodies for rare diseases. Is it more difficult when dealing with orphan diseases and orphan drugs than it is for more common issues? I have a specific example in mind, where drugs have been shown to be effective for Duchenne muscular dystrophy. How would you speed up the process? Is it more difficult in cases like that?

Sir Michael Rawlins: It depends. With Duchenne muscular dystrophy, the patients very often attend one centre so you have a pool of patients. I spent many years in Newcastle, where they have a special interest in Duchenne dystrophy, so they have a lot of patients. In a way, it is sometimes a little easier. It depends on the circumstances. But there was a second part to your question.

Q11 Graham Stringer: Is it more difficult generally with rare diseases to deal with clinical trials? That was the implication of your first answer.

Sir Michael Rawlins: Yes; if they are the sorts of diseases that do not coalesce on specific centres, it can be. There is another condition called Bell’s palsy, where the nerve to your face gets paralysed, and that comes on just like that. There aren’t centres that specialise in Bell’s palsy, so you might need to have 15 or 20 hospitals looking for cases to join a trial. It depends on the circumstances.

Q12 Graham Stringer: How effective has the Health Research Authority been in simplifying the process of getting to a clinical trial?

Sir Michael Rawlins: It has made a good start. It has obviously brought in the National Research Ethics Service, which has served well. It is bringing together other ethical approvals, so it has made a good start there. It is also looking at the feasibility and, hopefully, piloting a single sign-off, or at least a single arrangement, for governance so that the business of having to go round and get all these checks from 62 different hospitals disappears.

It is making good progress, and I don’t want to gainsay its efforts. It is also consulting with stakeholders and patient organisations.

Q13 Graham Stringer: The HRA did not take up the recommendations of the Academy of Medical Sciences in its recent report, did it?

Sir Michael Rawlins: It is the governance area that it is pursuing. It is bringing together the research ethics functions, and that is what we wanted. We also wanted it to take responsibility for doing the checks for the research governance arrangements. It is now looking into that and doing a feasibility study.

There is another way in which this might happen. In some areas, one trust is taking the lead on behalf of a group of trusts, and the academic health science networks may well also take on that sort of responsibility. I, for example, am going to chair the eastern academic health science network when I finish at NICE at the end of this month. We hope to have a single sign-off for all of the 20-odd hospital trusts in the network. That is another way of doing it.

Q14 David Tredinnick: I want to ask some questions about prospective trial registration to begin with. How important do you consider prospective trial registration to be in the move towards the increased transparency and scrutiny of clinical trials, and why do you hold those views?

Dr Godlee: It is a very important initiative. It is something that has been suggested for many years, but only recently, since about 2005, has it become something that is more widespread. Even now, it is by no means all trials; in fact, a large proportion of trials remain unregistered.

The reason why it is important is because of the problem of positive publication bias, which is well known to affect the whole of medical literature. It is optimism bias about the effectiveness of treatments being greater than it is and the safety being better. If you have a situation where more positive trials are published and we don’t know about the negative trials, people would assume that that is the truth about the treatments.

If prospective trial registration were working-at present, it is only partially working-it would ensure that we would have a full record of all the ongoing trials and, therefore, the potential to chase up and obtain the full results of those trials.

Q15 David Tredinnick: What would be the potential problems if it were to become mandatory to pre-register all clinical trials, including those unrelated to drugs? You touched on this in an earlier reply but perhaps you could expand on it.

Dr Godlee: I can see no problem with that. I can only see benefits.

Q16 David Tredinnick: Dr Godlee, this is a question for you. The BMJ has committed to publish research papers only on clinical trials that have been prospectively registered. How is compliance to this policy monitored, and how successful has the policy been in encouraging trial registration?

Dr Godlee: When the major journals first said in 2005 that they would not only publish but only peer review and consider randomised trials that had been prospectively registered, it sent a strong message. If you look at the data, it is from that point onwards that the acceleration in registration has occurred.

As you will know, there is evidence that the policy has by no means been 100% successful. Since the study was done that found that some trials that were still not registered were published in major journals, not to mention those trials that were unregistered that are published in the multiplicity of many other journals across the literature, the major journals have tightened up their processes. Certainly, the BMJ has, and, as far as we are aware, in the last two years we have not published any trial that has not been prospectively registered.

Q17 David Tredinnick: Sir Michael, this question is for you, as you generously offered to speak as chairman of NICE. You recently approved the use of acupuncture for lower back pain. I suggest to you and to other doctors that we need to broaden our analysis of available treatments. One area where a lot more could be done is with traditional Chinese medicine. Acupuncture has a 3,000-year history, and when I looked recently there were 50,000 hospitals in the People’s Republic of China using acupuncture.

I wonder, Sir Michael, if you would not agree with me that there is a great opportunity for NICE to broaden its research and to get more of these treatments approved. I myself used acupuncture for carpal tunnel syndrome to save having an operation.

Sir Michael Rawlins: We were attacked in some quarters for recommending acupuncture for lower back pain, but my colleagues who developed the guideline were convinced about the evidence for its efficacy. There are other areas, however, where acupuncture is ineffective. It is probably ineffective for migraine, where comparable trials have been done. Although we do not really understand the mechanism, in a way I do not mind if I don’t understand it. It is nice to know the mechanism by which things work, but what I am really interested in is whether they work for the benefit of patients even if I do not understand why.

Q18 David Tredinnick: That brings me on to my next point. In American institutes, trials tend to compare treatments against each other rather than against a placebo. This is a new trend. Is this not a significant development?

Sir Michael Rawlins: Well, yes, comparative trials like that are very helpful, but they are very difficult to do because of the choice of the comparator. The comparator differs in different countries. What is standard care varies, even within a country like the United Kingdom. Comparative trials are quite difficult to do if you want to generalise broadly from them.

Q19 David Tredinnick: Finally, with your indulgence, Chair, when Professor George Lewith of Southampton presented to another House Committee, the Integrated Healthcare All-Party Group, a week ago, he was talking about trials that are now being conducted to see whether herbal medicine is capable of replacing antibiotics because of the problem of antibiotics being ineffective. I wonder whether that is an area that you think should be taken more seriously.

Sir Michael Rawlins: Of course, many modern drugs are based on herbal products, such as digoxin and so on. Artemisinin, which is a herbal medicine in China used for the treatment of malaria, has now become widespread. We should take compounds wherever they come from, whether they are herbal or whatever. All I want to know is that they work.

Q20 Stephen Mosley: Going back to the regulation and regulatory reporting of clinical trials, before a medicine can be sold in the EU it needs approval. As part of that approval, you are meant to submit all clinical trial data, whether it is successful or not. Is there any evidence to suggest that all clinical trial data is being submitted, or is there any evidence that some clinical trial results are being withheld?

Dr Godlee: Shall I go first?

Sir Michael Rawlins: You go first.

Dr Godlee: As long as you follow up. It is fair to say that there is a great deal of evidence. Some cases that have been very well investigated and looked into suggest that evidence is withheld, whether on purpose or as a result of the system that we currently use. You will be aware from the written evidence that was submitted, which summarised a number of well-known cases, that drugs have been approved based on incomplete information, and that when information was provided subsequently the drug has been found either to be ineffective or even harmful.

In a famous recent case, the German regulator required the drug manufacturer Pfizer to give all the evidence on its antidepressant Reboxetine. It had not been widely used but was being used in practice. When the 75% of the evidence that had remained unpublished was finally provided, at the threat of not allowing the drug to be made available in Germany, it was found that the drug was ineffective and harmful. That is one case, but I could quote you many others that are in the written evidence. It is important to say that we are talking not only about hidden data in commercial trials but also in academic research. We recognise that there is a problem across the research enterprise.

As for whether the regulator gets all the information, this is something that is still rather murky. The industry will claim-Keith may be able to say more on this-that it does provide all the information that it is asked to provide, but I would like to hear Sir Michael tell us about his view of the current situation at NICE, how NICE relies on the EMA, and what it is that the EMA is provided with. We have good evidence on a number of drugs that the EMA is by no means provided with all the information that it would require to make a proper regulatory decision. The EMA and the FDA in America make different decisions because they get different quantities and quality of information.

Q21 Stephen Mosley: Professor Rawlins, while you are answering that, perhaps you would explain what the loophole is and how we go about filling it.

Sir Michael Rawlins: When looking at a new product at NICE, we do two things. First, we require the medical director to confirm that all relevant information has been provided to us. Secondly, we look at the scientific summary evaluation of the EMA, the European regulator, to see whether we have indeed seen all the clinical trials.

The problem is that, if things are being concealed from us, and particularly if they have been concealed from the medical director in Britain, we will not know about it. It is no good pretending that one can make a law, because the law will only influence the activities of the subsidiary company within the United Kingdom. Much as you might not like it, your writ does not go further than the English channel, as Stephen Dorrell said not long ago in a session of the Health Select Committee. If a medical director lied to us and knowingly withheld information, we would formally make a complaint to the General Medical Council.

Q22 Stephen Mosley: What would happen to that complaint?

Sir Michael Rawlins: Almost certainly the person would be struck off. Medical directors have in the past been struck off for that sort of behaviour, although not that in particular. Keith would probably throw him out of the faculty, too.

Dr Bragman: I am sure we would, Michael. Clearly there is a problem, and Fiona Godlee and Michael Rawlins have just summarised it. We need remedies. Most people who are practising pharmaceutical medicine and developing novel medicines are doing the right thing. Unfortunately, a number of individuals-I hope a small minority-need to be compelled to do the right thing. That means that we need registration of all clinical trials, we need results and we need access to data so that people can validate observations.

We are talking about the pharma industry, although I am not here to represent that industry, but it eventually comes back to the patient. We are eroding belief in medicines because people cannot trust the results that are published. I hope that it is a very small minority, but I think, Fiona, that you may have a different perspective in terms of the amount of data that you see that is potentially flawed. If we do not know what we can believe, the whole system suffers. Belief in medicine suffers, patients do not take their medicines properly, the benefits that can be derived from medicines suffer, and it costs the NHS more money at the end of the day. This is something that we need to put right.

Dr Godlee: I pick up on the point about the limits of your jurisdiction. We heard recently from the Chief Medical Officer about the need to see antibiotic resistance as a global problem and that we must act globally; we must act at the European and international level. This is another of those examples where we absolutely have to see this as an international problem. It is an international challenge, and the solutions have to be international. We look to Europe, and recent events suggest that the EMA is taking this seriously, but whether it will be able to do what it now wants to do is obviously something that we are watching with great concern. It may in some way be prevented from achieving the transparency that it wants.

Dr Bragman: May I add that the UK, whatever it does, needs to feed into the international scene so that we do not create a more bureaucratic problem or expand upon the problem, or make it more difficult for people to come and do research in the UK? We want international solutions.

Q23 Stephen Mosley: I was going to ask a question about whether the research ethics committees have a role, but from the sound of it, maybe, given your previous answer, that might not be the ideal way to go then.

Sir Michael Rawlins: I think it would only help a little bit. You have to remember that the big pharma companies are in Switzerland, North America and Japan, and your writ does not run there.

Q24 Sarah Newton: I would like to stay within this area of the reporting of information and clinical trials in relation to doctors, patients and the public and push you a little on that. Quite a few of our witnesses have said that the full publication of clinical study reports would be the way to go, but others have said that that would be quite an expensive thing to do and could add to the whole cost of the trials and therefore some things might not happen. Do you think that that would be a good route to go, and, if you do, how do you think that information should be presented? Obviously, doctors will have a degree of training to be able to understand it, but what about patients and the public?

Sir Michael Rawlins: May I start? Dr Godlee probably does not agree with me entirely.

The business about transparency has three components. The first is publishing what are often called the summary results of the trial in a conventional journal such as the BMJ. When I say conventional, I mean superb.

Dr Godlee: Superb.

Sir Michael Rawlins: It will never publish anything of mine again now. Somehow, every single trial should be placed in the public domain. The negative trials and the positive trials should all be in the public domain. That is the first thing.

The study reports are voluminous. They sometimes run to thousands of pages, and I do not think that they add very much, except to the cognoscenti. I am ambivalent about whether it is worth it. I looked at many of them in the days when I was chairman of the Committee on Safety of Medicines. They are huge, and they do not really add anything very much.

The third thing is about looking at individual patient data from clinical trials. That can potentially be very important and helpful, either to reanalyse them in a different way or to merge the results of different trials so that, instead of a trial of 10,000 patients, you have 10 trials of 100,000 patients. These individual patient meta-analyses can be extremely helpful. My big worry is ensuring the privacy of the patients. However well you anonymise it, particularly if you are talking about the rarer diseases, you can de-anonymise it and identify them. That is my big worry.

Because these meta-analyses or this single patient data are so valuable, we need to develop a mechanism whereby there is some sort of safe haven to ensure that the people who have access to the data are going to use it responsibly, particularly when it comes to the privacy of the participants. We owe it to them to protect them at all costs.

Dr Godlee: I agree with a lot of that. We have talked about the prospective registration of trials and the protocol being published and made available. Summary results should be given in a timely fashion-within a year is the figure used-but we know that that is not happening despite FDA legislation in the States. A lot of trials are still not published after a year.

We are only just discovering what is in a CSR. It is all a bit new to the academic and meta-analytical community, partly because of the Tamiflu case. The Cochrane group has been looking at that Roche drug and understanding more and more as, by painful means and very slowly, it obtains additional information. Its understanding, despite what Michael says, is that the CSR is providing information that increases discrepancies and raises more questions. It makes people realise that the basis on which we have been stockpiling this drug is probably rather more dubious than may have been understood.

The value of the CSR is something that is being explored at the moment. The alltrials.net campaign is calling for the registration of all trials, with the summary results of all trials and the CSRs to be made available. The extent to which that involves individual patient data is still very much being explored, some CSRs having patient data and others not to the same extent. I would agree that individual patient data is crucial for proper meta-analysis in certain fields, and that will require very careful management.

On the idea of having a protocol before a request is agreed to, and releases to researchers for research purposes, we are going to have to find a way to achieve that level of oversight. My basic view is that the more transparency the better. We could create a level playing field for industry on this. If we asked patients’ consent at the outset when they joined a trial, you will find that patients not only will give their consent but will start saying, "I am entering this trial because I want my results to be used for the widespread advancement of science. I am not in this trial to help industry. I am in it to help science." Consenting a patient right at the outset would be our answer to patient confidentiality problems.

Dr Bragman: May I quickly add to that? These are extremely valuable resources in terms of the raw data content. I do not want to repeat everything that has been said but simply to say that we need to know who is accessing this data. What are the questions that people want answers to? What are they going to do with the data, and what is their protocol going to be? Simply to open up these data resources to anybody, and for them to do anything that they wish and perhaps come up with claims that cannot be substantiated, could create a chaotic situation. To be able to validate claims and to be able to answer novel and new questions relating to meta-analysis is extremely important, but we need to have some form of control-not to prevent it from happening, but simply to facilitate it and know that it is being used responsibly.

Dr Godlee: Just to add to that, and I think Keith would agree, there is the question of who oversees it. Who takes the decisions is very crucial. It has to be independent. The idea that industry or manufacturers should in some way decide who sees the data and for what purpose is entirely wrong. It has to be an independent decision-making process that looks at the science and not the commercial implications.

Dr Bragman: I fully agree with that. Again, the preface is that I am not here to represent the pharmaceutical industry, but I do not think industry-some of it has, but much of it has not-has really woken up to the fact that it has so much to gain from having its data and claims validated, and that will create much more trust in the whole process and in the medicines that people take.

Q25 Sarah Newton: In your discussion you answered a lot of the questions that I was going put to you, but there are the who and how questions. Who the independent person would be who was to make these decisions and how that is all going to be done are two very important questions. Would you share with us at least your initial thoughts on the who and how?

Sir Michael Rawlins: The Academy of Medical Sciences is planning to have discussions with the Institute of Medicine in the United States about this, because these are transnational issues. It is no good one country trying to do its own thing without involving others.

Saying that patient consent is okay and satisfied is not enough. Some of these things are very complicated. Just to be a bit anecdotal, my little grandson was in a clinical trial. He is fine; it was a trial in London funded by the National Institutes of Health, and they wanted to collect his DNA in case a genetic change could account for the allergies that he was getting. That was fine, but then they asked at the end of the study for the DNA to go over to the United States. I said to my daughter-the lad was only a year old-that she had not got the right to give his DNA away. My daughter had not thought about this; she thought that she could do anything she liked with her little son. This patient business and getting patients’ agreement can be very complicated, but it is not fair on patients to opt out by saying that we have patient consent. It is more complicated than that.

Q26 Chair: Is not the reverse also a problem? I personally had to sign a consent form for myself recently on whether the results of a particular test could go into a dataset. Instead of the hospital explaining first the positive benefits of me engaging in that large group piece of research, it started the other way around by telling me how to opt out. Is that not an unwise way for hospitals to operate?

Sir Michael Rawlins: It is; it is not being fair to people.

Chair: It is not a good way of conducting science, either, is it?

Q27 Sarah Newton: Are there any more comments on the who and the how?

Dr Godlee: We have two examples in front of us about the who. One is GSK, which has said that it will set up an independent body to look at requests to reanalyse its data. We have not yet seen the detail on that-I think that they are still working on it-but there is a commitment that it will be independent, which is obviously to be applauded. However, there are many details to be hashed out. On the other hand, we have Roche, with the Tamiflu example, which suggested a group of people, who are paid experts of Roche, but there is a sense among those that I talk to that this is clearly inadequate and it will not serve the purpose.

You then have the question, if it becomes scalable, at what level it should take place. Should it take place at the European Medicines Agency or some sort of super-terrestrial body that could look across global boundaries? I expect that we will experiment and have a number of approaches to this, but it overwhelmingly has to be independent, with science at its heart and patients at its heart. We hope that things will move forward fast on this.

Q28 Sarah Newton: A lot of the conversation has been around information that will be used by other scientists to advance the body of knowledge. When it comes to explaining the results of trials to patients and the public in order to build up confidence in the system and transparency, who should be doing it and how should it be done, assuming that we can get the sorts of improvements that you have all mentioned?

Sir Michael Rawlins: Increasingly now, triallists themselves are writing to the participants to tell them the results of the trials, and we ought to encourage that.

I add to what I said before. A lot of trials are done by academics, and we need to bring them into the fold, as it were, and make sure that, whatever the arrangements, they apply appropriately to academic types of trial, which is important. We also need to remember that clinical study reports are not produced by academics. Academic triallists do not submit things to the regulatory authorities. They rely on publications in excellent journals such as the BMJ.

Dr Godlee: Your point is very important. If patients are to take part in trials and if patients and doctors are to make best use of the results, we need to know that we have all the results, and that is a big part of what you are looking at; but, assuming, in a magical world, that we reach that stage, we have to have the ability to explain why the trial was done, what it found and what comes next. The HRA may have a role in that. Certainly, we have written in the BMJ about the need for patient-friendly information at every stage in the process, in terms of when they may participate and when they are trying to interpret the results.

The education of the public and the provision of information is another challenge. At every stage, we have to avoid patronising the public. They are brilliant and bright people, and they have a lot of information already, but there is often a tendency to say, "Don’t scare patients. Don’t scare the public. We mustn’t tell them about the potential risk of this drug." I look forward to a time when we have a much more grown-up conversation with patients and the public about the risks of drugs. All treatments carry risk, and the pharmaceutical industry would do itself a lot more of a service if it allowed us to have that public conversation.

There are some drugs that have been taken off the market. Vioxx is one; it was a good drug but it had some adverse effects, and those adverse effects were hidden in all the ways that we have described in the journal that it was published in. As a result, over a period of time the drug has been withdrawn. I do not know if others would agree, but that seems to be a lost medicine of a certain sort, which a proper grown-up conversation at an earlier stage would have avoided.

Dr Bragman: I think that there is room for improvement in terms of how we communicate trial results to patients. It should be done in ways that, as you say, are not patronising, but also use a language that people can understand and relate to. I would extend this not just through clinical trials but all the way through to when the drug is approved and ready for prescription purposes. The information that accompanies the medicine-the patient information leaflet-needs to be written in a way that is appropriate and that people can relate to and understand, and that is not uniformly the case at the moment.

Q29 Chair: There is a message here for the whole of the medical profession about how to communicate with patients on things like risk. Not everyone is a statistician, but explaining risk by using analogues that people can understand-you are more likely to be run over by a bus on your way home than this procedure is likely to fail-finding some such analogue, where it is appropriate, is a much better way of engaging in a conversation than throwing numbers at people who have not been trained in statistics.

Sir Michael Rawlins: Absolutely.

Dr Bragman: Absolutely, yes.

Q30 Chair: Whose fault is it that clinical trials are not always published in peer review journals?

Dr Bragman: I think that is one for you.

Dr Godlee: In the past, journals have been at fault, although never entirely to blame. There is evidence going back 10 years that suggests that authors also keep trial or research results in their bottom drawer if they do not come up with the results that they wish for, or they move on to another project and lose interest.

More recently, there have been studies suggesting that journals are much less at fault than they ever were, with open access journals and online journals that have lots of space to publish negative and neutral results. The majority of the fault, if that is the right word, rests with authors and the sponsors of trials, who are failing to publish all of their results. As I say, it is not only full trials that are not being published but also all the results from an individual trial.

In case I don’t get the chance to say this, in the long run the solution to publishing clinical trials does not rest with journals. Journals have a role in publishing summary results, in educating doctors and patients about what the results mean, and in reviewing them and summarising them, but, with regard to the actual data, increasingly, in a timely fashion, we should be looking to see those on open access databases rather than waiting for journal publication to provide that accreditation.

Q31 Chair: That is fine, but, going back to Dr Bragman’s comments about some of the metadata, some of the datasets are getting almost impossible to publish in that form, aren’t they?

Dr Godlee: Journals don’t currently publish datasets.

Q32 Chair: I realise that, but you are inviting everyone to put their data in some public space.

Dr Godlee: Such spaces do exist. We work closely with a database called Dryad, which publishes research data, and we encourage and will increasingly mandate our authors to deposit their data in research databases of that sort. Again, it is new and developing, and these things will need funding. A piece of research will require money, not only to publish the research but to make sure that the data are available in a form that will make it re-analysable by others.

Q33 David Tredinnick: I have a quick question about Tamiflu. How significant do you think the Tamiflu incident was? This is something that the Health Committee looked at too, being concerned that the Government bought a product without having the information that they needed to assess it properly. Dr Godlee, you touched on this earlier.

Dr Godlee: The Tamiflu incident is iconic increasingly on a number of levels. It is helping us to understand the relationship between the regulator and the information that is made available by the drug company. Michael may be able to give us more on that. It has shown how much a drug-in this case a public drug-was being prescribed and made widely available, and vast sums of money were being spent both in the UK and internationally, based on information that was only available to the manufacturer.

In that case there were 10 trials, only two of which were ever published, and those two were to some extent ghost-written by paid employees of the company. The other eight trials were only published in abstract form, including the largest trial ever of Tamiflu, in terms of its effect on complications. You have a picture there, but beyond that the Cochrane group has discovered that for 123 trials, of which the vast majority if not all were Roche-funded and Roche-performed, the data were not available to anyone other than the manufacturer.

We know that NICE had some data, and we know that EMA had some data. I gather that the EMA has provided information on 16 trials. That may be all that it had. The public are being asked to believe in this drug. I don’t know if it works or not. I have no view on whether it works or not, or whether if my child was ill I would want to obtain it for my child. We are just left not knowing. The public have a right to be rather surprised that that was the basis on which the Government spent £500 million in stockpiling the drug.

The reason that it is iconic is that it may be an individual case. We are convinced by other evidence that it is not an individual case; there are other examples that suggest an endemic problem across the system. To some extent, the three of us here represent three failed groups who have tried to tackle this. They are the regulator, the academic pharmaceutical and the researcher.

Dr Bragman: I have not given up.

Chair: There are no accusations from us.

Dr Godlee: We may be failing, and we hope that you will help us to come up with a better solution.

Q34 Graham Stringer: You answered the question that I was going to ask about whether all raw data relating to clinical trials should be made available to the public, but having listened to all three of you I am not sure that I understood the answer. You seem to say that all the raw data should be made available, but you then put fairly heavy restrictions on it-that it should be made available to accredited or credible academics. Is that the right interpretation of what you were saying?

Sir Michael Rawlins: I really want to protect the confidentiality and privacy of the participants. That is my real goal. With regard to putting it out into the world so that individual people could look at it, I am not sure whether it would help anybody if you had 10,000 patients in a trial, as you could have line listings going on and on. Whether anybody who was not an expert statistician and had a huge amount of time to spare could devise anything, I am not sure, but I regard that sort of individual patient data as very important. As I said, I am very concerned about ensuring the privacy of the participants.

Q35 Graham Stringer: May I take you back to when you asked me what was the second part of my question? The question was in three parts, but I asked only two of them. The final part is this.

When you have a drug that might be effective for Duchenne or other diseases, how do you speed up the process of getting it to the patient?

Sir Michael Rawlins: Ah, isn’t that the subject of another inquiry, almost? I have been critical of the linear way in which we traditionally develop drugs, with phases 1, 2 and 3. We need to start producing swifter and sometimes leaner ways of doing it. For example, with Duchenne dystrophy, if an early stage trial had suggested benefit, I would allow it on the market subject to very close observation of those people during the first year, or two or three. That is sometimes known as adaptive licensing. We need to be much sharper about how we develop drugs. It takes such a long time, it is hugely expensive, it erodes the patent life, and at the end of the day it becomes increasingly costly.

Q36 Chair: That requires a slightly better form of informed consent, coming back to an earlier question.

Sir Michael Rawlins: Yes. This might be the subject of a separate inquiry on adaptive licensing or something like that.

Dr Bragman: May I quickly add to what Professor Rawlins is saying? We have to look at diseases like Duchenne very differently from those such as hypertension-high blood pressure. People who are developing medicines need also to understand that they can use the regulator; they can use experts to gain much more insight and discuss how to speed up the drug development process than many currently do. That is why we have this self-fulfilling prophesy in terms of the linear approach to drug development simply because that is the way that it was done in the past. It need not be so.

Chair: The medical profession has this morning succeeded 100%, because the average blood pressure of the Committee dropped down following our previous session.

May I thank you for the clarity of your evidence? We were given a lot of information in just over three quarters of an hour. Thank you very much indeed.

Prepared 20th March 2013