UNCORRECTED TRANSCRIPT OF ORAL EVIDENCE
To be published as HC 163-i

House of COMMONS

Oral EVIDENCE

TAKEN BEFORE the

Science and Technology Committee

regulation of Medical implants

Wednesday 23 May 2012

Dr Carl Heneghan, Dr Thomas Joyce, Professor Stephen Westaby and Dr Suzette Woodward

Evidence heard in Public Questions 1 - 37

USE OF THE TRANSCRIPT

1.

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.

2.

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.

3.

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

4.

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 23 May 2012

Members present:

Andrew Miller (Chair)

Gareth Johnson

Stephen Metcalfe

Stephen Mosley

Pamela Nash

Graham Stringer

Hywel Williams

________________

Examination of Witnesses

Witnesses: Dr Carl Heneghan, GP and Reader in Evidence-Based Medicine, Director of the Centre of Evidence-Based Medicine, University of Oxford, Dr Thomas Joyce, Reader in Biotribology, Department of Mechanical and Systems Engineering, Newcastle University, Professor Stephen Westaby, Professor of Biomedical Science, Swansea University, Consultant Cardiac Surgeon, John Radcliffe Hospital, and Dr Suzette Woodward, Director of Patient Safety, National Patient Safety Agency, gave evidence.

Q1 Chair: I welcome our panel of witnesses this morning. Just for the record, may I ask the four of you to introduce yourselves?

Dr Heneghan: My name is Dr Carl Heneghan. I am from the university of Oxford.

Dr Joyce: I am Dr Tom Joyce from Newcastle university.

Professor Westaby: I am Steve Westaby from Oxford and also from the Institute of Biomedical Sciences at the university of Swansea.

Dr Woodward: I am Dr Suzette Woodward. I am from the National Patient Safety Agency.

Q2 Chair: Thank you very much. I have a series of questions for you and I will start on the issue of regulation in the field of medical implants. We have a regulatory structure that has a UK component, and it operates within the broader EU arena. Does that relationship-the decentralised nature of the structure-give you any concerns about safety in this field?

Professor Westaby: To a degree it does in that, as you already know, there are more than 80 regulatory bodies throughout the EU, and in order to obtain a CE mark you can go to virtually any of them. Once you have a CE mark for an implantable device, you can use it in any European country. The MHRA supervises the UK very carefully, and I have lots of evidence with my own endeavours to back that up. Compared with the United States and Japan, for instance, where I have substantial interface with medical devices, we are very fortunate indeed, because I can use sophisticated devices many years before my colleagues in the States and Japan, for instance.

Dr Heneghan: I have spent about three or four years trying to get my head around regulation and how it works, looking in terms of the evidence requirement. The key is to understand, from the three European directives, how the evidence is brought about. You submit your device-a design dossier-to one of the notified bodies. There are 76 across Europe; you submit to one-wherever the country is-and you get access to the whole of the European market. When you submit that design dossier, you submit your potential technical standards and clinical data. When you dig down into it, what does clinical data mean? There is a crucial issue called equivalence. All manufacturers more or less use it. In the US they use it about 95% of the time; here, I don’t know, but I expect it would be round about 100% of the time. Equivalence means that you can write a literature review that says, "My device is very similar to somebody else’s on the market and so I should have access to the whole of the European market." That is then looked at by a notified body. As of now, that is a non-transparent process between the manufacturer and a private company, and nobody, including the MHRA, gets to see the quality of that data, as submitted, what it looks like and how much it establishes effectiveness. I am not going to come to safety yet-we will come to that later. Is this device equivalent or is it better than something on the market?

Right now, probably nobody in the European Union has a perspective on that, because it is a non-transparent process that hides behind commercial confidentiality. I have access to that in an ongoing case right now, as an expert witness in evidence in transvaginal tape meshes, where I have the information submitted, and that is the only way I can get access. That non-transparency in the quality of evidence is a real problem.

Q3 Chair: Professor Westaby, going back to your response, are there examples you can give of approvals that have been given by one of the 80 regulatory bodies that specifically give you concerns?

Professor Westaby: To be honest, not in my field. I am sure that the recent PIP breast implant is the focus of attention here, but I think that was a fraudulence issue. If the regulatory authorities are told that a certain content is present in a device, they are justified in taking that as accepted. I don’t personally have experience with a device that I think should not have been used. I have had experience with a heart valve that was licensed and then caused issues. There were two well-known ones. The LabCorp valve was implanted at six or seven centres in the UK; it was a safe and durable device, but it was sterilised in a material that could cause corrosion of biological materials. There were clear instructions as to how long that particular valve should be washed before it was implanted. In certain cases, because of certain techniques, the valve was not washed for long enough; there were nurses in the operating theatres whose English was not good enough to scrutinise the instructions before use, and there were issues, but it was not the valve itself. I have not actually used a product that was questionable from the point of regulation.

Dr Joyce: Much of my work has been concerned with a group of devices-metal-on-metal hips. Within this group, we are seeing substantial revision rates and substantial failure rates-certainly a lot more than would be expected from conventional types of artificial hip joint. Those that have failed have all been through the regulatory process and have been CE marked and checked, but when it comes to pass, they have failed a massive amount. That is with hips.

If we look at something like finger joints, they have been fitted in relatively small numbers. If we look at the literature, the LPM finger prosthesis failed in about 50% of cases after about 15 to 18 months, and the Mathys finger prosthesis failed in 76% of cases, at 15 months again. Although these are small numbers, they are substantial failures and they have presumably been CE marked and checked, yet they are going into patients and performing incredibly poorly.

Q4 Chair: Is there any evidence that the regulatory structures outside Europe-say, in the US or Japan, or indeed anywhere else-are better than the European one?

Dr Heneghan: I can answer that. The US system is slightly different in that they have two systems. They have one called the 510(k) route, which is their equivalence route, but they have a second route called the PMA approval process, which they use for the highest-risk devices. The idea there is to say, "I’m sorry but you can’t use equivalence and you have to establish clinical trial data to use this product in our population." I can give you a whole list of about six different ones that have not been cleared in America but were used in the European market.

Hips are a classic example. You had a 3% failure rate, which is a very impressive revision rate, at about 10 years, and then a new device comes along on equivalence-and it is a technical equivalence, not a clinical improvement-and now you have failure rates of 13% or 15%. Whoever is responsible for the regulatory framework there is in trouble, because their job is to catch that. The key is that we have a system, and you have to remember that there are about 76 different joints available, which does not include all the combinations. There are loads of these floating out there trying to gain access to a very competitive market.

Q5 Graham Stringer: Are you saying that the metal-to-metal hip joints were allowed on the basis of an equivalence test alone?

Dr Heneghan: Nobody knows this. A notified body has looked at the data and approved it. Nobody knows the make-up of notified bodies or their skill base, and nobody knows their transparent process of approving. At least in the US, when somebody puts in a process, I can get hold of all the 510(k) information and the clinical PMA off their websites. If we were to do one thing the next time, it should be to provide a bit of transparency so that at least you could look at it and say, "I wonder what clinical data was submitted." There will be manufacturers who are conscientious and do this really well, with an evidence-based system in place, but there will clearly be manufacturers who usurp that. The answer is that I don’t know, and I very much doubt that anybody knows apart from the manufacturer and the notified body; the MHRA certainly does not know.

Professor Westaby: I have just one comment. For many years the top American companies, such as Medtronic, Edwards, and Boston Scientific, have routinely tested their products in Europe. I have been in a position on many occasions to have had the privilege to use excellent new American technology five or six years ahead of the Americans themselves and then gone across to the FDA to present our findings, which they then approve. There is no doubt that the US uses European patients to test their devices, but that is a two-edged sword. We get fantastic technology earlier.

I am very positive about MHRA and how it works. I took a concern to it about the particular heart valve that I mentioned and how it was being washed and so on, and the fact that I had confidence in the valve but not in the way that it was being dealt with. Its response was very prompt and effective; it did a very good job in preventing the sort of scandal, which would have been completely inappropriate, that we saw in the PIP implants. In my view, MHRA is very much under-resourced. It does not have enough people. There are hundreds of thousands of medical implants, and it is simply not resourced to look at them all. PIP was very unfortunate.

Q6 Stephen Metcalfe: Do you believe that there is no role for equivalence, therefore, in assessing class III implants-that it is too risky?

Dr Heneghan: You have to get your head around it. Say that Stephen here makes a wonderful device and takes it to a clinical trial, and it is shown to be effective. The problem is that you don’t incentivise anybody to do that, because the problem is that manufacturers around the world will then use equivalence to start making devices that may be slightly cheaper and come to the market. The problem is that we have no system in place to incentivise people to produce evidence.

The problem with the system is that it is expensive to do clinical trials, and most device manufacturers are not like drug companies with £20 million or £30 million floating around; they are in the innovation stage and don’t have the money to do the trials. We have to get our heads around equivalence. The FDA in America will be moving their high-risk devices to PMA and asking for clinical trials on all those devices. You might say that that is anti-innovative, but you should ask yourself where all the companies are located around the world for making devices. People should be aware of the framework of quality, if we want to build high-quality products and not lots of tat and rubbish. I am sorry about that, but that is the problem that we are facing. We will build these systems and then have other companies with equivalence come on board.

Professor Westaby: I don’t think that equivalence works. My biggest research thrust is with miniaturised artificial hearts, which are rapidly becoming equivalent to cardiac transplantation. I believe that they are very important, because the heart failure issue is a global epidemic and we have an off-the-shelf solution to it. However, you cannot say if one device works in an animal for five years that it will work in a human. We have had some big surprises that way.

The way we have to test devices in this country and in the States is with animal work. I have tested an American device; in fact, I did the first permanent implant here and got the world’s longest survival with any type of artificial heart. People like me have to do this work with animals in order to get class III devices implanted, but there is overt hostility in this country to animal work and we don’t feel protected, so I have had to go elsewhere to do it.

Q7 Stephen Metcalfe: For those particular types of device, do you think that there should be an explicit, specific requirement to have some clinical evidence before they are approved?

Professor Westaby: Yes.

Q8 Stephen Metcalfe: Does anyone disagree with that?

Dr Heneghan: The system in 1976, as it was brought around in the USA, was expected never to have equivalence for implantable devices, but they were in a bind for pre-existing devices. So it was said that there were some devices from before 1976 where you could use equivalence but any device from 1976 onwards should have clinical trials. The great thing about industry is that if they can get round anything they will do it, and they have got round that to such an extent that the majority, even in the US, is equivalence, and they are worried about that. Even the metal hip went into the US system for equivalence, passed through the FDA and got used in their market. That has been a real bind for them and a problem, so they are coming to what you are insinuating. They are calling it total product lifecycle evidence: you build pieces of equipment, and they may go wrong in 20 years, so you have to build with quality for the long term, and that requires trial data at the outset.

Q9 Stephen Metcalfe: If we move from the use of equivalence data to more clinical data for the higher-risk devices, will that delay their arrival into market? You commented on having the ability to use products perhaps five, six or seven years before they get approved by the FDA. Will there be less of a benefit to the patient by moving to an alternative system?

Professor Westaby: There could definitely be, but I sit on a Food and Drugs Administration committee relating to artificial hearts and it has recognised that things have got to change dramatically. Your artificial heart is an example of the epitome of implantable devices. If one stops, the patient dies, and I have had that in this country. To test one of these miniature artificial hearts costs a company $1 million per week-their clinical trials cost $1 million a week. I have a device, and I want it to go into humans soon, but my little company cannot afford $1 million, full stop. The FDA recognises that it has radically got to change the way that it looks at devices and licenses them. The clinical trials have got to be made accessible and far cheaper than they are now, because right now it just does not work. Animal work, as I say, does not necessarily translate into what you find with a human; there have been big surprises in the translation of animal work to humans. It has to be clinical, but it has to cost less. In my view, the NHS has got to support it far better than it does.

Dr Heneghan: Can I give you a perspective? It will delay things, but that is not a bad thing actually.

Let us take a simple device, such as a blood pressure machine, which is an important device. If that machine is not accurate, it could have profound implications for who gets treatment and who does not. Beyond the CE system, the British Hypertension Society and the European Society of Hypertension have worked out that they need clinical studies, kitemarks and protocols to stamp on these devices to say they are accurate. That is profoundly important if you are going to put a million people on treatment, because if the machines are not accurate we will have major problems. We do it better for simpler devices than we sometimes do for the high-end devices.

Will it delay things? Yes. With a new drug on the market, would you expect to say, "Let’s just see how it goes on for a few years", and after a few years you put it through trials and then get NICE to look at it? The perspective is that MHRA is a different body from what NICE is and does for you. NICE says, "Is this drug safe and does it have any benefit for society?" That body is not in place, so there is an argument that you could look at something like NICE and say, "What is NICE’s remit?" There is a medical technology committee, a device committee and a diagnostic guidelines committee. Its remit should be to put a stamp mark on products and say, "Yes, this will give us some benefit."

Q10 Hywel Williams: I don’t want to go off on a tangent about NICE, but doesn’t NICE operate differently in Scotland from the way it operates in England and Wales? It is a quicker process in Scotland. I might be wrong about this. Is there anything to be learned by comparing and contrasting as far as this particular issue is concerned?

Dr Heneghan: Certainly, and there is a lot to be learned about international collaboration as well.

Q11 Stephen Metcalfe: Why do you think those two separate paths have been developed? Why do you think NICE went down the route that it did, with evidence-based clinical assessment, and we have not done the same for implants, which you might argue could have an equal impact on the patient?

Dr Heneghan: Yes, you could say that. NICE was set up for a postcode lottery in drugs and medications. They said, "We’ve got people using drugs over here and they are not using it there, particularly in areas such as cancer; we need a perspective across the NHS." It is an amazing innovation to say that that is what we do, but then people moan because they have to wait a bit longer-but you are better waiting.

Professor Westaby: May I just say that I am on the NICE Medical Technologies Advisory Committee? Its remit is only to look at devices that are submitted for assessment by companies, and there is a paucity of devices submitted to NICE to look at. The American companies will not go anywhere near NICE; they can market their products without being involved with NICE. There is an understanding that NICE is a lottery; it is based on cost containment, and most important devices never go anywhere near NICE.

Dr Heneghan: However, the flipside to that is in the US medical area in their purchasing activity. They do their own sort of technology assessments and work out the value for money and effectiveness before they decide to pay for it.

Q12 Chair: Before we move on, may I ask you, Dr Woodward, what you see as the patient perspective on this part of the debate, on delays versus the benefits?

Dr Woodward: May I slightly clarify the role of the National Patient Safety Agency? It is about looking at the whole system and trying to make it as safe as it possibly can be in a proactive and reactive way in the instant reporting system that we run. We run a massive national reporting system, and anything that could possibly go wrong in the NHS gets reported to us. From our perspective, we look at what it feels like and how it is perceived by patients, but it is also very much with staff and how staff feel when things go wrong.

From a patient’s perspective, you absolutely expect that when you are going to go in you are going to have the right thing put in the right place at the right time. That is what you should have, and that is also what you expect, so you don’t question it. If somebody says you need a replacement, you don’t question where the replacement is coming from, whether somebody has tested it in the right place or whether it is the right piece of equipment for you. If somebody tells you that you need a hip replacement, you sweetly say yes, sign the consent agreement, and go into surgery. When you come out, if things go right, that is fabulous, but if things go wrong, it is a very hard system to cope with. When things go wrong, you are rarely told that they have gone wrong, and you are rarely told why, and you don’t quite understand what can be put right for you as an individual. You also start to become very fearful of the system. You don’t come back into it because it doesn’t feel right; that frightens you, so you don’t want to have it happen again. It is a very confusing scenario for patients when things go wrong, and it is almost an ignorant system before they go wrong because you believe that everything is going to be right.

Q13 Gareth Johnson: Professor Westaby, I presume that decisions that you make about which particular implants to use for your patients are based on medical need and the available information that you have at the time about medical data. Do you suspect that any decisions you have made would have been different if you had had full access to the pre-market medical data?

Professor Westaby: No, I don’t believe that any clinician working in the NHS ever gets access to the pre-market data at all. At clinician level, you are provided with an evidence base to some degree, depending on experience, and this is where registries are very important. There are now registries for all types of implant, and registry data tends to be more instructive than clinical trial data, because a lot of what is published in terms of clinical trial is done by enthusiasts in the best units under ideal conditions. In contrast, registry data tells you how everybody has got on with the product and you can pick up variables from registry data.

Q14 Gareth Johnson: Professor Westaby, we have heard calls today for a far more transparent system than is currently the case. Presumably, one of the reasons those calls have come forward is because it would have an impact on patients.

Professor Westaby: Yes.

Gareth Johnson: In what way do you think it would change the medical decisions that you have made if we had a far more transparent system?

Professor Westaby: I have always been an academic to a degree, so I have always looked for an evidence base for products before I have used them. That comes from a massive amount of literature in journals. I have to say that most devices in clinical use are sold by enthusiastic representatives.

It is not so long ago that trips to international conferences, pens, biros and everything else were given out by companies, and clinicians would make subjective assessments on which type of heart valve, pacemaker or stent to use depending on the representatives that they liked the best and the companies that looked after them the best. That is the way that the English market has worked for an awfully long time. I think there is regulation about that. In the United States the regulations about these things are very tight now, and I think it is similar in the UK now. You are not allowed to make judgments depending on which rep you like the best.

Dr Heneghan: That is not the case. In America, they have a thing called the Sunshine Act, which means that the companies have to post exactly how much they may or may not be paying the individual, so you will be able to look up and see if Dr Carl Heneghan has received any money from one of these companies or manufacturers. That does not occur in the UK. Again, that is a great transparency issue. A lot of it is about being able to trust your clinicians to make good decisions and them not having conflicts of interest, and having access to the information to make the best decisions-that is what we are here about. At the moment, that clearly is not happening and most problems arise because it is still going on and there is a potential for conflicts of interest to occur. I, and you, have no way of knowing who has such conflicts of interest, but I can tell you that I have not been paid any money by anybody.

Dr Woodward: May I talk about the transparency issue in relation to the work that we have done, which might give you some clues as to how to think about taking things forward?

We were very nervous in the early days of making our data as open as it could be. We collect 1 million incidents a year-around 3,000 a day. That, to a member of the public or a patient, could be a huge cause of concern-3,000 things are going wrong in the NHS every day. If you put it in the right context and look at the levels of harm associated with those incidents, only a very small minority lead to severe harm or death, but you also need to be open about that. That creates a feeling of trust between the public, the patient and the data that you have and, therefore, the system in itself. What you have to do is to back that up to show that the system wants to learn from that information, so we don’t just collect it and stick it out in statistical form; we collect it, send it out and say, "This is what we are doing about it." I think that creates a different relationship.

That also leads to a much better informed patient. For example, with surgery we have something called a surgical checklist. When patients go through the system, lots of things are checked off as they go through. Many doctors were very reluctant to use that, surgeons in particular, because they felt that undermined their clinical expertise. "I know what I’m doing. Why do I need a checklist?" But, actually, it empowers surgeons, the teams and the patients. They then use that to go through the system, feeling as if they are part of that system and not just being done to. The transparency of the data is incredibly important in creating a different relationship between patients and the health service.

Q15 Gareth Johnson: May I follow up on that? In your particular role, if we had a greater degree of transparency in the system, would any data protection issues for patients or patient confidentiality issues arise because of having a far more transparent system?

Dr Woodward: We have to adhere to all of those things, obviously, and the things that we report go through a certain level of filter so that individual patients would not recognise themselves in something that was made public and so on unless they had consented to it. You would not issue incident data that was very much related, say, to one speciality that was dealt with in one hospital that probably only 10 patients had experienced over the last year, because that has the potential to identify those people. We are very careful about data protection and confidentiality. However, in presenting this data, you can be very creative in explaining the statistical data that you have in a way that does not mean that patients say, "That was me." They could say, "That kind of scenario sort of happened to me", but not, "That was me."

Q16 Graham Stringer: There are obviously exceptions, but there are many different standards among the regulatory bodies in southern and eastern Europe compared with northern Europe. Is there any sense that manufacturers can cherry-pick where they go for authorisation for devices?

Dr Heneghan: Is there any hard evidence? No. Is there anecdotal evidence-for instance, when you look at the minutes of meetings of the FDA or the MHRA-and they bring that out and say that there is an aspect that is a cultural issue that is known about? Some exists in the minutes of MHRA meetings, but nobody has any hard evidence of that at the moment.

Professor Westaby: There are very definitely tendencies to direct your research to areas where you are likely to get ethics permission early for certain things.

I have a lot of trainees in Greece in units. I support heart failure units there that are starting to use artificial hearts. I can do far more in Greece that is wholly ethical much faster than I can in the UK. As an example, I am using miniature blood pumps instead of cardiac transplantation and using stem-cell therapy to make the native heart better. Now, I can get ethics approval for that in Greece, where it would take me two years in Oxford. All clinicians do this; there are American companies that will go straight to South Africa to test devices first. It is very easy to selectively go to areas where it is easier to get ethics committee approval in order to make your advances more quickly. That is not a lack of integrity; it is a very practical aspect of getting medical devices and new techniques accepted. You have to target your patient population and get ethics committee approval for doing your clinical trials; and, naturally, companies will go where they can do that quickest.

Dr Heneghan: We did some work with the BMJ on metal-to-metal hips, and one of the notified bodies in this country is BSI, which put the kitemarks on. We got into trouble because we did not realise this. We said that BSI must have looked at it and given it their stamp. It was interesting that they came back and said, "We don’t even look at the product, so you got it wrong," whereas it will look at prams or toasters and put its BSI stamp on them. We made a mistake, and we held our hands up to it when they said that. That gives you a whole host of loopholes, if you don’t even see the product. It is a completely odd system that allows all these things to happen.

Q17 Graham Stringer: If it can be used to make the treatment or the medicine more efficacious, can it also be used to get round the system so that less good equipment can get into the market quicker? Do you know what the balance between those two things is? It is probably an impossible question to answer.

Dr Heneghan: You would need to have access to look at that. It is simple.

Q18 Graham Stringer: Going back to the metal-on-metal hips, was that a regulatory failure, inasmuch as they should never have been allowed on to the market, or was it a regulatory success, in that they were found to be dangerous in some way and were withdrawn from the market?

Dr Heneghan: That is two questions. There was a bit of both there.

Dr Joyce: It is a failure. We speak to many thousands of people, and these lives have been absolutely ruined. I would like everyone to go away with that reality.

We have been here before. With the 3M Capital hip in the 1990s there were questions in Parliament, and the Royal College of Surgeons set up a committee, which said that there is no such thing as a small change. So we come back to the idea of substantial equivalence. I have this amazing sense of déjà vu. That was approximately 5,000 patients in the UK and Ireland. When the DePuy ASR hip was withdrawn in 2010, there were nearly 100,000 worldwide. We do not know the exact number of this group of devices, which are largely metal-on-metal hips, but somewhere between 500,000 and 1 million people worldwide have been fitted with these devices.

Dr Heneghan: It was a success of the registries, but it was a success of the Australian joint registry in 2007. We have a system, but there are problems with our registries and our regulators. No. 1 is that they are not independent enough, potentially. No. 2 is that they are not taking heed, and I don’t know whether they don’t have the skills or the framework to say, "We could have known this in 2007." Interestingly, the Australians had a very comprehensive joint registry in place, and Tom would know more about this. It was telling the messages then but they were ignored. There is plenty of information that I could send to the Committee to show that they were systematically ignored because the make-up of some of these bodies, including our National Joint Registry, was not independent.

Professor Westaby: The important point here is that the public get access to a lot of medical science through the media, and they get hope related to medical advances. If you are dying of heart failure or cancer and something appears that you can see will clearly help you with your last couple of years of life, you want access to that treatment. With long-term medical implants such as hips and blood pumps-artificial lungs are on the way-and all sorts of exciting technology, you would have to embargo widespread use for something like 10 to 15 years before you got your outcome data. When the BBC and Fergus Walsh say that stem cells and blood pumps are the future for heart failure and so on, and it goes out on the BBC and Sky, the patients are on top of you, and you simply cannot say, "We have to wait 10 years for long-term trial data."

We are up in the air with this, and there is no good answer. Registry data, as I intimated earlier, is important. Surveillance is important, and MHRA does it well. You have to take a leap of faith with devices. You have to look at them carefully when they appear, but you have to give the public access to the technology, and you then have to watch it with registries and regular surveillance. That is the way that you will pick up problems, but you cannot deny technology to the patients.

Dr Woodward: One point that you made was that regulation was a success because it picked it up. One of the crucial things is the speed of picking it up and the speed of response. If you take the aviation industry as the gold standard and places such as Toyota, they have a system whereby, if something goes wrong with one airline or even one flight, it is known throughout the system within about 12 hours. That is what we need for the NHS-a system whereby, if you pick up really quickly that something is going wrong, rather than wait for months and months, you need to be able to spread it around really quickly. Toyota has a system called stopping the line. The moment you know that something is not quite right, you stop the line. What we would want to happen when somebody picks up that things are going wrong is that you stop any future surgery or implanting or whatever, until you know that it is safe to start again.

Q19 Graham Stringer: Collectively, your evidence is very interesting and getting increasingly scary. Who should be responsible for post-market surveillance? We have had some evidence that clinicians do not report failures and difficulties as quickly as they should. What are the barriers to clinicians reporting problems?

Dr Woodward: We have studied this to a great extent, because our work is not wholly dependent but largely dependent on having a very good reporting system. If you know about the stuff, you can actually start doing something about it. So we work very hard to identify those barriers and to address them. There are a large number of them; I shall give you some but not necessarily in the right order, as it were.

There is the lack of feedback. If you report something and you hear nothing back, you say, "What was the point in me doing that?" There is the issue of immediacy. If something is going wrong, you deal with what is going wrong. You deal with the patient, the family and the issues. By the time you have done all that and ended your shift, you will want to go off and do not necessarily come back and fill out an incident report. Incident reporting systems are seen as highly bureaucratic. Performance management systems make you end up feeling that you as a person or your colleague are being blamed, rather than the organisation, the group or team that you are in saying, "This has obviously gone wrong. Let’s see if we can learn from it." We need a culture of learning rather than punishment.

All those issues to do with time, bureaucracy, immediacy of action and the lack of feedback lead to not reporting. There are a number of crucial things that you can do, which are obviously the opposite of all of those, to encourage people to report-make it feel worthwhile, give them lots of feedback, make it a learning system and make them feel supported when things have gone wrong. There are some fantastic lines about "The best people make the worst mistakes". People don’t go into the health service wanting to make it go wrong; they go in wanting to do their absolute best. They are traumatised themselves when things go wrong, and they want to learn just as much as the system wants to learn. You need to create an environment within which they can do that.

Q20 Graham Stringer: What is the proposed market surveillance system?

Dr Heneghan: There are a couple of issues. First, we need to ensure that manufacturers do this, and this is coming back to the PIP implant, which is quite an interesting one. In America, if you make a change even to the industrial grade silicone, you have to tell the FDA-you don’t have to do that in Europe, by the way; we can change products in Europe. When they asked manufacturers to look into all breast implants at the time, they lost 80% of the follow-up because it is quite technically challenging to run registries. I do think there is a role, and No. 1 is to think about having a national system for all implants and it becomes mandatory to put in anybody who has a joint put in. There are systems coming on board that will allow these devices to be trackable. That is simple; it is like a barcode system. We can be world leaders in this with the NHS and we can ensure that everybody gets into the system.

The flipside of what we are talking about from the National Patient Safety Agency viewpoint is the second thing. For drugs, there is a very valuable system called the Yellow Card system, which is now embedded even within the electronic systems of primary care. If I see a drug that is giving you an adverse effect, I can send that off. That has been going for about 50 years, but there is no equivalent system for medical devices. It is quite absurd; it is like a website, if you can find it: it is so hidden in the back of the MHRA that you have to really search for it.

Have we got a scary problem? In the five years that we have looked at this, there has been a 1,220% increase in the number of field safety notices. That is the number of manufacturers saying that problems are occurring. We are trying to help you, I hope, with a system. At the moment it is continuing to rise and be problematic. How do we get a system that is innovative but is also safe for patients? It can be achieved, but we have to start using some of these systems to help.

Professor Westaby: I have done over 11,000 open heart operations now, and I have used more than 2,000 heart valves, hundreds of vascular grafts and many types of artificial heart. I would like to give you the confidence that primary device failure is extremely rare. It is very, very, very rare in the NHS. If you hear that a device has failed, there are several reasons why that might happen. Devices can wear out earlier than expected because of patient factors. A lot of devices go down because of poor patient compliance. Heart valves and blood pumps need anticoagulation, and if the patients don’t take their pills the devices can go down with clotting and that sort of stuff. That is not because the device does not work but because the patient has not been compliant. The last thing is that inappropriate utilisation of devices can account for alleged device failures. However, I would go right back to the original statement. Primary device failure in the medical profession is, in my view, extremely rare, and you should have enormous confidence in the way things are.

My biggest issue in this country is access to life-saving technology. I won’t go into it, but the NHS rations life-saving technology in a way that many, many more lives are lost through that than through devices going down.

Q21 Pamela Nash: A few people who have submitted evidence to the inquiry suggested that it might be helpful to expand the National Joint Registry to include all implants, although Dr Heneghan has indicated that its independence is compromised at the moment. Would an expanded registry-or an equivalent, if you would like to suggest one-be helpful?

Dr Heneghan: That is a great point. The first thing is, how many hips failed?

Dr Joyce: It depends on the data you look at, but it was about one in two DePuy ASRs or about 50%.

Dr Heneghan: How many was that in absolute numbers?

Dr Joyce: Somewhere up to 100,000.

Dr Heneghan: I am sorry, but I don’t see how that can be rare. This is one where I could give you a number of examples. Medtronic was mentioned earlier; its bone cement was used off-label. That has been all over the world, and they are looking at that. There are a number of devices, including breast implants, so it is not rare.

With the joint registry, there are two things we want to do with devices. No. 1 is to work out whether it is cost-effective and does it make a difference or at least whether it is equivalent to what we already know. You can do that with benchmarks as well, saying that after three years you should expect to see things in the context of a trial. The second thing is that, when you put it on the market, a registry is set up to be a quality assurance system. You don’t know, but in 10 years cancers could appear, or in 15 years devices may fail catastrophically. It should be a quality assurance system; it should not be designed to be about causation-"Oh, it went really well." The argument now is that we have only had device registries since 2003, so we don’t know enough about the long-term consequences. We should expect devices to fail, but we should also see them as being complementary to sets. There is a three-year benchmark for hips with NICE to say that you should get 3% revision rates at least and if you are above this you are out of range.

Q22 Pamela Nash: Do you think that this can be made from the existing NJR or do you think that we need to rub it out and start again?

Dr Heneghan: No, I think it can be within the existing NJR, but you need to think about who is going to run that, and its nature, structure and independence. You should think about how it ensures quality and that it provides information really quickly when it finds out what is going on-not that it delays information. We could sit in a room and think, "Well, let’s wait a year," but that is potentially damaging.

Q23 Pamela Nash: The key point of my question is whether we should expand this to include all medical implants and not just joints. Is that something that would be helpful?

Professor Westaby: There are thousands and thousands of medical devices that impact directly on the body, and many hundreds of implantable things, if you include fillings in the teeth and so on. There is a big difference between one device failing lots of times and a large proportion of devices failing. I will go back and say that the vast majority of devices have been proven to be safe.

All devices will wear out sooner or later. Clearly the hip issue is very serious, because it needs revision. If a blood pump packs up, the patient can die. One of my patients died when his blood pump stopped, but it was because he had not taken a spare battery out. It is as simple as that; it is about compliance. When implants fail, there is a large amount of emotive stuff around it, but I personally have confidence in what I said previously and that surveillance right now by MHRA is good. What you have to do is to persuade the clinicians to provide you with the data.

Q24 Pamela Nash: You also said that you did not think that it had the resources to do the job properly. Do you think that we now need to set up a registry of all implants that are used within the UK?

Professor Westaby: Registries are very good, and it is worth all class III implants having registries. Many companies have their own registries. The valve companies and the pacemaker companies want feedback. They are very keen to know if any of their devices go wrong. The current level of integrity in the device market in my view is very high and the regulation of the device companies in the States is very good, and-let’s face it-most of the devices that we use come from the States.

Q25 Pamela Nash: The fact is that most of that information is not centralised anywhere if it is registered by the company.

Professor Westaby: Yes.

Pamela Nash: And the next part of my question is whether we should be publishing data from the registry as it stands at the moment and if we were to expand it, which we cannot do if we have different registries.

Professor Westaby: Cardiac surgeons a long time ago set up a heart valve registry, a registry for mechanical assist devices and so on and so forth. You will find that the medical profession in many areas sets up its own registry, and this is happening more and more often. We feed information to the Department of Health and MHRA, and, if all branches of the profession did the same as the cardiac surgeons, then you would have the information that you want.

Dr Heneghan: You can take a perspective that is cynical or of being on the sidelines, but let us go back to the analogy of Toyota. If you own a Toyota car and there is a brake failure in one car, they know how to alert and get hold of everybody and draw them back in. If you have a failure in your joint, or in your x, y or z, we have no system in place for saying, "Let’s use this to alert the 8,000 people who now have the system to make them aware." That does not happen, and that is absurd. Your car is in a much safer position than the stuff that we put inside your body. That is what we are talking about.

Q26 Pamela Nash: Given my understanding of this, which you are expanding rapidly, would that not be a function of MHRA rather than a registry?

Dr Heneghan: No. The MHRA’s job is just to respond to manufacturer alerts and/or clinicians who alert them to x, y or z. Given how it is set up and its role in the system, it is in a bind. It can only impact on what it can do. All that it is saying is that it responds. It could take a more neutral viewpoint and say it could have more clinical data and so on, but it says it only responds. You could argue, in the case of the hips, that it could have responded earlier and taken a more uncertain viewpoint as opposed to this certainty viewpoint.

Dr Woodward: The MHRA sends out some alerts that alert the service to certain concerns on the use of devices or medicines. There is something about the timeliness of that which could definitely be speeded up. Actually, it is similar to the system that we have when we produce things called patient safety alerts or rapid response reports, as they are currently known. Lots of national bodies can send out as many alerts as they like and as much guidance as they like, but the crucial thing is implementation and compliance. There is a big gap between what we tell them should be happening and what they actually do. The key question is how we manage that bit. Is that the role of regulation, or is it the role of the local provider to police or monitor that, and how do we monitor it?

We currently have a system called the central alerting system, or CAS for short, and people self-report their compliance with an alert. If we send an alert in January 2012, we expect it to be implemented in full-we usually give them a year-by January 2013. Over a certain period we expect them to report back to say, "Nearly compliant", "Nearly, nearly compliant", and "Compliant". That is self-reported, and you can only spot check some organisations to see whether they truly are compliant or not, and we expect the Care Quality Commission in our respect to do that.

At the moment, we have 99% compliance. Lots of people say that that is fantastic, but I disagree. I think that 1% non-compliance is really poor. You would not expect to go on an aeroplane or take a train journey and think that 1% of the things that they are supposed to be doing have not been implemented. You would be scared from point A to point B on your journey. Why should it not be the same in the NHS? If you are issuing an alert and it is because there is a problem-you have really good evidence that there is a problem and you can tell people that they should be doing something about it-people should be compliant with that, and that is a big issue.

Dr Heneghan: There is a real problem with timeliness here. The MHRA committee on safety of devices said in 2006 that there is growing concern over the biological risk of metal wear debris, yet it was March 2012 before we came to a decision. Going back to your other industries, I cannot imagine getting on a plane and saying, "We have a problem with wear debris." How does that get communicated? People need to say, "Hey, we need to draw back now. We need to stop and think about patient safety." There is an issue here that we should not be sitting here doing this now; we should have been sitting here four years ago.

Q27 Pamela Nash: Before we move on, Dr Joyce, you mentioned in your evidence that you thought it would be helpful to conserve explanted joints. First, would you address the earlier points I made? Do you think that we should have a registry of all medical implants, and should it be published? Secondly, on the final point, if you are able to give us any examples of when that has been of advantage it would be really helpful.

Dr Joyce: I think that the registry should be expanded to all artificial joints. On implants, I am not sure because it depends on the implants. One thing that we need to point out, which was alluded to earlier, is that, if you are going to have a heart valve put into a patient, they are probably in a life-threatening situation. That might be one situation, but with a hip with osteoarthritis it is not the same.

The other thing is that post-market is crucial. It is certainly my experience, in the case of the DePuy ASR, that there was no need for one of them to go into a patient. That could have been identified by pre-market testing. They had the machines and the equipment, and those tests could have been done.

To answer your other question on explants, again I mention Toyota. If there is a problem, what do you do? You take the thing to pieces and look at it. That is all that I am suggesting. Similarly, with plane or train crashes, you look at the bits and pieces afterwards. At the moment, it is just not done with hips or knees. Most of them are just thrown away. To be honest, without the work that we did I am not sure whether ASR might still be on the market now. We want an independent centre examining it and then sharing that information, saying, "We think there is a problem," and having scientific publications produced which share this information.

Q28 Hywel Williams: We have already talked about public perception and public understanding of this entire issue and how important that might or might not be. We are being broadcast this morning, so there are at least a few people listening to your evidence. How important is it that the public understand the role of the MHRA, notified bodies and manufacturers in implant regulation?

Dr Heneghan: If you go for an implant, you are consented. That is the first thing; you are consented to the benefits and the harm. I wrote this contentious thing, and Stephen is at the university of Oxford, but around the world it is the same issue. You want to say, wherever you are, that we have access to the information that this is beneficial or this is the harm that you face. It is really important that the MHRA and our joint registries promote what we are trying to do for patients-improve safety.

Does that happen at the moment? No. We had a recent thing where these patients get up and tell you what is going on in their lives, and they question us. They come to me as a GP and say, "Why did somebody put this in me when they already knew this? Why did we not have some fail-safes? We were doing this a year ago, six months ago, so why have I now got this long-term risk that I don’t understand?" We have to get a perception of how to communicate directly with patients. The flipside from the NPSA is that we have to get a system where they report in more. That would be wonderful. The patients could start saying, "I’ve got a problem with my hip. This is what I want to do. Where do I go and report that?" That would be amazing.

Dr Woodward: I am not sure as a patient that they really want to know about the role of the MHRA or what the NPSA or NICE do. They just want to have their problems addressed and to be pain-free and cured if at all possible, or helped and so on. It is usually when things have gone wrong that they start to try to find people in the system to help them figure that out.

On the issue of consent, a lot of people compare the system in the US with that in the UK. In the US you are told about absolutely every single possible thing that can go wrong in major, major detail. In the UK we are veering towards that, but what you have got to do is to explain that in a way that is genuinely understood, because risk is a really hard thing to understand. If you say that you have a one in 10 chance of this happening to you, what does that really mean? Do you think that you are the one, or do you think that you are one of the nine? If you happen to be the one, will you think that the nine people behind are okay? I know that I am being simplistic, but it is a really hard concept to get across to patients.

The other aspect is choice. If you are in a situation where you need a lot of help and support, you are in a lot of pain and you are very vulnerable, Dr X or nurse Y will say, "You need to have this. Please consent." It is a bit like looking at the terms and conditions to some of the things that you sign up to on the internet. You just tick "I agree" and get on with it. It is only when it has gone wrong that you start to rationalise why you said yes to this.

Professor Westaby: After 40 years in the medical profession and at least 35 at the sharp end, I can tell you that very few patients actively take an interest in their treatment. They want to know that they can trust their doctor and that the doctor will give them the best advice. I am sure that this resounds with you.

No one has heard of MHRA or at least until the recent issues. There are many bodies involved in safety now, and it is bewildering for the patients. But, because of the internet, let us say 5% of patients are avidly interested in what is going to happen to them. They tend to be the academics and people who are obsessed with healthcare. For the sake of those 5% of patients, I think that we are absolutely going to have to make information like this available on the internet. We are going to have to publish registries for different types of implant. We are going to have to publish mortality rates for different procedures. It is already happening in cardiac surgery. Many surgeons are having to drop out because their mortality rates are already perceived to be too big.

I shall shut up now, Mr Chairman, because I know that you don’t want to divert into another area.

Q29 Chair: Having listened to Dr Woodward’s last response, I think that some of your colleagues are going to have to learn how to communicate to patients in English. Issues of probability are not easy concepts to put across to patients, especially when they are ill.

Professor Westaby: Absolutely.

Chair: There is a serious challenge to the medical profession about how to do that better.

Professor Westaby: Absolutely.

Dr Joyce: We sit down with patient groups and let them talk together. We have focus groups, and they produce minutes like this; I shall pass these round. They say things like, "Why does the GP ignore tests?", "My GP does not understand the information data", "It feels like nobody cares about us", "When you stop trusting, it can have a big impact", "We move from place to place; you are in so much pain you will go anywhere." From the MHRA there is a lack of communication. They have failed; they shared responsibility. There are massive issues for the patients that we talk to about knowing what has happened and why it has happened, and they do not want it to happen again. It is an incredibly difficult place.

Dr Heneghan: May I come in on this issue? In my experience as a GP-perhaps this is a reflective-I think that people do want to know what is going on. Increasingly with the internet age, they are becoming more sophisticated and more knowledgeable. Boy, every time they walk in I say to them, "What have you looked up? What do you think is going on?", and they know what is going on. People do want to be informed and they do want to know. They take an interest in what we do, and we should particularly explain the decisions that we make better, why we do not have access to certain treatments and why we need to do this. We do it really poorly, and we need to take a perspective on that.

Q30 Hywel Williams: How could notified bodies and manufacturers be more publicly accountable for their role in certifying products? Should the work of the notified bodies be conducted by public organisations such as MHRA?

Dr Heneghan: You are in a bind there, and the Americans find this. If you try to do it with a public organisation, you will need a massive organisation and it will become very costly very quickly. Whichever way, the manufacturer will have to pay for it. The US FDA does this. The solution is to make it transparent. It is to say, if it is a notified body, we are in the UK and we need the information to be made available. Let’s put it online; let’s look at the design job. They will argue that there is commercial confidentiality in those documents, but the clinical data should never be commercially confidential. There may be design aspects that they cannot give, and I am happy with that, but let us publish the clinical data. If you did that tomorrow, I would be on the case working out what is going on. I have been trying to do it for three years, and I cannot do it.

Dr Joyce: I agree with that need for transparency. With artificial hips or knees, they should tell us how they have been tested and what the results are, and share it with everyone. If there is nothing to hide, why not?

Dr Heneghan: There is another route as well. You don’t have to be totally anti-innovative here. There is a system that the FDA is adopting, which I have written about, called the IDEAL system. The idea is that, when you are very innovative, you are in an investigational device exemption zone. You could apply to the MHRA, saying that you are going to do this in the conduct of trials, that all people are going to be followed up and that for the next three years we are going to consent people as though they were in a trial. Actually, we should promote more use of that-of how these things get recorded and registered and how we understand it-so that more innovative people can look at it. We do not want to completely shut the gate down, but we can use that system to promote innovation at the same time.

Q31 Hywel Williams: I move on to my final question. Who should bear the costs of corrective surgery and public health messages when an implant is found to be unsafe? For example, with PIP we have a difference between Wales and England.

Dr Heneghan: If you wanted to fund independent registries, they are very cheap. When I was in Australia, it worked out at about 20 Australian dollars per patient entered; that is one. The first is registries from a manufacturer.

The second is what is wrong with an insurance-based system? Here is another analogy. If your plane crashes or your plane company goes bust and you are in Spain, they have to get you all back. They will fly you back. What is wrong with having an insurance-based system equivalent to the level of evidence? It should go down as you promote it. If the device has 20 years and not many problems, you won’t need much insurance; but in the early days it is going to cost a lot of money for some of these devices.

Q32 Stephen Mosley: Dr Woodward, you talked about risk in your previous answer. When we were talking to the MHRA, it gave the quite simplistic case of bed rails, saying that patients had been trapped by a bed rail and injured. They looked at it and thought that it was bad news, but when they actually looked at the data they found that you were at much greater risk of falling out of bed and injuring yourself in that way than you were from a bed rail. You therefore have to take a risk-based approach.

You can see, with some of these implants, that if someone is going for a hip replacement they need the replacement, but if problems occur two or three years down the line where does the ratio of risk apply?

Are you trying to suggest, between you, that if you have a situation you should be upfront and honest about it and say, "Look, if you have this operation there is a 50% chance that it might go wrong in a few years"? Are you saying that that is okay if you are upfront and honest about it so that professionals such as yourselves can make the right decision, or are you saying that that should never happen and that you should never have a situation where someone suffers because of a problem?

Dr Heneghan: Let me clear about it so that you understand. There are three classes of device-class I, class II and class III. The bed rails are a class I device and you get a CE marking, rather like you get for a teddy. It is a bed rail, it is to standard and you hold that documentation. Yes, they will fail, but that is okay. Actually, there is a system in place that probably deals better with class I devices because there will be slight problems-hoists fail and all these sorts of things. That will happen regularly. For that, the current system is okay because they are low-risk devices, and you hope that somebody would put up a better system for patients and the public, probably like the Yellow Card system, so that you can communicate that better.

However, let me be clear that when you are talking about class III the capacity to do harm to a number of patients is so great. It is like with a drug; if there is a problem with a drug tomorrow, we can stop you taking it. You can have hundreds of thousands of patients with a hip inside them, who are saying, "What do we do? Can you take it out tomorrow?" You can get rid of the bed rail tomorrow; you can change it; but you cannot change something that is inside somebody. So you want to know really quickly what is going on if it is going badly.

Dr Joyce: On the point about being upfront, most hips are excellent. It is just some of the modern designs that have failed. However, when we talk to patients, they tell us that when they went to see the surgeon, he said, "This will last 30 years. You will be horse riding and skiing." The patient will say, "I have never skied in my life or ridden a horse." The patient is left in a situation where they are expecting some brilliant device that will last for 30 years, and then suddenly they have pain and they are ignored. Eventually, they find out that there is a major problem with the hip. They are not alone. Then there are the long-term concerns about the potential cancer risk, with these metal particles travelling around the body, and we still don’t know the long-term implications of that. I think that no one was upfront about that with the patients that we speak to.

Dr Woodward: People underestimate the actual skill required to do a really good risk assessment. You obviously need to understand the risk of doing something versus the risk of not doing something. Then you need to show the risks in the short term, the long term and the medium term. You need evidence behind all of that if you are going to do it really well, so it is not about gut feelings or intuition but about history and evidence over time. That would cover your clinical patients and your clinical outcomes, the research, the audit, the patient safety and incident data and the case note review methods that people use. There are also fancy tools that you can use, such as failure modes and effects analysis and hazards analysis and all sorts. You need some skills in order to do that.

Once you have made the probability assessment, you need to use those skills to turn it into something incredibly simple in order to communicate it. It is not as simple as saying, "This could happen to you tomorrow and you need to be aware of that." It is a balance of probabilities, as they always talk about, and that is a really hard thing for clinicians to do and a very hard thing for patients to take on board. Then we have to weigh all that as a matter of judgment, asking ourselves whether or not we do it and what are the options.

Dr Heneghan: In the current situation, it is impossible to do that because nobody has any data on the effectiveness of devices. In lots of situations you will say, "We will tell you how it should perform at some point in the future." It does not matter how much we discuss the ins and outs; the key for lots of devices is that we cannot generate the information to inform patients in the first place-until perhaps at some point in the future.

Q33 Stephen Mosley: Do the MHRA and the other notifiable bodies-the regulators-have the expertise and the skills to do that?

Dr Heneghan: It is interesting when you look at the make-up. In terms of organising a regulatory body and what it should look like, the answer is yes. In terms of how they look at the data and how they communicate that data, it is no. The Americans have realised this and set up a network to support them from epidemiology and statistics-based centres and made them independent.

The key is that you want people who will look at the data and be able to unblind it and produce that data in an independent and transparent fashion, in a meaningful way. We already do this with drugs. In clinical trials you have stopping rules that are built by smart people, and if things are going wrong they can unblind it and say that it is not working. There is a nature problem in the make-up there. It is the same with the joint registries. You have to think about who you want, particularly for understanding the data and the independence.

Professor Westaby: I sense that we are getting down the line with this. Right now, I would make one caution. It is that the NHS is becoming so stiff in bureaucracy that it can hardly move. If we are not very careful, it will just grind to a halt.

I will tell you frankly that the amount of scrutiny that was levelled at heart surgeons after Bristol has put people off in this country from going into the profession. We have to do the right thing for the patient, but virtually the whole conversation this morning has revolved around one breast implant, one hip and one valve that was not really faulty but which was not used properly. We would take a long time to make a list of 20 devices that had gone wrong in this country. I would reiterate that, if we are not careful, we will make the whole system so bureaucratic that it will not move at all.

I shall finish by saying that, when we contacted MHRA about the anxieties about that one heart valve that was not being washed properly, an alert went out about it within one week to every single cardiac unit. In two weeks the MHRA knew everything about how many valves had been used and in what centres, and everything else. I thought that was a very fine response and they did the right thing. They cannot do that for absolutely everything because there are not enough people and, in my view, they are under-resourced-like the whole of the NHS. You will not be surprised to hear that several eminent members of MHRA are going to leave imminently. Quite simply, you can only put so much attention and flak in one direction before people do not want to be involved.

Dr Heneghan: The key here is combining innovation with patient safety. That is what we want. We want to be global leaders in industry and to bring work here, but when you look at where all the companies are located most of the device companies are located in America; most of the drug trials are done in America; and two thirds of the regulatory trials are done in America. They have a tougher, tighter regulatory framework and they are about to make it tighter, which will incentivise their companies to build higher-quality products. In the end, you will acquire quality and you will win. In the short term there will be lower-quality products; yes, you will have lots of little industries making a lot of little bits of devices here and there, but we want to be the Formula 1 of the industry. That is what America has realised. So the argument does not wash when people go, "We’ll not have a business if we become more regulatory." American has already done it, and it is winning.

Q34 Stephen Mosley: Turning to the PIP breast implants, I think it was you, Professor Westaby, who said that it was not a problem with the regulations as such. It was more that the application was in effect fraudulent. They were claiming that there were different things in them, basically. Should the regulators just be looking at the data that manufacturers supply, or should they be looking to see whether they are doing as they say? Should they be looking beyond the data? I know that you can do that in the long term if you have a registry and look at how effective things are, but when someone comes upfront to get approval, should you just be looking at what they are telling you or should you be trying to get behind it?

Dr Heneghan: There is a perspective here. You may say for very high-risk devices that the MHRA should at least take a sample of them and say, "Let’s audit a sample of these"-perhaps 10% or 20%. You have to start somewhere. We cannot suddenly say tomorrow that we are going to put a load of work on here; you will need to increase the work load. You would start with some perspective and say, "Let’s start towards it with a sample and then move forward." If you find problems you will have to expand that, but if you don’t find problems you can carry on with the audit. That is exactly what the FDA is doing. The FDA even go out and visit manufacturing plants-surprise, surprise. Do we do that in Europe? No. Will we be able to effect the EU regulatory framework to do anything? Probably not, but we can effect our regulator.

Professor Westaby: You would have to disassemble and chemically analyse every implant that was presented to you. The number of independent experts needed to do that, even before you started the clinical assessment of a device, would be massive. You are pushing us towards a situation where the whole thing will be so stiff with bureaucracy that it will not move. On cardiac transplantation, I can tell you that if we applied the stringent criteria that we are suggesting for devices to donor organs and donor hearts you can forget it. That would be transplantation gone right now.

Q35 Chair: Let me assure you, Professor Westaby, that we are not pushing you anywhere. We are simply seeking evidence.

Professor Westaby: No; I know that.

Dr Heneghan: May I come in here? In the USA, Steve Nissen looked at the American system and the number of devices that had failed. For cardiovascular devices, there were 76; that was the highest, so there is a lot going on. What Steve Nissen found, which was a really interesting issue for the FDA, was that 90% of them had gone through the 510(k) route. Of the ones that had gone through the pre-market approval process, very few failed. Some manufacturers are incentivised to produce evidence in clinical trials, but those who build a wonderful piece of kit have not got a system in place to stop somebody six months later using a 510(k) equivalent to bring the same piece of kit to the market. With global expansion, the number of equivalent devices is going to get worse.

Q36 Stephen Mosley: How important is the MHRA’s committee on the safety of devices? How effective is it at giving expert opinion? Should it operate more transparently that it currently does?

Dr Heneghan: I have been looking at this for three years, so I know quite a lot about it. First, there is an issue about declaring and thinking through the conflict of interest issue, what it should be, what the make-up of that should be and what is the best perspective to do that. At the moment you have a mixture of people from industry and non-industry on it. At the end of the day there will be conflicts of interest. I have conflicts of interest; everybody does.

The second issue is that we should start to expand it, because it is not big enough to deal with the numbers, the size and the amount of work load going on. You have just one device safety committee meeting every few months, and it is not big enough. The complexity of the issues is important and difficult, and I don’t think it is currently equipped for that role.

Q37 Chair: I have one final question. One of the things that is happening in the world of medicine is this extraordinary convergence of technologies-for instance, the way in which engineering is now so fundamental to medicine. Are we heading towards a regulatory problem in some respects? For example, Professor Westaby, some of the stents that some of your colleagues put in will release chemicals.

Professor Westaby: Yes.

Chair: And there are other devices that are designed to release medication into the patient. Is a point arriving when the convergence of pharmaceutical and engineering technologies and surgical skills will require a different approach to the regulatory structure?

Professor Westaby: It is there already. For instance, I use plastic tubing that is coated with heparin. I am working on a total artificial heart with a professor of bioengineering, who is going to line the device with stem cells so that the propensity for clotting within the device is that much less. Yes, the complexity of artificial organs is moving on at a great pace.

One very interesting thing that has not been mentioned is cost containment in devices. Every one of the small artificial hearts that we can use instead of transplants these days costs the same as a Porsche car. It is a tiny thing, so why does it cost the same as a Porsche? It is not that it is more complex. It is because of all the regulatory issues and risks associated with an implantable device. To get around that and to make this sort of thing accessible to British patients on the NHS, we put together a company to make a far less expensive device that is probably even better, but it is here in the UK. We are doing that, and there is a lot of interesting work going on in the UK that will combine biology with mechanical technology. It is becoming very interesting, and it is going to be very good for patients. For enormous groups of patients, like heart failure patients, there is an enormous imperative to get this technology available quickly. If we make regulation more bureaucratic, there will be another whole generation that we cannot treat in the way we should, according to current medical science. That is why I am kind of worried about anything that will put the brakes on more than we have already.

Dr Heneghan: That is a very good question. What you have described is a device with a drug in-heparin or stem cells. That means that it has to go through the drug regulatory pathway and through the usual clinical trials. It will require two randomised trials to be implemented, and it will then be submitted to NICE guidance. Take the heparin out of the stent and you don’t need any of that-you can use equivalence to bring it to the market.

The key of what you allude to is whether we are going to look back in time and think that it is ridiculous. Yes; it is a bit like 50 years ago with drugs, when we had a situation with Thalidomide, and look what happened then. Under the current system, if we don’t do something different, at some point a device will come on the market that will cause something catastrophic that will make people stand up. Currently, we have not got that yet.

Dr Woodward: This may be a very simplistic answer, but my simple approach is this. As the system gets more complex there should still be the same principles, no matter how complex it becomes, or how innovative or technical it becomes, or how much more technology there is. The simple principles are: is it safe, is it effective, does it meet patient needs, does it pose the least risk and all those things? There are some key principles that will apply no matter how complex the health service becomes.

Chair: May I thank the panel for a very informative session this morning? If you have any additional thoughts, we would welcome a note from you because this is clearly an issue that is considerably more complicated than most people realise. Thank you very much indeed.

Prepared 8th June 2012