THURSDAY 14 OCTOBER 2004

MR RICHARD BROOK, PROFESSOR DAVID HEALY AND PROFESSOR ANDREW HERXHEIMER

  Q180  Mr Burns: Professor Healy, I do not know if you were present for the earlier session?

  Professor Healy: Yes, I was.

  Q181  Mr Burns: So you will know what I am talking about. There was an exchange of questions with Dr Wilmshurst where he in his written evidence has suggested that drug companies would pay eminent cardiologists and others up to £4,000, plus expenses, for an hour long talk on their products. Maybe you will be able to help us, because I see from your CV that your background is in consultancy work with pharmaceutical companies and in recent years you have acted as a consultant, conducted clinical trials, spoken or attended foreign meetings, and then I see about three and a half lines of pharmaceutical companies that presumably you have spoken for or acted as a consultant. Is it your experience that you could earn up to £4,000, plus expenses, for an hour long talk?

  Professor Healy: I think, as you might actually expect, general psychiatrists probably come somewhat cheaper than cardiologists, but, yes, if you are asking me in the course of a day, in the course of work for a day or so, could people like me be paid of the order of £4-5,000 per day, the answer is, "Yes". Do I think that there are people in the field who are at a higher echelon than me who perhaps have closer links to the regulatory apparatus than I have who would be being paid more, the answer is, "Yes". The industry is also very clever in how they organise these things. For instance, if I get perhaps contacted . . . If I am working in a consultant capacity for one of the pharmaceutical companies, I will have had media training often, and the understanding is if the media were to get hold of you . . . Let's say some issue blows up about some pill and the media get told, "You can approach Dr Healy", for instance, I will be able to say, and the media and also the pharmaceutical company will be able to say, "Well, no money passed hands." When I was asked by the Guardian, for instance, "Is there a hazard with these pills?", there will not have been any money that has actually passed hands from me doing that piece of work for the pharmaceutical company, but the money comes from elsewhere; it actually comes from the trips to the Caribbean; it comes from being asked to chair meetings which involve no work at all; it comes from having my papers written for me and then I am paid as though I have written the papers. That is where the money comes from.

  Q182  Mr Burns: Do you have that done for you?

  Professor Healy: I have had papers written for me, sent to me, and I have said, "No, When I actually planned to get involved in this meeting, I had actually planned to write my own paper." The pharmaceutical company in question was rather surprised at this. When they saw the product that came back to them, they said, "This article that you have actually written is quite good. We think we will hold onto it, but there were certain important commercial points in the one that was done first, so we are going to have that as well", and they altered just one word, the name of the author, Siegfried Kasper, Professor of Psychiatry in the University of Vienna. There could be not a more symbolic name in all of psychiatry than professors of psychiatry from the University of Vienna. His name ended up on the piece. How much he was paid for it, you will have to ask him.

  Q183  Mr Burns: Per se do you think it is always wrong for money to change hands or is there a valid area where it is justified?

  Professor Healy: No, I do not. A great proportion of what I have had has gone into research funds. There is a charity that I have also fixed up out of which I am unable to get any funds at all; but the issue is not the funds changing hands, and the issue is not the articles being ghost-written per se. I could live in a world where the editors of journals had links to all of the major companies, where people speaking on company platforms had links to the companies; I could live with their articles being ghost-written. The crucial issue is not all of those. You do not want to tinker with things at the edge: the free pens, the ghost-written articles; the key issue is whether these articles correspond to the raw data that comes from clinical trials that your children may have been involved in for instance: that is the key point that you need to get at.

  In the absence of the Chairman, Dr Naysmith was called to the Chair

  Q184  Dr Naysmith: The more perceptive of you will have noticed that the Chair has changed. David has now left us and a different figure is in the Chair. David, I think, apologised earlier that for family reasons he had to go. I think we have a number of other areas we need to ask a few questions on yet, if the panel are happy to go on. I was going to ask again Professor Healy, who seems to be rather dominating this panel a little bit—we will have to let the other two have a say in a minute or two—you were talking just then in answer to a previous question about secrecy in drug trials and that there are things that are known that are not published and made public knowledge. When I had a proper job I used to be a scientist and one of the difficulties always was publishing negative results and what you do with things, goodwill designed experiments that come up with the answer that something does not happen which I supposed to happen; and I imagine the same sort of thing happens with drug firms, to a certain extent, that you get your correct information and it does not really tell you much one way or the other. Before I finish the question, there are two aspects to this, efficacy in whether the drug is any better than something that is on the market, and obviously the firms will try to argue that it is, and then there is the question of danger and possible risk and hazard. What is your answer to what we can do about all this: getting negative results that are meaningful?

  Professor Healy: Very quickly, and this is where I ought to hand over to people like Andrew, who has much more experience in this sort of field, but awfully quickly, the SSRIs, for instance, take one of them Seriraline, of the first sixteen trials done ten showed that the drug did not seem to work at all, two-thirds of the trials do not show this drug works. The adverts for the drugs say, however, that this is a drug that has an unparalleled evidence of efficacy. In essence the position that New York State took recently viz-a"-vis GlaxoSmithKline not publishing clinical trials where the drug had not been shown to work was that this was fraud. How can a physician give a drug to a person like you or your wife or your children when they do not really know what the true picture us. I think there is a real issue about the clinical trials that do not show that the drug works not being published, but there is a bigger issue about the ones these days that are being published. The biggest hazard, the thing that influences me and all the rest of us the most is the trials that are in the BMJ and are in the Lancet but actually are a distorted picture of what the raw data looks like. These things really do influence us even more than the trials that were left unpublished.

  Mr Brook: If I could just say, I do have an issue here, which is that, for instance, on the expert group over the 10 years Glaxo were asked to provide trial data on Seroxat and they produced over 180 trials and pieces of work, much of which has never been put into the public domain. Interestingly, in the issue around children and paediatric work, what happened there was they actually wrote to the MHRA asking for an extension of their license and they actually referred to the depression trials that were subsequently looked at by the expert group; and what they actually say is, "These trials have not quite worked out, but rather than seeing them as negative trials", ie giving a negative result, "we believe they are failed trials"—in other words they are flawed—and actually there is correspondence between Glaxo and the MHRA bringing that out as a very, very clear picture; and I think that happens quite a lot, that people are actually saying, if a trial does not quite work, as there seems to be common acceptance in the work that I was involved in, then actually you just call them failed trials rather than negative trials, and it is, of course, very hard because the trial data is not in the public domain, people do not look at it and those assessments are very hard to be challenged; and so I think we have a real issue about trial data, and I think there is a real issue about our regulator and how robust it is in actually seeking that.

  Q185  Dr Naysmith: At what stage should such data made available?

  Mr Brook: I think we are beginning to see some changes, which are very welcome but again need to be driven by regulation and law, in my view: one is obviously the beginning of having to say that all trials need to be registered before they start so you can actually track what is happening to trials, how many are getting out into the public domain; secondly, again drawing a much better line on this, so there is guidance about how good trials can be constructed; and I think trials should be signed off as being constructed appropriately, because time after time in the work I was involved in people would say, "We cannot use this trial because it has actually been done the wrong way basically", but I know Andrew has more to say.

  Q186  Dr Naysmith: He is really saying, Professor Herxheimer, that you are the expert?

  Professor Herxheimer: I think that there are two aspects to it. Negative trials—I agree with what Richard Brook has said—you would have to have registration of trials at inception, and ethics committees have to demand that a trial be registered before people are recruited to it, otherwise it should not be legal; and I think that requires a change in regulations or law. The other aspects of negative are the adverse effects. These are unwelcome to everybody and they are euphemised or disregarded or just not noticed; so that the resources spent on detecting adverse effects or suspected adverse effects, adverse experiences, by the people who take part in trials, how you actually collect those and investigate them makes a huge difference to what the data set is about that drug. To give one very simple example, if you do nothing to investigate them and only record those that are mentioned spontaneously by people taking part in a trial, you get a very small number saying, "I had this and this happened." If you ask them a general question, "Was there any change that you noticed while you were on the drug of any sort", then you get more, you get some, but if you ask for specific effects then you get much more specific answers: people know what you are talking about, what you are expecting. These methods are not described in the clinical trials reports very often and the amount of space given to adverse effects in clinical trial publications is on average as much space as is given to the title and the authors and their affiliations.

  Q187  Dr Naysmith: This sort of criticism has been going on quite a long time about the design of clinical trial. Is there any evidence that they are getting better?

  Professor Herxheimer: The International Committee of Medical Journal Editors has made recommendations about that and the journals that take part are applying better standards, and there is an international statement, consensus statement, on the reporting of randomised clinical trials which also has raised the standards enormously. The second version of that, which deals with adverse effects, is coming out in a month or two; previously it did not deal very well with adverse effects.

  Q188  Mrs Calton: I was going to ask whether you had a specific example, Professor Herxheimer, of where a pharmaceutical company did not adequately investigate side-effects?

  Professor Herxheimer: Well, because their investigations are secret, I have no idea.

  Mr Brook: Could I just say that there is actually a role for the regulator here as well, because the regulator in March of this year, when I asked them, had never bothered, for instance, to look at the adverse report data from the FDA, and there is no arrangement, for instance, for adverse reporting data in other regulatory areas to actually exchange data, so we are only looking at the UK data anyway when actually all of these drugs are used worldwide. So there is a real issue about that as well.

  Professor Healy: Can I add in on that point. It is clear that in the trials of the SSRI's, people who have gone on to become suicidal have been coded under the heading of "nausea", they have been coded under the heading of "treatment non-responsiveness", but in the group of trials Richard has seen, the group of trials in children, children becoming suicidal have been coded as being "emotionally labile", and this is a thing that very few physicians or people in the street—actually it was a person in the street from the media that picked this up; it was not actually any of the physicians that read the articles that appeared or heard the talks given that actually picked this issue up—children becoming aggressive and even homicidal were coded as "hostile". This is the kind of thing that slips under the regulator radar awfully well and under the physician's radar also, and this happens widely.

  Q189  Dr Taylor: Can I go on, Professor Herxheimer, with the adverse effects of drugs, because in your paper you gave us you give some recommendations about what should be done. Would you like to spell those out and expand on them? We have already got the message that certainly the consumer is somebody who should be reporting these sort of things. Can you expand on your recommendations?

  Professor Herxheimer: Yes. Because the regulators are funded by industry, they have no funds for doing any work of their own, and the work on adverse effects of medicines is absolutely to do with the public health, it is not to do with industry; and that is a public responsibility and it should be possible that the official organisations, including the MHRA, Medical Research Council and so on, should be able to investigate adverse effects independently—academic institutions. I think there is an argument for separating the whole analysis and collection of data on adverse effects from the MHRA, but clearly there are various ways in which that could be done. One way would be to have separate organisations doing it in the way that scientific research is done. There is a scientific community and there is an adverse-effects community and a pharmacovigilance community, but that is all governmental, and I think that is very unhealthy because it has this close relationship with the industry. So, independent collection of analysis and investigation could be funded by a modest levy on sales of pharmaceuticals. This problem is also reflected in the very small staff. There are far more people employed on evaluating applications for licences than there are who are evaluating the whole pharmaceutical market and what happens to the drugs afterwards and the people who take them. There is an enormous disproportion, and these are very hard working people, but it is impossible for them to do a proper job.

  Q190  Dr Taylor: You have certainly given us some suggestions of where we should be looking to make recommendations. I am sorry to go back to the yellow card system, but it has suddenly occurred to me that in the West Midlands we had to send yellow cards to the university department. Is that widespread? Did that mean that more reports were sent in from professionals?

  Professor Herxheimer: No, I think the reason for that is that you want these reports to be discussed locally and people to learn from them locally and to have an involvement—people should feel involved—not send them off to a black hole in London, and I think that has happened to some extent, there has been more discussion and there have been better reports, but it is still quite inadequate; so it has been a bit disappointing.

  Professor Healy: If I could just quickly add, this is a point about the reporting of adverse events that should not be seen as industry hostile. Industry to date, as you have heard earlier, has not actually been bringing new drugs on stream all that well. The research, the hypothesis driven research, to get new drugs is not working. Our biggest single source of new drugs still remains adverse events. Viagra was noted because of an adverse event; so it is the kind of thing that should be feasible to have sold to industry as a thing that both they and the consumers and the rest of us can gain from. There is a value in trying to see what these drugs do.

  Q191  Dr Taylor: The other effects—

  Professor Healy: It is not just trying to penalise the pharmaceutical companies.

  Mr Brook: Can I also just say that a large number of patients do not manage to succeed in getting their adverse effects reported. That is a consistently big issue for Mind. We have evidence over several years of people trying to report going to their GP, asking for adverse effects to be reported and the GP saying, "I do not think that is actually what has happened and so I am not doing it." I know patient reporting is now starting, but it still, I think, raises a real issue, and again you have got a number of case studies of that. The other issue that really worries me is the fact that the adverse reporting is seen as very minor in relation to clinical trials, and time after time I have been told that adverse reporting only can give a signal and it is clinical trials that are definitive. I think that is wrong. Those two must be married up. If we have got a large amount of adverse reporting we must understand what is happening here. I think it is really interesting, Seroxat had the most adverse reporting of any drug world-wide and yet it has taken all this time to sort it out. The last point I would make, which I think is a very relevant point, is it takes 40 days for the MHRA on average, in the last annual report, to licence a drug and yet takes it two years to review anti-depressants; and so there is something about the balance that Andrew was talking about that raises a real issue. If it takes eight weeks to get a drug into the market, why does it take two years sort out its safety afterwards?

  Q192  Dr Taylor: So, to come back, there must be a formalised easier route for customers, for patients, to report?

  Mr Brook: And they must be alert to what is happening and we must not just dismiss them as a signal; they are actually really big evidence.

  Professor Herxheimer: I would also like to add that the reports from patients, the MHRA has no idea how to deal with them. I think it would be far better for some other body to deal with those, obviously in connection or consultation with the MHRA, but I have no confidence in the MHRA being able to analyse and understand them.

  Q193  Dr Taylor: Does such a body exist?

  Professor Herxheimer: No.

  Q194  Dr Naysmith: Does it exist anywhere in the world?

  Professor Herxheimer: Yes, in the Netherlands there is an organisation which does that, which deals with all the reports, which is independent of the regulator, which does it for the regulator, but I think that because we do not know how to set up such a body we need to do pilots of various kinds and then work out which is the most effective.

  Q195  Mrs Calton: Professor Healy, in your evidence you say that the industry can engineer a clinical consensus that will favour their products and that they will shape assessment. You have already mentioned ghost-writing and some of the other issues, but how does such engineering go on beyond ghost-writing? What examples do you have?

  Professor Healy: We could be here for the next hour, but let me just be very brief and focused on one issue that I have grave concerns about. At present the most commonly used anti-psychotic in the UK is a drug called Zyprexa. What will happen is you will get a group of experts in and you can get the most disinterested experts who have no links to the pharmaceutical industry into a room to work out just what drugs we should use to treat people who have got schizophrenia, and you can, if you present them with the clinical trials that have been published, get them to agree that it would be a good idea if we replace older drugs like chlorpromazine, or whatever, that cost eighty times less than this drug, with a new drug like Zyprexa. You can do that quite easily. It just comes back to the issue that experts will say, "What we go on is the clinical trial evidence", but in actual fact what you have got here is that none of these clinical trials faithfully report the incidence of suicidal acts on this drug. Usually people like me, if we do not think the clinical trial evidence, the articles that are published in the BMJ, or elsewhere, are all that good, we used to be able to do a thing, we used to be able to go to the FDA website and get FDA reviews of these drugs. The MHRA has not got any website like this which would let anyone get access to any information on these drugs at all. In the case of Zyprexa, if you go to the FDA website you will find that there is not—usually, with all these drugs that are used for behavioural problems, you will get data on the number of suicides and suicidal acts that have happened in the course of the clinical trials that you cannot get in scientific literature. If you go into the bit where they have the reviews for Zyprexa you find that this section is missing completely. What this means to me is that I could not give it to you, or anyone linked to you, on an informed consent basis. This is a serious hazard of this drug: and I do not know what the hazard is. Let's say it was a huge hazard: I could say to you, "There is a huge hazard here, but I think this drug is good. We should still perhaps try using it for you, or perhaps your child". Clinical trials are happening on this drug in four-year olds at the moment. I have written to the Minister for Health here when it was Alan Milburn and he has been, it would seem, unable to get access to the data either. I had thought, in the usual course of events, the Minister for Health here cannot get access to the data that comes from clinical trials, but, given the clinical trials were happening here in the UK and there is requirement for informed consent for those trials, I had thought there was some onus on the Minister either to get access to the data or to stop the trials, but he did nothing. I have not actually, in essence, had a reply to the two lengthy letters that I wrote to him on the issue. So you can have a situation like this with a drug like this which becomes the most popularly used drug in its group here in the UK which grosses something like $4 billion a year for the pharmaceutical company in question happening; you know there are hazards of this order and you can still get experts, ones who have no links to the industry, into a room and get faced with the clinical trial evidence written up by the pharmaceutical companies and they will endorse the use of the drug, which is what has happened in this instance. This is why this drug is the drug that is the most used in the UK.

  Mr Brook: If I may refer you, you may not have seen it—I am sure you have had a huge amount of evidence—but in the Mind submission there was the GlaxoSmithKline sales manual for Seroxat and it was talking about moving towards the second billion; and if you actually look at that—its strategy is all about developing new conditions for that drug and demolishing the arguments of other competitors about why their drug was not any good. Actually the evidence that was around then and now would not back up those claims, and that is a very different, very, very useful piece of information. You can get from the manufacturers their sales manual and it clearly shows that they are actually promoting the drug into new conditions and rubbishing their competitors on the basis such as it is not addictive, and yet, five years on, we have got a leaflet produced by them which says it is 25% affecting people on withdrawal.

  Q196  Mrs Calton: I think you have already indicated, Professor Healy, that the fact that you cannot get at the information is what is causing some of the problems. How much of a problem is it that official secrecy seems to cover over all of the information that might be there, and do you think that greater transparency would make a difference to the pharmaceutical industry's contributions to public and personal health?

  Professor Healy: Yes, I think at this point the pharmaceutical companies regard the data that comes from the clinical trials as theirs, even though I do not think there is a legal basis for that. I have tried to find out what the legal basis might be, but I am not sure it is there. They regard it as theirs to be put into articles that are ghost-written up to be used to promote drugs off label for conditions that are in the process of being created, conditions like socio-phobia, what in the US is called premenstrual dysphoric disorder and things like this. What you have is a situation where they are using the data for just the kind of purposes that circumvent the regulatory apparatus, but we cannot get access to the data.

  Q197  Dr Naysmith: You have mentioned ghost-writing again, or ghost-writing has been mentioned. How does this work? In terms of footballers, famous footballers, journalists write and they stick their names on. Are you suggesting that eminent clinical scientists, academics, add their names to papers that they do not really write?

  Professor Healy: My estimate is that, even in journals like the BMJ, the Lancet, the New England Journal of Medicine and JAMA, the leading journals in the field, if these articles have to do with therapeutics, with drugs, it may be worse perhaps for psychiatry than elsewhere, but I doubt it, 50% of these articles are ghost-written. It may be higher.

  Q198  Dr Naysmith: How would that work?

  Professor Healy: It works in the sense that the industry get the data from the trials, they write the articles up, they approach authors to have their names put on the articles. These authors may not have seen the raw data at all, but they put their name to it, and they may be the most distinguished authors from the most prestigious universities. They are approached precisely because they are the most distinguished authors from the most prestigious universities.

  Q199  Dr Naysmith: I must say, that is pretty disturbing stuff?

  Mr Brook: I think what is also very interesting is that it has moved within the regulator, because the regulator actually receives summaries of trial data from the companies; and one of the things that struck me so strongly is that it is extremely rare that they do anything but read the summary as provided by the company and actually do not—they have not got the capacity or the ability to look at the raw data themselves; so it is only when you get an issue such as the Seroxat issue and the Vioyx issue, or whatever, that they start to look at the raw data. So they are totally dependent on the companies whose commercial profits are made by these drugs to produce summaries of what has happened. So it is really quite an amazing situation inside the regulator as well as ghost-writing.