UNCORRECTED TRANSCRIPT OF ORAL EVIDENCE To be published as HC 151-i

House of COMMONS

MINUTES OF EVIDENCE

TAKEN BEFORE

HEALTH COMMITTEE

PATIENT SAFETY

Thursday 15 January 2009

PROFESSOR ANDY ADAMS, DR DAVID WHITAKER and PROFESSOR BRIAN TOFT

PROFESSOR MIKE MURPHY, PROFESSOR BRYONY DEAN FRANKLIN

and MR ROGER LAMB

Evidence heard in Public Questions 215 - 329

USE OF THE TRANSCRIPT

Oral Evidence

Taken before the Health Committee

on Thursday 15 January 2009

Members present

Mr Kevin Barron, in the Chair

Charlotte Atkins

Mr Peter Bone

Jim Dowd

Sandra Gidley

Dr Doug Naysmith

Mr Lee Scott

Dr Howard Stoate

Dr Richard Taylor

________________

Witnesses: Professor Andy Adam, President, Royal College of Radiologists, Dr David Whitaker, Immediate Past President, Association of Anaesthetists of Great Britain and Ireland, and Professor Brian Toft, Professor of Patient Safety, Coventry University, gave evidence.

Q215 Chairman: Good morning, gentlemen. Could I welcome you to our third evidence session on our inquiry into patient safety? I wonder if we could ask you, for the sake of the record, to give us your name and the position you currently hold.

Professor Toft: I am Brian Toft; I am Professor of Patient Safety at Coventry University.

Professor Adam: I am Andy Adam; I am Professor of Interventional Radiology and President of the Royal College of Radiologists.

Dr Whitaker: I am David Whitaker; I am a consultant anaesthetist at the Manchester Royal Infirmary; I am the Immediate Past President of the Association of Anaesthetists; I am Chairman of the European Board of Anaesthesiology Safety Committee and a member a member of the Safety Committee of the World Federation of Specialist Scientists and Anaesthesiologists.

Q216 Chairman: Thank you for that. A general question to all three of you, although some of the questions today will be specific to individuals and we can try and tie that down. We have got one or two pressing statements being made in the Chamber a bit later on, so we are hoping to run through this morning pretty quickly. Do you think that technology is a problem for patient safety or a solution - or both?

Professor Toft: Of course, as the committee will know only too well, it is a two-edged sword: it can be for the good, it can be for ill. For example, Professor Mike Murphy is going to speak later on this morning about the excellent work he has done at Oxford Radcliffe on bar-coding and blood transfusions. The stuff there has reduced errors phenomenally. On the other hand, we can see that people have a moral hazard when it comes to technology; that is to say, they rely on the technology in such a way as they actually make mistakes - they believe it to be safe. A good example of that would be in the United States of America in 2006 when six neonates were given 1,000 times too much heparin because the drug was pulled from an automatic computerised draw. They swiped the draw believing that they were taking out a weaker solution and, in fact, the person that stopped the draw just stopped it with far too high a concentration of heparin and that was given to the children and they did not check it; they assumed that the technology was okay. Recently, at St Mary's Hospital, at the back end of last year, there was a case where a lady had a gall bladder taken out that she should not have had done. It is quite interesting that that particular hospital was using the WHO Safe Surgery Checklist. That is new technology to the medical profession. So it failed on that occasion. That does not mean to say it is a bad thing; it just says that it failed on that occasion.

Q217 Chairman: Is this the one that was announced on the radio this morning?

Professor Toft: I could not say. I was travelling at half past five this morning, so I do not really know. St Mary's is the one where Lord Darzi is the Honorary Consultant. They were using that technology and it failed. The bar-coding technology used by Professor Mike Murphy may well have saved the day there. Besides getting things like that, you have also got things like work rounds. Nurses and doctors will engage in new technology and if it does not work in the way in which they expect it to work, then they will do work rounds. They will do things like take bar-codes and stick them to their jumpers and then put them onto people, which can then lead to people being wrongly identified, and there are papers for this; there have been studies to do with this. By the way, I should perhaps inform the committee that I only study failure. I am not interested in success; it is just the bad things in life I look at because they are far more interesting! In terms of technology more generally and trying to move technology forward, we then have improvements. A good example of failure occurring in an improvement in technology is at Papworth Hospital. Originally they had a system where they used a thing called a check digit in the symbiology, which is like the bar-code, which is intended to make a check of the number so it is accurate, so that the machine can actually say this is an accurate number, and they implemented a new system quite recently, in May last year, and this new system does not have the check digit in. The check digit is actually being mandated by the Department of Health saying that the GS1 technology which contains this should be used under these circumstances. Now we have a system being put into a hospital which neither complies with the Department of Health's own requirements on this thing and it has made it more unsafe. So you can see, from that perspective, technology in one sense is very good because it can improve matters and in another sense it can also create lots of problems. At that point I will stop, Chairman.

Professor Adam: Yes, I agree with Professor Toft that it can bring challenges in terms of training and patterns of practice that must be met if it is not going to increase risk, but I think in the Health Service in particular the balance between the advantages and disadvantages of new technology in terms of patient safety are overwhelmingly in favour of increasing patient safety. I think new technology in terms of treatment brings in minimally invasive methods of treatment that are much safer than traditional surgical techniques; new methods of diagnosis, which again tend to be less invasive and more informative, so in that sense they improve patient safety. If there is anything to be said in general terms about technology and patient safety, it is that unfortunately the National Health Service is slower than the health systems of most developed countries in adopting new technology, and this is why I am very pleased that the NHS Next Stage Review has made innovation one of the aims of the new review of the new system that is about to be introduced, and that is why I am also pleased that the Health Innovation Council has been set up by Lord Darzi, which is looking at the speed of adoption of new technology and how to encourage that in the National Health Service. I applaud that and I believe that should make a major contribution to improving patient safety.

Dr Whitaker: I agree with both my colleagues. I think it has a bigger part to play in solutions rather than hazards but, I think, to put the whole thing into context, in all high-risk industries about 75% of accidents are caused by human factors and I think you have to bear that in mind. You have had, I think, Martin Bromiley, who is one of the experts in that, to talk to the committee, and I think you have to bear that in mind. We do not know what the percentage is within medical accidents at the moment, but the MPSA Critical Incident Reporting System has now got a new question added which may give us the handle on that in the future. What we need to do is the right thing to most patients. We do the right thing to most patients most of the time, but what we really need to do is do all the right things to all of the patients all of the time, and technology can help in this. A safety culture is very important to get that 100% compliance, but technology can help in this way. A simple thing is the pin index system on oxygen cylinders which means you can only connect an oxygen cylinder to an oxygen yoke, and that has been established for many years. The association first discussed that at its AGM in 1932, and all the technology that anaesthetists use now for monitoring oxygen saturation, ECG, blood pressure makes a significant contribution to patient safety and medical indemnity insurance subscriptions have gone down in this country and have started to go down in other countries as well. Also oximetry technology, which I spoke of at the Associate Parliamentary Health Group last year on healthcare in Africa, has now been highly recommended now by the World Health Organisation (and that is the highest level of recommendation that they give) and it should now be used in every one of the 230 million operations that take place every year in the world. It is already in used in this country but it will obviously help in the poor resource countries. Also, the laryngeal mask, which is a world-beating British invention, which is a novel airway device invented by Professor Brain 20 years ago, has now been used on over 300 million patients worldwide, it has made a major contribution to safety in anaesthesia in difficult airway situations and also when resuscitating patients. Patients never see this, because they are unconscious when it is being used, and this remarkable technology deserves very widespread recognition. Technology can become part of the problem when it fails or it is over relied upon, as my colleagues have mentioned.

Q218 Chairman: A specific question now to you, Brian. You did the investigation into the death of Wayne Jowett?

Professor Toft: I did.

Q219 Chairman: What conclusions did you draw from the case and has the National Health Service acted on those conclusions?

Professor Toft: My conclusions regarding Wayne were that there had been procedure failures. There was inadvertent human error (that was certainly the case) but there was also systems failure. There were systems simply not in place that would have protected him, but people did not know that. It is unknown, if you will. One of the things was the failure of the technology. You can connect any Luer syringe to a spinal needle, which meant that vincristine could be put into his back. Once they got to that stage, that was it, finished. One of my recommendations was they should change the technology and design a spinal needle to which a Luer syringe could not be connected. That seemed to me to be a reasonable way to progress. Since that point in 2001, and we are now in 2009, we still have not got that connector. There has been a lot of reasons for this. In the first instance, although I have had no close supervision of it, there is the commercial interest - who will be the manufacturer who would be prepared to put money into creating such a syringe - and, then, will the medical profession buy it? The next thing, of course, would be the designing of the syringe - how would we design it - and then a whole range of other issues surrounding whether professionals will be prepared to use it when they are used to using their own methodologies. Would the needle work in the way it is supposed to do? What about the law of unintended consequences? In systemic terms, will the creation of this create another problem in another area and lead to a patient death for another reason? And so forth. It has been those kinds of things. I did ring the Department of Health yesterday and ask them how we were doing, as the saying goes, and they said they had some prototypes that they were working on now and they hoped, within the not too distant future, they would have a suitable solution to that problem. Of course, I cannot say how long the future is.

Q220 Chairman: Were there other conclusions that have been acted on?

Professor Toft: Absolutely; yes. I would say that there was reticence. I tend to be, as you can tell, a bit voluble, and my recommendations tend to be a bit the same way, but I do not like people dying particularly, so I am very thorough in what I do, at least I believe I am, and there was certainly some resistance by professionals in the field to my recommendations. In fact, when they asked for my recommendations, I put them forward. They were all accepted by government by the way, Kevin. I then found that when they actually did the guidance for my recommendations they only incorporated half of them, or approximately half of them, into the guidance to be implemented in hospitals. I then had a very robust conversation with Professor Mike Richards, who is the Cancer Csar for the country, who saw my point and took on absolutely my point and, eventually, all the recommendations were implemented apart from the one with the technology. It was a hard-fought battle, if you will, but they are now all there and I am very happy to report that when I contacted the MPSA yesterday, as far as we are aware, there have been certainly no deaths in this country as a direct result of inadvertent administration of vincristine and there have been no near misses either, so I would like to think that the recommendations actually work.

Q221 Chairman: Why is it? Manufacturing? Does it have too close a relationship with some or all clinicians?

Professor Toft: It may well be. There are certainly sensitivities around people from outside telling professionals in another sphere precisely what they ought to be doing and certainly in terms of the technology, commercial manufacturers of medical devices do not necessarily want to change their production lines because, thank you very much, they have been making some money. A good example of that would be a category of medicines called look-alike sound-alike medicines. I had the great honour to do an investigation at United Bristol Healthcare Trust into why four children were given 500 times too much heparin consecutively, one after the other after the other. It turns out that it is because of a National Heath specification on the way in which the labelling on ampoules is produced - it is all, I think, Pantone yellow and black. I asked why this had not been changed and why did they not do something about it, and they said manufacturers do not want to change their branding and their labelling as it costs money and they will lose their branding position in the market, and so forth. I did, in fact, speak to a gentleman called Lee McGill, at the European Commission, about this, and he very kindly sent me an email back saying that he would try and bring this to the EU's attention. However, he did say that the different directorates in the EU have different opinions, and very often there are silos in between them and, therefore, it becomes extremely difficult for people to do these things. So I thought I would bring this to Parliament's attention, because perhaps Parliament could do something if I cannot.

Q222 Dr Taylor: Again to Brian. Root cause analysis. The first sentence in paragraph 4.4 of your submission is, "There is evidence, however, from the root cause analysis of four patient safety incidents I carried out that the current national reporting and learning system is not as effective as it needs to be." Can you tell us something about root cause analysis? Is it being used enough and, if not, why not?

Professor Toft: It is certainly being used - that is a fact - but there is evidence from a study done by Wallis et al in this country that the depth and level of understanding of the people undertaking it is not sufficient for the job that it entails.

Q223 Dr Taylor: What sort of people are undertaking it other than you at the moment?

Professor Toft: It is not just me. The MPSA, very wisely in my view, did a three-day course where people were trained from hospitals how to engage in root cause analysis. This was at a very superficial level but, nonetheless, given from a standing start, they did extremely well; it was well done. The idea was that these people would then be the trainers of the trainers, but if you have only been on three-day course yourself and you suddenly find yourself endowed with lots and lots of other jobs to do, it then becomes very difficult, if not impossible, to actually fulfil that promise. You will do some but perhaps not as well as you would want to do it; perhaps you do not even get an opportunity to do it. The problem there was that, although they were well trained from that level, they had great difficulty inserting that training into the hospitals concerned, and the study certainly showed that, and the study done in Australia by Braithwaite also came up with very similar things. So we have got people doing it, but not trained to a sufficient level that you could pull things out of it. They work at a more operational level; so they can tell you what went on but they cannot tell you why it went on. I have been doing what I have been doing for 30 years now and I think I might just be getting competent at it; so it is not an easy thing.

Q224 Dr Taylor: What sort of people are being trained, are going on this three-day course?

Professor Toft: Generally speaking, there have been doctors and there have been nurses. I would have said mainly nurses were the larger part of the people that have actually gone on the training courses.

Q225 Dr Taylor: Who are the people that should take it on?

Professor Toft: I think everybody should be trained in it. That is my view, of course, and I have been known to be called radical and extreme. I think that everybody should be trained in root cause analysis, because if you understand the way in which errors occur, then you can usually see them unfolding before they do. It is a case of seeing the symptoms. If everybody was trained at least in part in doing them, then they would not have the problem of some poor soul who has got 10,000 jobs to do being given a job and they then do the root cause analysis in their part-time, in their spare time, pretty much like you do select committees in your part-time, other than your real job of being MP.

Q226 Dr Taylor: Should it be in medical student training, in nurse student training?

Professor Toft: Absolutely. It should be in medical student training. Right from the very beginning that they start their training there should be a gradual build up of notions of error, how human error is created, how the whole system works together, how it leads to the creation of errors but, most importantly, everybody should be told directly that nobody is perfect - nobody. We all make errors, me included. I have studied it for 30 years and I make the same ones, much to my embarrassment, more than once. Nonetheless, nobody is perfect, and therefore the doctors need to know that so that they feel more able to come forward and say, "I have had a near miss today", and so forth, because very often the opprobrium that comes with having made an error stops them from doing so.

Q227 Dr Taylor: Ought the people doing the root cause analysis to be from outside the particular institution being investigated?

Professor Toft: Without doubt. There is undoubtedly pressure put on people from within their own institution and from within their own profession to, if you will, massage or bowdlerise their own results, and certainly I have had an experience of that quite recently.

Q228 Dr Taylor: So the training for everybody would be to prevent untoward incidents happening but, if they happen, the investigation should be done by somebody outside.

Professor Toft: I absolutely believe so. I am running an investigation at the moment, a review, and it took me nearly ten months to get two retired surgeon (and I say the word "retired") to come and sit on my panel. The active surgeons, the ones that were operating, were too frightened to sit on it in case the results that were produced were against what had already been produced by the British Orthopaedic Association. So I am doing a review of a review, and one surgeon actually said to me, "I have to think very deeply about this, Brian, but I am probably not going to sit on your panel because I am quite frightened for my livelihood."

Q229 Dr Taylor: How can you get across the profession's almost natural intention of protecting themselves?

Professor Toft: That is a huge question, and I am afraid I have no easy answer for that. One of the ways I would say is by influencing the culture of all organisations and having them recognise that people will make mistakes and that they should then support the people and their colleagues. They should support them heavily when they have made a mistake because, apart from the Shipmans and Beverly Allitts of this world, over which I have no trouble, of the remainder, I have never yet found a physician or a nurse that has been involved in any of the incidents I have looked at, which have been quite serious, that has not been profoundly disturbed by what took place. If you spend all your life looking after people, keeping them free of pain, and you suddenly find out that your inadvertent action has resulted in exactly the opposite, your raison d'etre for living has gone. In some cases I would say that the healthcare professionals involved actually needed more care, help, advice and support than the families did.

Q230 Dr Stoate: Your memorandum refers to involuntary automaticity. Could you explain to the committee what that is, reasonably succinctly?

Professor Toft: I will do my best. I am going to start by talking about automaticity. You may remember that when you started to drive, on your first day behind the wheel, you would put the ignition key in, twist it on, sit there with white knuckles and set off and everything that you did you thought about consciously, you paid attention to: "I must change gear, depress the clutch", and so forth. By the way, I do not drive, it is far too dangerous. I let my wife do that. You have to concentrate very heavily on what you are doing, so you process the rules one by one cognitively. Eventually, a year later, you step in, click, yes, you have got it, and away you go, and that is automaticity. You become so well practised that the cognitive system says, "I am in driving mode; I will now just tell you what the rules are", and then you slip in, you just do them automatically. Generally speaking, people talk about automaticity and its benefit because it is very helpful. When you start driving you cannot speak to your next-door passenger because you are too busy concentrating; when you have been driving for 12 months, you can talk to everybody, you can look at the sat-nav, you can look at your map and all that kind of stuff, so you can do more than two things at once, so it is a very useful thing. However, it has a cost. One of the first people to recognise this cost was a guy called Barshey who works for NASA. What he found was that pilots, when they are doing landing routines and cockpit routines, sometimes get into a litany a bit like in church. The priest calls to the congregation and the congregation call back. In church you do not have to do anything. If you are in an aeroplane and you do not do things, it is likely to be complicated. The example he gave was of an aircraft coming into Casper, Wyoming, and he said that the co-pilot said, "Landing wheels down", and the pilot repeated, "Landing wheels down." He said, "Three greens", the pilot said, "Three greens." He said, "Landing wheels down and locked", and the pilot replied, "Landing wheels down and locked", and then the plane landed with the wheels up. Just remember that when you go on your holidays! Clearly, that kind of behaviour is not conscious and it certainly is not wanted. You have done something without being effective; so you do the check without being effective. That is what involuntary automaticity is. In a medical sense I did an investigation where a patient received two and a half times too much radiotherapy. To get this amount of radiotherapy the patient had had 14 separate treatments and she had been attended to by 12 different radiotherapists working in assorted pairs, using a verbal double-checking protocol - so a standard operating procedure. They would call the patient's name out, they would call out the amount of radiation which was to be given, they would call out whether or not a thing called a wedge was in the beam of the radiation. Because it is a constant source of radiation, the wedge acts as piece of lead, so it absorbs radiation. So if it is supposed to be in and it is not in, what do you think happens? Clearly, the patient gets over radiated, and this is what happened. On the console, the controls of the linear accelerator, there was a thing that said, "Wedge", and it would say either "in" or it would say "out". The word that should have been there was "in" and the word that was actually there was "out", and on 14 consecutive occasions two dedicated, conscientious radiotherapists, using a verbal double-checking safety protocol, read the word "out" as "in". I could not understand how people who were so dedicated to their patients could do that, so I did the research and I came up with this notion of involuntary automaticity, which Parliament very kindly, through the Right Hon Alan Webb, asked questions about, and that is how it came about. But that is what it is; it is about people seeing what they expect to see and it not being so.

Q231 Dr Stoate: That was the short answer.

Professor Toft: The very short answer!

Q232 Dr Stoate: The slightly longer answer, but hopefully not too much longer: what do we do preventing it? Are there technological solutions which are failsafe?

Professor Toft: In terms of radiotherapy there are. What happened was that the data was entered manually. The data could have been entered in technologically; it could have gone from one source directly into the other source, into the (?) database, and if that had taken place, then I would not have found that and the incident would not have taken place.

Q233 Dr Stoate: So there are ways of preventing it?

Professor Toft: There are ways, but the one thing I would just point out is involuntary automaticity is forced upon the healthcare professional by the system in which they are operating; they have no control over it whatsoever. If you are captured by this, you are captured by it and it is completely unconscious. Therefore, if a person is put into that situation, what I have argued in my latest paper, which is coming out in February in the Quality and Safety in Healthcare Journal, is that since managers are responsible for the environment in which clinical people work, the actual work environment, if a clinician should say that these working conditions are so onerous that I could easily fall into involuntary automaticity and the manager does nothing, then I have recommended that in law (because I have written it with a lawyer) in fact the manager should be held accountable for the incident and not the clinician.

Q234 Dr Stoate: It is fair to say, it is not only a good thing, it is an essential thing to have?

Professor Toft: It is an essential thing.

Q235 Charlotte Atkins: Professor Adam, one of the conclusions in the report Towards Safer Radiotherapy is that over reliance on automated procedures is a factor in some patient safety incidents in radiotherapy. Can you tell us a bit more about that?

Professor Adam: To some extent, to a large extent, it really overlaps with what you have just heard from Brian, that one can used an established method of doing something without second checks and making sure that what is supposed happen has actually happened, and this is partly why one must not rely on just this kind of process and why there are recommendations, for example, in the document Towards Safer Radiotherapy for using additional checks such as in-vivo dosimetry. In-vivo dosimetry, essentially, uses a machine that checks that the dose that is going to be delivered to the patient is the correct dose. It is not in itself sufficient, and all the things that we have just heard from Brian are also very important, but there are additional things that you can do to make sure that you do not just rely on certain processes that are subject to human error.

Q236 Charlotte Atkins: So it is a mix of better training and better technology, is it?

Professor Adam: Exactly, you need both, and the further training also has to have processes such as back-checking, checking that what you are supposed to do has actually had the intended effect - the kind of thing that you have just heard from Brian.

Q237 Charlotte Atkins: The fairly horrifying concern, I think, for any patient is that you have these experts who are following procedures but somehow just overlook the fact that what their eyes are showing them their brain does not recognise.

Professor Adam: Yes.

Q238 Charlotte Atkins: When people are confronted with results of investigations into these accidents, into this misjudgments, in your experience, do they learn from those experiences or do they just internalise, move on and carry on as they did in the past?

Professor Adam: Of course, it does depend on the individual, but the system does try to encourage methods of learning, and the Royal College of Radiologists is putting a lot of effort into that. For example, we really favour a no-blame system where you report near misses, so that we have much more widespread publicity of where things nearly went wrong, because there are many more of those than serious injury, so that people can refer to those and learn from them. Also, the few cases where there is serious injury---. By the way - one has to put things into perspective - real, serious clinical injury in radiotherapy is very rare, something like three in 100,000, but those also ought to be publicly available so that people can refer to them and learn from them, and having them inadequately accessible is a problem. Of course, there is a difference between individuals, but that is just part of the problem. Accessibility of the information is also an issue, and that is something that is being looked at.

Q239 Charlotte Atkins: Do you think the culture is changing? Do you think that with new people coming into the profession the culture is changing and it is improving, or do you find there is a real mix between the younger new recruits and the more established radiotherapists who have been in the business a long time?

Professor Adam: I believe that there is much more openness now and much more widespread recognition that information is good. For example, as I am sure you have heard, the National Patient Safety Agency will say more reporting to them is an indication of a good organisation, so that organisations that report more near misses should be considered better than ones that do not, rather than the reverse. I think that is something that is becoming generally accepted and people are recognising that, and that is a very good thing.

Q240 Charlotte Atkins: With hospitals worrying about their ratings in terms of patient image and everything else, is there not any pressure from managements, chief executives of hospitals not to report near misses on the basis that it might affect their rating in the public mind?

Professor Adam: I have not found that, and I think that is partly because people are realising that there is real benefit from doing that, but it has to be done in the proper way. There has to be this no-blame culture---

Q241 Charlotte Atkins: Is there not a different objective? A chief executive might have a very different agenda than perhaps a radiotherapist?

Professor Adam: I think where problems arise is where lawyers get involved. If there is serious injury and it is a legal issue, then, of course, there are pressures from lawyers about what can be disclosed and what cannot be disclosed, and even the doctors involved do not have complete control over that because you would be going against legal advice. I think there, there are there real issues, but lower down the scale, where you are talking about near misses or accidents that are not particularly serious, although there is an issue there as to whether there are concerns about disclosure, people feel that it is a good thing and I do believe that there is a real change in culture. As I say, when the courts get involved it is more complicated.

Q242 Mr Scott: Professor Adam, can you outline the risks and benefits of technology as regards patient safety in two speciality areas of clinical oncology and clinical radiology?

Professor Adam: Yes, I think technology has benefits and risks in both, but, as I said in the general answer I gave at the beginning, I believe, in both of these disciplines the contribution towards increasing patient safety is vastly greater than the risks introduced by new technology. Perhaps I could give examples in both. In a study of clinical oncology there is new technology, such as image-guided radiotherapy, that is much safer than past techniques, and I will explain that. You have tumours in organs that move. The most obvious example, perhaps, is a tumour in the lung. Obviously, as the patient breathes in and out the tumour will move up and down. Having the x-ray beam which is treating the patient tracking that tumour and shining the treating beam to exactly the right area obviously is a lot safer than having the tumour being missed because the tumour has changed its position. That is image-guided radiotherapy. Intensity modulated radiotherapy (that is controlling the intensity, the strength, of the beam, using a computer, so you get a beam that is the three-dimensional shape of the tumour, avoiding structures that are vulnerable and should not be radiated and giving the dose where it should go) again, is an example of technology increasing safety. Proton beam therapy, which does not exist in this country, sadly, and we are sending patients abroad to be treated with that - protons are positively charged particles that have a very, very short range and so, using protons to treat patients in critical areas, for example, next to the spinal cord, you can really give the dose near dangerous areas, destroy the tumour and not harm adjacent structures. The cyber knife - again there is only one, I believe, in the country, in the private sector and none in the NHS - uses a system of robotic control of a pencil beam and introduces an even greater level of safety, where the pencil beam has sub-millimetre accuracy and destroys tumours with great safety without damaging adjacent structures. Those are just four examples. In clinical radiology, both diagnostic and interventional--- Let us start with interventional, because I believe that with regard to patient safety there is perhaps no other discipline in the Health Service where the gap between the potential of discipline to increase patient safety and reality is greater, and this really relates to the lack of availability of interventional radiology, both in and out of hours. Of course the problem is even more serious out of hours. There are only 10% of trusts in this country that offer formal interventional radiology on call ratings, out of hours. Interventional radiology is pinhole surgery controlled by imaging. You are really using tiny instruments that advance to the right part of the body, under x-ray control, to do what you need to do, and there are many examples of procedures that increase patient safety but perhaps an obvious one is stopping haemorrhage. If you have someone who is bleeding from the large bowel that comes into hospital and they are losing a lot of blood and their blood pressure is low, opening them up and performing open surgery in that setting has great risks. You can introduce a tiny catheter under local anaesthetic and block the artery that is bleeding. That is called embolisation. That can be done at a tiny fraction of the risk that surgery carries, and it is almost always successful. That should be available in every acute hospital. Every patient should have access to it. There have been lots of recommendations regarding this. For example, when there were ten maternal deaths at Northwick Park between 2002 and 2005, the Healthcare Commission made two national recommendations. One of those two national recommendations was that interventional radiology should be available to all patients that need it. We are nowhere near achieving that goal, and this is sad because---

Q243 Mr Scott: Is that purely financial?

Professor Adam: It is not purely financial. It is partly financial, but not financial in the sense that the money is not there to fund it. Because interventional radiology is extremely cost-effective, you would actually save a lot of money if it were used more widely, so in that sense it is not financial, but it is financial in another way, that the funding structure of interventional radiology is a huge problem. There is a paradox there. We have to look at how it developed, and it developed through diagnostic radiology, it developed through the invasive techniques that diagnostic radiologists were using to reach a diagnosis, for example in angiography, where you put a catheter in, squirt some dye that shows up on x-ray and that shows you what is going on. Those techniques were developed and adapted to do the kind of thing that I have been describing. Therefore, interventional radiology is the only discipline that directly treats patients. It is essentially pinhole surgery, as I said before, but it sits within a service department that does not have the infrastructure for clinical practice, and that includes the setting of contracts with PCTs and where money comes in and funds what you do. Interventional radiology is funded out of budgets of radiology departments; the same budgets that fund CT scans and MRI scans There are huge pressures on those, as we know, and when interventional radiology techniques are introduced in a hospital, new techniques, actually that is usually a drain on the resources of the x-ray department, so there are huge disincentives to introduce new interventional radiology. So funding is an issue, but only in that way because, as I said before, it is extremely cost-effective. For example, if you treat the fibroids of a woman that is bleeding from the uterus with hysterectomy, which is the usual way or one of the common surgical ways of doing it, it costs twice as much as blocking the arteries to that uterus and treating the fibroid and destroying the fibroid using interventional radiological techniques and, on top of that, the woman can still have children, whereas, of course, if the uterus is out she cannot have any more children. So it is safer, it is effective, it is cheaper and we are not using it because of structural problems. What we really need to develop is designated post and interventional radiology so that the career structure develops, so that young people go into it. They are very enthusiastic. Lots and lots of radiologists really want to do this. So the people are there, they are full of enthusiasm, they want to do it, but we do not have the structure to offer them, for the reasons that I have explained.

Q244 Sandra Gidley: I want to move on to your memo, which mentions teleradiology and says that there were a number of problems associated with it. I wondered whether you thought it was more of a threat to patient safety than a benefit.

Professor Adam: Again, it is both, but it depends on how it is used. There are huge benefits in some situations. For example, the British Army in the field uses teleradiology. Images are beamed sometimes thousands of miles away so that one can get a specialist opinion quickly when it is needed, and that is wonderful. In the context of the NHS, one can get a super specialist's opinion, again, quickly and improve on an opinion that you would have given without the benefit of that expert opinion. So there are great benefits to teleradiology. Where the risks come in is when one uses this wonderful new tool in a way in which it should not be used. If you are using it simply to save money rather than improve the standard of patient care and increase patient safety, then problems can arise, and there are two areas that affect that. One is the actual pattern of practice. Modern diagnostic radiology is a clinical discipline that relies enormously on the interaction between the referring physician and the radiologist. When you are reporting films, very often you would get additional information about the clinical state of the patient, you will have access to pathology results, biochemistry results and, most importantly, you will have access to the previous images of that patient. There is a saying in radiology that the world's greatest consultant is the old films, because very often comparison with the old films will give you invaluable information. If you are trying to do that from thousand of miles away, especially if you are doing it from abroad in a completely different system, then problems can arise and your report may end up describing what is there but it may not tell the referring physician what it really is, and that can result in an uninformative report and can lead to the wrong treatment. That is one issue. The other one is the issue of regulation. I think we often take it for granted in this country and we underestimate how well off we are in the way that the medical profession is regulated and what the benefits are which accrue from that in terms of patient safety. A radiologist in this country is regulated and is accountable at various levels. First of all, you are accountable to the patient, who, if something goes wrong, can complain about you, and quite rightly so. You are accountable to the referring physician or surgeon, who may operate on the basis of what you have said. You are accountable to the clinical governance arrangements within your own department, and there are many now in the x-ray department. There are many methods of increasing patient safety, including regular discrepancy meetings, for example - that is, error meetings where you look at errors that are made to learn lessons. You are accountable to the clinical governance arrangements within your own hospital, you are accountable to the General Medical Council, you are accountable to the court ultimately and, as from this year, doctors in this country are going to be licensed; they are going to have a license to practise that will last for five years. So there are seven levels of accountability. These things, especially if teleradiology is used to get reports from abroad, certainly do not apply in that same way, and you may have a situation where the people that are providing these reports are essentially, in practical terms, accountable to the company that employs them, who then gives assurance to those that commission that service. I think that is an important consideration that really needs to be addressed, because there are real issues for patient safety there. At the very least, patients ought to know when their images are being sent abroad, because they may not realise that this is happening and they do not have the same structure as they take for granted here.

Q245 Sandra Gidley: Have there been problems in this direction, or is it just a concern that there could be problems?

Professor Adam: There have been problems. It is not the kind of thing where a large case study has been done (and it would be extremely difficult to do a comparative randomised study), but the problems that arise is that reports will come in and you will see that they are of the type that I have described - descriptive reports, reports that tell you that something is there but where there is a non-committal conclusion and where it is less useful than a report that you would expect to get from a local radiology department.

Q246 Sandra Gidley: I may have misread this, but one of the concerns seemed to be that there was no guarantee of the picture of the patient that you thought you were looking at matching the existing clinical records so that you have the whole picture there. Why is that a problem when we have got an NHS IT system?

Professor Adam: That you are looking at the wrong image, you mean?

Q247 Sandra Gidley: Yes.

Professor Adam: That can happen both in radiology and there are all sorts of ways---. It would be very rare, but, nevertheless, that is the kind of issue that sometimes is down to technology, is sometimes down to human error. It is not specific to teleradiology. Where there is a problem is that you may not know who the person is that is reporting that film - there are no ways of checking it. We often take the databases that we have in this country and the systems that we have for granted. For example, we have a specialist register, so every radiologist is on the specialist register. There are many European countries where there is no specialist register accessible to the public. So one assumes that something that we have exists elsewhere, but sometimes it does not.

Q248 Sandra Gidley: What is the best way to get some evidence on this? Has there been much research done? What research needs to be done to establish what are the safest processes or where the dangers are?

Professor Adam: Most of the research that has been carried out into teleradiology has not been on regulatory issues, it has been on image quality and technical aspects of this. I think that it is high time that research was carried out on precisely these issues and actually finding out what the regulatory processes are that are used and what is acceptable and what is not acceptable when contracts are being drawn up.

Q249 Dr Naysmith: Dr Whitaker, you have been sitting rather quietly all morning listening to what has been a fascinating morning so far.

Dr Whitaker: I agree.

Q250 Dr Naysmith: Now is your chance to contribute on anaesthesia, your specialty. Obviously, the safety of the apparatus and the bits of equipment being used in your speciality are of paramount importance in this aspect, but you say in your memorandum that the regulatory aspect of this, which is overseen by the MHRA, is done well but, as regards buying the safest equipment, "the National Health Service rarely uses its purchasing power as it could and", in your opinion, "should". Could you expand on that a little?

Dr Whitaker: Yes. Perhaps I can say a little bit about standards. We were talking about standards just now. The EU regulations are more balanced in mobility of labour than patient safety, but there you go. The standards that the MHRA use are drawn up by a standards committee, British Standards ISO, which can take three to five years to draw up, and on the committees there are representatives from the manufacturers and also clinicians. We are very lucky to have a number of clinicians who are interested in this work. They have to go through painstakingly and look through all iterations of the documents and make sure that it is all going to be safe, and they often do this in their own time and often now the meetings are in Europe and they are having trouble getting expenses to do this. This is a historical thing, because prior to 1990 all this money was top-sliced and paid centrally by the scientific and technical branch of the Department of Health for all specialties. At that time I believe it was about a quarter of a million pounds for the whole NHS, but, unfortunately, this was devolved down to trusts at that time and, effectively, has been lost and, frequently, when these consultants go to their own trusts and ask for the expenses they have trouble getting them because the local trust does not see the purpose of them going to an international standards meeting. Recently there have been several European initiatives that have wanted to encourage clinicians to be involved in these committees rather than dominated by the manufacturers. I think BERR (the Department for Business, Enterprise and Regulatory Reform) has made a statement on this as well, and we think it is now time for this money to be top-sliced again and come back maybe to be administered by the MHRA itself so that there can be adequate standards made. Because these standards take three or four years to develop, often because lessons learned from critical incident reporting and other developments can take place, some pieces of equipment that meet the minimum standards will actually have new safety features that are more advanced, and that is where a PASA purchasing for safety policy can come in, because they can more immediately take account of those new developments. Brian was talking about the Luer connectors. In fact, to get a new standard in this process for a Luer connector will take five, seven years, but the MPSA, I think, has recently, before Christmas, issued a statement saying that PASA will only purchase this new type of connector after 2010; it will only be on their purchasing list. This is a way of responding more quickly to these changes, and it is also a way, I think, of influencing manufactures to develop safety features.

Q251 Dr Naysmith: Can I just be clear what you are saying? You are saying that the MHRA is administering safety standards adequately but these standards are out of date, that there should be more up-to-date standards?

Dr Whitaker: The process takes a long time to develop, three or four years, to get a new standard for something and other developments can take place in the meantime. Also, they are minimum standards. So there could be new ideas come in, new radiology things that come in, that will not yet be in the international standards, but, I think, for the patients' sake we should be getting that technology if it is thought to be valuable and safe.

Q252 Dr Naysmith: How should this purchasing power be being used then? How should the purchasing power of the NHS change from what you have just outlined?

Dr Whitaker: I think it is changing. I think they are starting to take account. This Luer lock thing, I think, is very interesting, and they are developing product councils, with which clinicians are getting involved, that can give them direct information to try and influence this. The value of PASA is, of course, that the NHS is a massive purchasing body and a manufacturer would like to be doing business with the NHS, and so I think that is how they can influence things. One reason why they are not investing in new Luer lock connectors is probably they all wait for the standard, because they do not want to start new tooling equipment until the standard comes, but if the NHS says, "We are not going to buy the connectors unless they meet these new rules", that will be a good thing. Another example is the MPSA have said that all drugs should be in a pre-filled syringe that has maybe some tagging system on it to identify it; and for a common anaesthetic agent, propofol, one company makes a syringe like that, but not all hospitals purchase it, and that is an example where purchasing for safety would help support manufacturers who are making safety innovations and bring the others up to standard.

Q253 Dr Naysmith: Could I quote you something that the Royal College of Anaesthetists said in evidence to us? They said, "Patients may be placed at risk where equipment is old or poorly maintained and where new equipment is introduced without appropriate training for the user." Would you agree with these statements?

Dr Whitaker: Yes, traditionally equipment has not been replaced as quickly as it might have been. Often the NHS finance manager will probably have had three or four new cars in the time that he has replaced one anaesthetic machine. Each hospital needs to have a rolling programme of either renewal or a rolling programme of finance so that it does not have to always become a crisis at the end of ten years when the equipment is starting to fail. Things are improving but there is still a culture there that responds to the crisis as it occurs. I would say, though, that manufacturers are very responsible in the UK. We have good relations with BAREMA (the British Anaesthetic and Respiratory Equipment Manufacturers Association). They exchange ideas of good practice with us.

Q254 Dr Naysmith: Do they participate in training? Do they fulfil their training---

Dr Whitaker: The manufacturers, yes, they are very keen on that. The manufacturers appreciate that it is necessary, if they are going to bring in a new product, that the new product goes well in hospital, because it is bad commercially if it does not. We have just got 12 new ventilators, 12 theatres re-equipped, and the manufacturer trained three of my consultant colleagues to be trainers and then they trained the rest of the other 50 consultants in, I think, a one and a half hour programme. We went through a training manual to make sure we knew what all the knobs and buttons did. That sort of thing happens. Patient controlled analgesia pumps - when we got new ones of those the manufacturer sent a nurse that worked for them that was trained to come and train the staff, including the night staff as well, because often when you have these training programmes the staff that work at night never get to find out about it, it all happens during the day - so they are starting to understand that.

Q255 Chairman: Is the issue of training part of the procurement process? Is there a figure on it to your knowledge?

Dr Whitaker: When equipment bids come in, the hospital can either ask for---. Certainly the PCA one that I am familiar with, the companies that were tendering said they would provide some different levels of training in with the price. So it was part of that.

Q256 Chairman: So it was priced, even if it was not absolutely transparent?

Dr Whitaker: Yes.

Q257 Jim Dowd: Dr Whitaker, I want to look at bar-coding being used for double-checking prior to drug administration in anaesthesia. They do this in New Zealand, I am led to believe. Could you briefly describe how the system works and whether you think it would be beneficial in the UK?

Dr Whitaker: Yes. This is really all about medication safety and getting the right medication for the patient, and anaesthetists are expert in this. Someone at my stage in my career has probably given over 500,000 intravenous medications to patients, and some of them will have been incorrect and some of those I will know about. I remember in the 1980s I gave 30 mg of morphine to a patient when I intended to give ten. The reason for that was the only difference on the ampoule was a three and a ten. Then in the 1990s, using a hand-labelled syringe, I gave a cardiac patient a drug labelled PRA when I meant to give them one labelled PCB. They did not come to any harm, but I remember those things. Now we all use these stick-on labels. They are colour-coded, they have an international colour-code for the group and also they have the name on so there is no misunderstanding. The two situations both occurred because I had mistaken the labelling. The other suggestion that has been made---

Q258 Jim Dowd: We do not always get audio visual aids!

Dr Whitaker: I have not brought a bar-code machine! ---is a pre-filled syringe, and this one comes from the manufacturer with the drug in, with the label with the right colour, and it is clear for everyone to see. That is slightly more expensive, but it cuts out the ampoule, it cuts out the transport, it cuts out the storage and it produces a much safer product. The idea with the bar-coding is to go a stage further. This would then have a bar-code put on by the manufacturer and you would put this against the bar-code reader and the thing says to you: "Ephedrine hydrochloride 30 mgs", so you know what it is - it is a double-check with a computer. This has been piloted in New Zealand and now the MPSA, in partnership with the college of the association, has been piloting it at Papworth Hospital and Wrexham. The pilot project has just finished and I understand it has gone well and we are going to hear about this in the coming months. Just to be a control situation, nine of the hospitals have been piloting double-checking with individuals. So the ODA in the operating theatre, the anaesthetist, said, "Can you tell me what this is?", and the ODA read it out for you so you know that you have checked the drug and everything, but bar-coding takes out that even further human element, because people have shown - the sorts of things that Brian was talking about - that even situations like that can fool a human being.

Q259 Jim Dowd: Do you think there is any reason why it could not be extended to the UK and why we could not utilise it here?

Dr Whitaker: There are two things. In New Zealand I believe one of the inhibitions has been getting pre-filled syringes bar-coded and centrally provided. I understand in Wrexham these have been provided by the local pharmacy, and that has worked very well. That is a bit of a cultural change that needs to take place. Also, the equipment, obviously, compared to some of the new x-ray equipment, is not in the same ball park figure, but there will be a cost there as well and there will some training issues, of course.

Q260 Jim Dowd: What about going to the manufactures to put the codes on if they are pre-filled?

Dr Whitaker: Yes, again, that could be done. Again, it maybe the person for safety (PASA) would have to say, "From 2010 we are only going to buy ephedrine looking like this." There are some pharmacy issues, mainly with stability of drugs in these situations, but maybe certain high-risk drugs, such as heparin that Brian has done a report on and potassium, should automatically come in pre-filled syringes so that it cuts out error of drawing drugs up into the wrong ampoule. Sometimes pharmacists will, usually for costs reasons, change their suppliers of, say, heparin maybe three times in three months and you end up in the drug cupboard with three different boxes looking different with the same drug. This is because they have not had sufficient buffer stock of these important drugs, so that if the manufacturer does have a manufacturing problem you have still got six months' supply of an important drug so you do not have to change the packaging and confuse the operator. The doctor, patient, the three or four metres round the patient where the end-user is working is the most dangerous area and everything should be focused on making sure that that is as safe as possible and the end-user does not inherit the errors of the system further back.

Q261 Jim Dowd: As a lay person, the danger, the weakness in this system is the accuracy of the labelling, because everything is staked on that, is it not?

Dr Whitaker: Correct. A number of years ago there was a muscle relaxant put into a steroid ampoule. I think the company made these two drugs and they carried on making one a bit longer than they thought, and, fortunately, an anaesthetist spotted it. They had given steroids to a patient and they noticed that they had the effect of muscle relaxation and were able to pick it up. Errors will occur; the thing with patient safety is to try and reduce harm as much as possible. To say it will be none, I think, is possibly unrealistic.

Q262 Sandra Gidley: Can you clarify: is the patient bar-coded and is the drug checked against a bar-coded patient as well?

Dr Whitaker: No. Barcodes have been used for blood transfusion, but in this particular situation the patient is checked to make sure Dr David Whitaker is the patient and then all the drugs that are given are checked - the anaesthetist is giving the correct drugs.

Q263 Sandra Gidley: I am just wondering. To quote your example earlier, that you gave the wrong morphine, if we juxtapose that with the example we heard earlier of the involuntary automaticity, you might say, "Oh, morphine 30 mgs, yes, that is fine", forgetting momentarily - that is the human error bit - that actually 10 had been intended, but if it had been written up before hand and the patient was bar-coded that error would have also been discovered at that stage, so is there a stage further that we could go?

Dr Whitaker: Yes, there are lots of further stages we can go to tighten these things up, and bar coding---. If you watched the children's Christmas lectures, there was the IT chap who was talking about very small chips that have got infinite numbers on, so you could use those. Bar-codes is tested technology.

Q264 Jim Dowd: They do that for pets, do they not?

Dr Whitaker: In the newsagents where you buy your copy of the Times they are bar-coded.

Q265 Dr Stoate: It could be a condition of registration of birth: you have to have a chip put under your skin!

Chairman: Can I disassociate from that. Richard, I think you have probably got a final question in this session.

Q266 Dr Taylor: Thank you very much. As we are unbelievably ahead of schedule, could I ask a very general question. Have each of you got any ideas of accidents actually waiting to happen. In the old days we had things like chlorpromazide and chlorpromazine, which was one of the classical errors that were made. Have you any views of things that we should be getting onto now that are accidents waiting to happen?

Professor Toft: Very definitely. Look-alike drugs are a killer. The National Patient Safety Agency, between January 2005 and July 2006, had 800 incidents of the wrong drug being administered to patients per month. Twenty-five patients died and there were 28 serious injuries. Look-alike drugs are real killers. People think they have got the right drug in their hand when they have not, and it is not because they are stupid or because they do not know what to do, it is because they are captured by the situation and they think they see what they see, but they do not, and that is a real killer. If I could recommend anything that you do, you would do something to the manufactures to stop that from happening.

Q267 Sandra Gidley: But they are doing that. Some of the generics manufacturers are producing packs.

Professor Toft: Wockhardt is doing it because I asked them to.

Q268 Sandra Gidley: I still think there are potential dangers though.

Professor Toft: Wockhardt were involved in the incident I was involved in investigating at UBHT, and so I contacted them, and when I told them what had happened they were literally aghast, and their head of safety, reliability and quality took it on his own back to inform the company, and more generally, and as of November last year they changed all their heparin packaging and labelling so that it did not resemble the Pantone yellow and black that had been previously the case, which I think is well worth mentioning because they are clearly proactive - reactive in one sense but proactive in the general sense - but they took it on board and acted on it.

Q269 Jim Dowd: Can we ask you to contact all the others then!

Professor Toft: Jim, I would be delighted to do it. If I can have the power of this committee, I would be delighted to do it.

Jim Dowd: Tell them we said so.

Q270 Dr Stoate: You have got it.

Professor Adam: I have not got an example of a process error, but what I will say to you is that somewhere in the country today someone is going to either have a serious injury or die because interventional technology is not available to them, and they will either get surgery or no treatment at all, and that is an accident waiting to happen.

Professor Toft: Following on from Brian, the National Blood Transfusion Service blood bag, I think, needs to be redesigned. Blood transfusion is very safe in this country, but the commonest form of error is the administration at the bedside or it is an administrative error. It is covered with all sorts of information, you only need to check about six things to make sure you have got the right blood for patient: it has got the address of the manufacturer that made the bag, a number of different bar-codes that are all used in the process way back and the end-user is confused by it, and the whole label is actually upside down. Again, the thing was designed for the donor session where the bag hangs up the other way round as it comes out of the patient and, when you give it back to a patient, the label is upside down. This is a thing that has grown up out of evolution that I think could do with a serious look at. I have been to a number of countries around the world. I always ask, "Can I see a bag of your blood?", and they have all got the labels upside down.

Chairman: Can I thank all three of you very much indeed for coming and helping us in this inquiry.

Witnesses: Professor Mike Murphy, Consultant Haematologist, John Radcliffe Hospital, Professor Bryony Dean Franklin, Centre for Medication Safety and Service Quality, Imperial College Healthcare NHS Trust, and Mr Roger Lamb, Healthcare Manager, GS1 UK, gave evidence.

Q271 Chairman: Good morning. Could I welcome you to the second session of our third day's evidence taking in relation to our Patient Safety Inquiry. I wonder if I could ask you, for the sake of the record, to give us your name and the current position that you hold.

Professor Franklin: I am Professor Bryony Dean Franklin; I am Executive Lead Pharmacist for Research at Imperial College Healthcare NHS Trust and Director of the Centre for Medication safety and Service Quality.

Professor Murphy: I am Professor Mike Murphy; I am Professor of Blood Transfusion Medicine at the University of Oxford; I am a consultant haematologist for NHS blood and transplants at the Oxford, Radcliffe Hospitals and I am the Secretary of the National Blood Transfusion Committee.

Q272 Dr Naysmith: You are going to get asked why blood bags are upside down!

Professor Murphy: I was thinking that.

Mr Lamb: My name is Roger Lamb; I work for GS1 UK, which is an international standards organisation. Some of you might know us better by our former name, which The European Article Number Association. I am the Healthcare Manager for GS1 UK.

Q273 Chairman: Thank you. I have got a question to all of you and then the other questions will be specifically to individuals. Automatic identification and data capture has been seen as a revolutionary technology with the potential to transform the efficiency of healthcare. Do you think that is true or is there an element of commercial hype in relation to this?

Professor Franklin: I think it is somewhere in between. AIDC is a tool and it is a necessary tool for many of the systems that can be used to improve patient safety, but it is only that. It is a tool, it is a means to an end, and actually it is the bigger system that uses it that has the potential, or otherwise, to help patient safety. I ought to say for the committee, the experience and the examples that I am familiar with are in relation to medication safety in particular, and certainly for medication it is an important tool for automated dispensing robots, verification of patients or drugs at the point administration, but, as I say, it is a tool; it is not in itself the answer.

Professor Murphy: I would agree with that. To improve patient safety, I think, requires a combination of both technology and training, and it is very important that the user has a good understanding of the technology, how to use it and its limitations in terms of improving practice.

Professor Franklin: Clearly, the technology alone does not solve the problem, but the technology does help in improving patient safety because it ensures replicability of processes and repeatability of processes, so it gives you that reduction in error that helps make the whole process of dealing with patients more efficient and more effective. It is a tried and trusted technology; it has been around for 35 years. We do not think about it, even when we go into a supermarket nowadays and we get to the checkout, that we are not being charged for a colour television when we buy a can of beans, and that is the technology being able to automatically identify the fact that the items are different through the bar-code.

Q274 Mr Syms: Can I start, Mr Lamb, by asking you to explain what your organisation's role is and how it relates to patient safety in the National Health Service.

Mr Lamb: Yes, certainly. GS1 UK is part of over 100 member organisations around the world who are part of the GS1 family. We are an international standards organisation. We are a not-for-profit company. Our standards are used in AIDC, which is automatic identification and data technology. You could think of us really as an issuing authority for numbers that go into barcodes. We ensure that those numbers are globally unique. Those numbers are at the heart of everything we do. We have a set of standards around numbers for different things, so you can identify not just items but you can identify assets, you can identify documents, you can identify people through our numbers. Our numbers have been taken up across a number of different industries, so our standards are cross-sectoral, our standards are open, and our standards are ratified by ISA. Our standards have been accepted within the Department of Health in a publication called Coding for Success that was launched a couple of years ago, and we are working very closely with a number of agencies to implement our numbering standards across a number of different applications.

Q275 Mr Syms: Is there a danger of thinking that the answer to patient safety is always a technological fix, when actually it is much more complicated than that?

Mr Lamb: I think it is a combination of things. Clearly, good technology can be implemented badly. Systems, if they are not in place, people will potentially workaround if they do not understand why they are being asked to do things. I think awareness, training, and education are very important, but, also, it is a question of understanding why people do need to use technology in helping them change. That is probably a change in management issue as well. It is about people and processes and understanding why the new solution is better than the old solution and the impact of why they have to follow the new processes.

Q276 Dr Stoate: Are there any countries that are particularly good at this from that we could learn from?

Mr Lamb: I think we can learn from a number of other countries. Through the GS1 Group we have a healthcare user group. We are capturing information that is going on in a number of other countries around the world. There is work going on in the USA with the FDA around ePedigree, which used to authenticate drugs at the point of distribution; there is work going on around unique device identification for medical devices; there is work going on in Japan, where different technologies are being used - and we mentioned about chips a bit earlier on - using radio frequency identification; there is work around business messaging using our standards going on in Australia; and there are lots of activities in other countries, including New Zealand, who are very, very keen on the technology and are implementing it in terms of bedside verification at the point of care.

Q277 Dr Stoate: Is there any of that which could be easily and quickly transferred here or would we need to start again in the NHS?

Mr Lamb: Because the work that is going on is going on to create global standards that will allow interoperability and compatibility, then, healthcare being a global industry, it means that it does not really matter where the manufacturer of the medicine or the device is, they can put something on that device to uniquely identify it that can be read in any other country in the world. I think it is a global solution.

Q278 Dr Stoate: Could you think of a particular thing, for example, that it would be very simple to bring over from another country that we simply should just get on with?

Mr Lamb: We have started to implement our barcodes in a number of different ways on medicines that are manufactured in hospitals. That is something that is being done today and more of that can be done. There is also the use of different types of barcode to carry additional information, like batch numbers, for instance, on vaccines. Some work has been done in Canada which suggests that the problem in Canada is that they do not know which batch number necessarily has been given to which person or whether or not they were even given the vaccine - and some people are having to be revaccinated - because the system is quite hard to manage. You could scan a barcode, pick up the batch number and record that against the individual, and then you would know exactly what vaccine was given to which person at that time.

Q279 Dr Stoate: Is there any reason why the NHS is not doing more of this?

Mr Lamb: I think it is a question of focus. There are a lot of initiatives going on. We have a programme at the moment in a number of areas and vaccines is an area in which we would like to do something more in the future, but at the end of the day it is probably about resource and focus and getting the message out to the hospitals so that they can pick it up. Hospitals need to have champions in them to make these things happen. I think the best champion in a hospital would probably be the CEO - so having the CEO aware of it and maybe it being part of his key performance indicators would help to drive that programme forward.

Dr Stoate: Thank you, Chairman.

Q280 Dr Taylor: Following up on that, is there anywhere we could go and visit where barcoding is being tried for prescribing and administration purposes?

Mr Lamb: Yes, there are a number of hospitals. There are about 40 hospitals which are using our system to put on medicines which go into robots. Medicines already have GS1 codes on them that are coming from the large pharmaceutical manufacturers. All of those companies would be putting the GS1 codes on their products anyway, if they were selling their products, for instance, in retail stores, because that is where our numbers are prevalent and used.. Tesco have 50,000 products and all of those products have a GS1 number on them and that is what keeps them unique. They are being used in hospital pharmacies today but the problem is that not all of the medicines have the GS1 codes on them. We are trying to encourage pharmacies which over-label and repackage medicines, as well as those which manufacture their own, to put the GS1 codes on those items, so that we can get them into the robots and so that dispensing errors can be reduced.

Q281 Dr Taylor: Is there are any specific site in London that we could visit to see things going on?

Mr Lamb: I can certainly arrange for visits. Moorefields Hospital is doing it. I do not have the list in front of me but I could certainly provide you with a list.

Professor Franklin: Perhaps I could interject there to say that at Charing Cross Hospital we have both a dispensing robot and an automated, ward-based dispensing system, both of which rely on barcodes for their use Certainly Members of the Committee would be welcome to come and visit us there.

Q282 Dr Taylor: That would be a good place to start a visit.

Professor Franklin: I would say so. It is in London and we have two separate systems. In fact, we have four systems that use barcodes in different ways, so it might show a good illustration of the different ways in which they can be used.

Q283 Dr Taylor: We are not waiting for the whole of the NHS IT system before barcodes could be used.

Professor Murphy: No, it can be done at a very local level, very quickly and very cheaply.

Q284 Dr Taylor: If they are used for routine prescribing, would they pick up things like dangerous drug interactions?

Mr Lamb: Once, you can uniquely identify an item, then from that point on the systems can be written to make sure that things do not go wrong. The aim of the project is that you would have the barcode on the medicine when it is being dispensed, you would have the barcode on the patient's wristband and you would scan the patient's wristband to check automatically that you have the right patient, you would then scan the medicine to make sure that you have the right medicine, and if there was contraindication then the system could come back and indicate: "Do not administer this medicine to this patient" - because his condition had changed or whatever it was. In the case of Dennis Quaid, the Hollywood actor, his children in America were given the adult dose of a drug which was to flush out toxins when they should have been given the infant dose. The adult dose was 1,000 times stronger. Because that medicine did not have a barcode on it, the nurse administered the adult dose, it effectively turned their blood to water and they nearly died. These mistakes are not discriminatory. The real tragedy in that is that the same mistake was made by a different nurse in a different hospital a year earlier, when the adult dose was given to six babies and three of them died. We can eliminate those kinds of errors through this humble barcode.

Q285 Dr Taylor: What are the obstacles? Why are we not adopting it straight away?

Mr Lamb: Potentially it is about resource priority, other initiatives going, and helping hospitals hear the message. A bit more money and resource might help butt some hospitals have perhaps taken this up a bit more enthusiastically and others have been a bit slower.

Q286 Dr Taylor: You have used a buzz word in your recommended actions. You said: "leveraging Coding for Success". What should we be recommending to get Coding for Success leveraged?

Mr Lamb: I did not realise that leveraging was a buzz word. As I said earlier, it is getting this message out to people. Coding for Success was published two years ago and a copy of it went to every CEO, every chief financial officer, and every CIO of every trust. I do not know if they read it or passed it down to somebody else, but adoption is what we would like to see. Anything to publicise that would help. We have just had our standards for the wristbands signed off and endorsed by Martin Fletcher, the CEO of the NPSA. We hope that with the help of local Patient Safety Agency teams we can drive that adoption down through the individual trusts. I think it is about publicity and endorsement at the most high level. I have met Ann Keen and she is very supportive of our standards, and at the Patient Safety Conference in London last year she mentioned our standards in her opening speech. It is not for the want of trying; it is about just getting that message through.

Dr Taylor: Thank you.

Q287 Chairman: Roger, are you familiar with the Papworth and Wrexham pilots that have been taking place on barcoding?

Mr Lamb: I am not familiar with those particular ones, no.

Chairman: Right, we will move on.

Q288 Jim Dowd: It occurs to me that if there is to be barcoding of pharmaceuticals, would there need to be an international standard for that? Would it need to be the same the whole world over?

Mr Lamb: I think it would be and it is - and we have one. We have a way of identifying the drug uniquely. Proprietary coding systems and proprietary systems to try to identify things do not tend to work. It is a bit like having lockers in a changing room, where it is all well and good if you go into a changing room and you know which locker you put your clothes it because they are all numbered uniquely, but as soon as those lockers come out of those changing rooms and are sort of swimming around together you potentially could have numbers which are duplicated. You do need to have an international system, therefore, which ensures that everybody does it the same way, that everybody records the date the same way and the batch number the same way, so that everyone knows that when they are reading the number they know how to interpret it properly.

Q289 Jim Dowd: I do not know if you were here when I was at the questions of Dr Whitaker about the New Zealand experience. Was your organisation involved in that at all?

Mr Lamb: GS1 New Zealand has very close link with the New Zealand government. They have done some work with Bruce Anderson over there and they have recently published a report which showed that through using this technology in this area over the next 12 years over 1,000 lives could be saved as a result of misidentification. But it is not just about saving lives; it is about potential harm, both permanent and temporary, and the numbers there are getting into the tens of thousands. Clearly, if those numbers were extrapolated to the UK you could see that maybe as many as half a million people could be saved from some form of harm over the next 12 years if implementation took place.

Q290 Jim Dowd: Professor Murphy, blood transfusions and the dangers inherent there is where you have done most of your work. Could you briefly outline to the Committee the dangers that are inherent in blood transfusion and, also, how you have managed both to identify and deal with these by the use of technology?

Professor Murphy: There are a number of risks associated with blood transfusion. We are fortunate in the UK that we have a national incident reporting scheme called SHOT, the theory hazards of transfusion scheme. They produce an annual report outlining the serious events that have occurred in relation to blood transfusion over a year and they make recommendations in terms of improving patient safety. Over the ten or 11 years that they have been producing reports there have been over 4000 serious adverse events and 70% of them have been incorrect blood component transfused, so the patient getting blood that is the wrong blood for them, often intended for another patient. The most serious type of wrong blood transfusion is an ABO-incompatible red cell transfusion. There have been 213 of those in the 11 years which have caused 24 deaths and 107 cases of major morbidity - that is admission to ITU or renal failure. Those ABO-incompatible red cell transfusions are due to errors. They are entirely avoidable. There are other risks of blood transfusions, as I am sure you are aware: the transmission of infection, for example, and immune consequences of blood transfusion.

Q291 Jim Dowd: You are talking about contamination of the blood there?

Professor Murphy: Yes, that is right. In relation to transmission of infection there have been 60 events in the last 11 years. Those have mostly been bacterial contamination associated with platelet transfusions. Platelets are stored at room temperature, so the risk of bacterial contamination is higher. There has really only been a handful of cases of viral transmission: hepatitis, HIV. There have been four cases of transmission of variant Creutzfeldt-Jakob disease (variant CJD). We are mainly concerned today about the errors resulting in serious adverse events to transfusion and, as I have said, there have been over 200 of those in the last 11 years and 24 deaths. There are about 20 of these occurring every year, two deaths every year. These are the sorts of cases that end up on the front pages of local press, as some of you may have come across in your own parts of the country.

Q292 Jim Dowd: As, indeed, they deserve to.

Professor Murphy: Absolutely - because they are avoidable. The transfusion process is complicated. It starts with the collection of blood from a patient for blood grouping and cross-matching. A sample goes to the laboratory that carries out blood grouping/matching. The blood is put in a storage refrigerator for collection by one of the nursing or portering staff and then there is a checking process at the bedside to make sure that the blood is being given to the right patient. Errors can occur in all those steps and result in the wrong blood being transfused. SHOT documents those and has clearly shown that errors throughout all those steps can cause problems. There have been a number of initiatives to try to reduce these wrong transfusions. A number of bodies in the UK - such as SHOT itself, the National Patient Safety Agency, the National Blood Transfusion Committee, and the UK CMOs have run a Better Blood Transfusion initiative over the last ten or 11 years - have all made recommendations about improving transfusion safety, largely about improving training for the many staff who are involved in the transfusion process and also exploring the use of technology to make it easier for the staff to get the process right - not to rely on the technology alone, but to make it easier for the staff to get it right to do whatever part of the process they are doing absolutely correctly. Shall I go on now and describe what we have done in Oxford?

Q293 Jim Dowd: I suspect so. You have already answered four of my questions and hopefully you will have answered them all by the time you carry on.

Professor Murphy: Are there any questions about what I have described so far?

Q294 Jim Dowd: As we have established from previous witnesses, you cannot eradicate the possibility of error. You have reduced it, obviously, by the way you have adapted your system. Have you identified the area where the greatest risk now lies? How do you ensure that the right patient gets the right wristband, because everything else springs from that?

Professor Murphy: Maybe I should just describe the system that we have developed. This is not something that has happened overnight. We have been working on it for six and seven years and we carried out lots of pilots before we launched into implementing it throughout all the Oxford hospitals. We started off in a day case haematology ward. The first thing we had to do was to get a barcode on a patient's wristband, so we had to interact with the hospital IT department to ensure that we could have printed wristbands rather than written wristbands - the sort of printed wristband that Roger has here - with the patient's ID details on and a barcode. Rather than a linear barcode which will just encode one number, we used one of these dotty ----

Mr Lamb: Two-dimensional..

Professor Murphy: -- two-dimensional barcodes which can encode a patient's surname, first name, date of birth, gender, and hospital number or, indeed, NHS number - and we might come on to talk about the NHS number later. Basically, the nurse or the doctor taking the sample for blood transfusion has a handheld at the bedside and they first verbally identify the patient. The normal identification steps will happen but the patient will be first asked to state their surname, first name, date of birth, and that will be checked against the written details on the wristband. Then the sample is taken, put into the tube, the barcode is scanned on the wristband, the patient's details come up on the wristband. The handheld is just reading what is on the barcode. The nurse or doctor will again check that those details match the written details on the wristband and what the patient is telling them in terms of first name, surname, date of birth, and they will check the hospital number too. The sample goes to the laboratory. The laboratories have automated processes, so they scan the barcode on the tube and that enters the patient's information into the laboratory system. The testing is carried out and then the blood bag gets a sticky label on it, produced by the laboratory, the patient's name ----

Q295 Jim Dowd: Which way up is this label?

Professor Murphy: It is the same way up as described earlier, so it is great for the laboratory. Anaesthetists clearly have difficulty reading things upside down.

Q296 Jim Dowd: Perhaps they should just print them all sideways.

Professor Murphy: It is a fair point that we need to make it easier for users to do things right. If we are making it difficult by having it the wrong way round, then we should look at addressing that. We have an automated system for collecting blood from the blood fridge. The laboratory puts the blood in the blood fridge. What happens in terms of collection is that, when a patient needs a transfusion, the nurse will scan the patient's wristband, produce what we call a pick-up slip with the patient's ID and a wristband, and that will be either taken by the nurse or collected by a porter to go to a PC by the fridge. There is a process of scanning the barcode into the PC, the patient's details come up, the fridge opens, the blood is pulled out, scanned again to make sure that it matches, taken to the bedside, and then there is a checking process at the bedside both involving verbal ID and scanning the barcode on the wristband and the blood to make sure that everything matches. The handheld is prompting the nurse through each stage of the process. That is key. When we carried out our work, the first thing we did was to carry out some observations of the current practice and watched what nurses were doing in terms of identifying patients and bedside checking. We worked out that the old process involved 27 critical steps. It involved two nurses, two wristbands in Oxford (we had an additional identification with a number for the sample which followed the process around called a Red Label System - so they had two wristbands) and we found that they only did this process correctly in 10% of cases. After we had implemented the system, we got it right. They were doing it right every time.

Q297 Jim Dowd: That is a significant improvement. Was it particularly expensive, the system you have now compared to what it was, and did you have any difficulty in persuading the trust to find the money?

Professor Murphy: Good question. We started off in a small way, carrying out pilots with research monies and we built up our experience and demonstrated that this system did work. We got feedback from the staff, tried to customise it so that it was good for the staff to use, and got feedback from the patients as well, and we gradually rolled it out from a day case haematology unit, with non-urgent transfusions, to more complex clinical scenarios like cardiac surgery intensive care. We built up our experience, mostly with research monies, and we highlighted what we did by going and presenting it at meetings, publishing our work, getting more feedback, and built up our experience over about five years before we felt that we would have a robust business case to take to the trust to say, "This is really going to be beneficial for the staff, it will save staff time, it will make it easier for the staff to get things right, it will improve patient safety, and by greater awareness of all aspects of blood transfusion make a cost saving also in terms of reduced blood usage." The trusts were well aware of what we were doing because of what we had done to raise awareness about our work over the previous five years and they agreed to support us. The cost was £1.5 million over five years for the equipment and for the training. The implementation throughout all the Oxford hospitals, about 100 clinical areas over three acute hospitals sites, took just over a year. It was completed at the end of 2007. As well as maintaining the equipment now, we have four members of staff who troubleshoot, maintain the equipment, and carry out training.

Q298 Jim Dowd: If it increased the accuracy rate from 10% to virtually 100%, the failure rate under the old process must have had a cost somewhere. Other than specifically to patients, it must have had a material cost to the hospital.

Professor Murphy: We estimated, with very rough estimates in terms of saving staff time, that we could save £0.5 million a year in terms of nursing time. That is working out how much time it would save and then costing that up. Nurses can be very busy on wards and I do not believe that the trust has employed less nurses because of the work that we have done.

Q299 Jim Dowd: They are off doing other things.

Professor Murphy: They are doing other things. It does save them time and they are absolutely confident that they are doing it right.

Q300 Jim Dowd: Could I just say that I think I have cracked this label business. With the sophistication of modern labelling machines, you could print on a single label one each way up. That would suit everybody.

Professor Murphy: Yes. I think we would be reluctant to put it on both sides of the bag because one of the checks involves making ----

Jim Dowd: Make two labels then, and stick one on each side.

Sandra Gidley: That was my idea.

Jim Dowd: Put one upside down and one the other way round, because clearly they are put that way up to suit the people who store the blood rather than what they are going to be used for.

Q301 Sandra Gidley: Obviously you have a very close, ongoing relationship with the trust, so it was relatively easy to persuade them of the benefits of your system, but is anything being done to ensure that the system is adopted more widely in the NHS?

Professor Murphy: We did write a national specification. We were given some money from Connecting for Health to write a national specification for blood transfusion using the electronic system. That is being trialled now at Mayday Hospital in a study funded by Connecting for Health.

Q302 Sandra Gidley: Why does it have to be trialled if you have been doing this for years? Is this yet another way of delaying things?

Professor Murphy: It certainly has held things back. For my sins, I am chairing the steering committee of this pilot at Mayday and I think we have all been incredibly frustrated about how slow it has been. The trust there has wanted to add in further refinements of this process.

Q303 Sandra Gidley: Can you expand on that? Do they know what they are doing or are they justifying their jobs? You may like to be a little more diplomatic in your reply.

Professor Murphy: Do I think they are doing it ----

Q304 Sandra Gidley: My understanding of a trial is that you do not just adapt these things willy-nilly. You have been doing it for some time, you have considerable expertise. Why do some people think they know better?

Professor Murphy: I share your frustration. Even as chair of the steering group I keep asking the same question. The pilot is overdue and there is no doubt that that is holding back some trusts from adopting the very basic systems.

Q305 Sandra Gidley: The other trusts are being stopped from adopting a system you have developed over a period of five years - you have written the spec. They are being stopped from potentially buying something they want to buy or investing in something in which they want to invest because on this pilot there are people who think they know better and want to change it.

Professor Murphy: They are not being actively stopped. All the trusts in the country are very aware of this pilot and it is being promoted as something that needs to be done before a national roll-out can be considered and so it may be putting off a lot of trusts from moving forward on implementing the system that we have developed.

Q306 Sandra Gidley: Who makes these decisions in the trusts?

Professor Murphy: The decision-making is individual. It is trust by trust. That is obviously a problem, in that the implementation of this system that is going on now is piecemeal, it depends on individual decision-making in each trust.

Q307 Sandra Gidley: Do you have people working in the blood service who want to adopt this but are being stopped at management level for whatever reason, or is there just a caution at the level of implementation and are there any other reasons for that apart from the delay of the pilot?

Professor Murphy: It is the latter. They are not being actively discouraged, but waiting for the results of this further pilot may be being used as a bit of an excuse for holding back.

Q308 Sandra Gidley: There are not any financial barriers.

Professor Murphy: Yes, there are huge financial barriers. As I was saying earlier, I described the cost to the Oxford hospitals. Finance for the equipment is one; finance for the training is another. There are IT connectivity problems within hospitals, between patient administration systems and laboratory systems and other systems within hospitals. We would like to use the NHS number, but we have problems with that, in that we cannot ----

Q309 Sandra Gidley: You said you wanted to talk about the NHS number.

Professor Murphy: I do not know whether this is ----

Q310 Sandra Gidley: If you are having problems with it and it is safety issue, that probably is something. I would like to think that I am a name not a number, but .... I gather we all have two NHS numbers now and it is all very confusing.

Professor Murphy: That is right. That is part of the problem. Sometimes we cannot get hold of an NHS number quickly enough, so that we would like to use an NHS number but we cannot get hold of it at the time we want to use it in 17% of patients. An NHS number has lots of advantages in terms of patients travelling between different hospitals and sharing information between different hospitals. We could very easily use it for our system instead of the hospital number but we cannot get it in 17% of patients. It would be very difficult to run two systems side by side, so for the moment we are not pushing on with the use of the NHS number, with putting it into the bar code for patient identification in Oxford.

Q311 Dr Naysmith: Before we leave the subject of blood transfusion, Professor Murphy, it is sometimes said that the use of a fresh and stored blood can be replaced by artificial substitutes in some areas, and that in many respects could be safer. Is there any mileage in that sort of area of research in saving the barcoding of individuals?

Professor Murphy: It is a great idea. There has been a huge amount of money spent on it, particularly in the United States, but it is one of those things in my career that always seems to be five years for the future, and five years on it is still five years away. At the moment it seems ten years away, because there are a lot of concerns of morbidity and mortality associated with the products which have been trialled in the recent past. So it is a long way away.

Q312 Dr Naysmith: It is still a long way off.

Professor Murphy: But you make a very important point about trying to ensure that patients only receive blood when they really need it, and that is a really important patient safety measure. There has been a huge amount of work through the UK CMOs Better Blood Transfusion Initiative, the National Blood Transfusion Committee and so on, to reduce the inappropriate use of blood. The use of red cells in England has reduced by about 20% in the last five or six years. As well as being great for patient safety, that has also resulted in a huge cost saving.

Dr Naysmith: Thank you very much.

Chairman: Bryony, you will be pleased to know we are going to move on to you.

Q313 Dr Stoate: We have heard a lot this morning about the uses of technology and we have heard great stories about how technology can benefit patients and reduce safety issues. Your memorandum, however, says that the research base and the evidence of technology is rather limited. Whereabouts are the limitations and what do we need to do about them?

Professor Franklin: I think there are several areas. Many of the technologies that are used within the medication field have been developed and tested in other countries. Particularly in the United States, where the systems for prescribing, dispensing, and administering are very different from those in the UK for many reasons, we cannot automatically assume that something that has benefits in the United States has benefits in the UK. One of the issues, therefore, is that we have often very little UK data and where systems have been tested and designed in a very different health system it is really important that we know if it works in the NHS, which is a very different health system. That is part of the issue. In particular, we have very little UK data on things like smart infusion pumps, which is a technology that is being widely advocated. We have little data on the bar code verification of medication at the point of administration - there is a very small amount but not very much data. We have slightly more data on the impact of electronic prescribing systems, but it is still very much in its infancy. There are even fewer studies promoting technologies used in primary care. I think we do need an evidence base for these products. They are expensive, they are very time-consuming. We would expect to have an evidence base for a medication that we were going to use in a patient and, in a similar way, I think we need an evidence base for a technology which is going to significantly change the way we use things. Some of those technologies have a huge benefits, others are neutral and cost a lot of money, and others might even cause patient harm if they are not used correctly. It is not a magic bullet; it is a very complex intervention and it depends how it is used. As well as the robust research evidence, we also need perhaps a toolkit of methods that trusts can use to evaluate things as they introduce them, that is perhaps not quite as robust in terms of the research but allows trusts to assess things that they can compare with other trusts. They can check that it is having benefits in their own organisation. We also have an absence of economic evaluation. Some of these systems, particularly trust-wide electronic prescribing systems, cost an enormous amount in terms of the capital, the training burden. We have already heard about the difficulties of training staff who might work one shift a week, only nights, locums, et cetera. The training burden is a huge as well as the ongoing maintenance and upgrade, so we need an economic evaluation of the benefits versus the costs. We also need a better understanding of the relationship between error and harm. We know that a many, many errors occur. Fortunately, the vast majority do not cause patient harm, however, they can affect the patient's confidence in the health system. I am sure we all hear relatives saying, "The pharmacist gave me 29 tablets instead of 28, so how do I know he has given me the right drug?" That is not going to cause the patient harm, but it does affect their confidence. On the other hand, you could say, "We should put the most resource where the harm is." That relationship between error and harm is not yet totally understood. Which are the errors that have the most potential for harm? In some cases we know specific drugs - vincristine administered incorrectly, methotrexate administered daily - but once you take those away from the others we do not really yet know what are the characteristics of an error that make it most likely to cause patient harm. Really we need an understanding of what the problem is, so that the problem drives the solution rather than the technology driving the solution, if that makes sense.

Q314 Dr Stoate: Yes, it does make sense. This morning we have heard huge amounts of evidence, virtually all of it about secondary care and about really high-tech stuff, in terms of radiology and very high-tech blood transfusion checking systems. You are the first person to have mentioned the words primary care, and even then you only did it very briefly. The vast majority of drug prescribing goes on in primary care, yet we have had evidence from the National Patient Safety Agency and others that we know virtually nothing about medication incidents and problems in primary care. What do we need to do about that?

Professor Franklin: You are absolutely right: 80% of drug use is within primary care and yet we have much less data. I think that will change soon. Certainly I am involved with several research consortia who have been studying errors in primary care. We have just completed a multi-centre study in the UK of errors in care homes. It was a Department of Health funded study and reports from that will be published shortly. We are also involved in a very large study evaluating the impact of the electronic transmission of prescriptions in primary care and the impact that has on safety, doctors' time, pharmacists' time, what do patients think of it, so the evidence base is growing but it is a long way behind secondary care. In the same way, we need evidence for the benefits or otherwise of technologies. Do they work? Do they create new problems? How can we introduce them to best benefit?

Q315 Dr Stoate: You are right about electronic transfer of prescriptions, even though that is still in its infancy and the vast majority of GPs cannot yet do it, but even that will only reduce dispensing errors, it will not reduce prescription errors and GPs simply making wrong diagnoses, issuing wrong treatments. There is currently virtually nothing to prevent a GP prescribing exactly what they want, entirely on their own, and with nobody checking them at all.

Professor Franklin: Most GPs have electronic prescribing systems. It is not my area of expertise but I think they vary in terms of the amount of decision support they have, so there is clinical decision support there. In fact that is one area where primary care is vastly ahead of secondary care, because most GPs do prescribe using computerised prescribing systems with decision support.

Q316 Dr Stoate: It does not give you decision support, but it gives you drug interaction support.

Professor Franklin: It does, yes.

Q317 Dr Stoate: The one thing GP computers do is come up automatically with drug interactions, but they do not otherwise say, "That drug is not indicated for hypertension" for example.

Professor Franklin: I do not believe so. I think some systems might have the opportunity to do that but you would then have to have the codes for that patient's diagnosis.

Q318 Dr Stoate: Where do we need to go in terms of sorting some of this stuff out?

Professor Franklin: Diagnosis is not my area, so I would not feel I could comment on that. Certainly the prescribing systems in primary care do have more advanced decision support in terms of drug interactions, dosage checking, et cetera, than those that we currently use in secondary care.

Q319 Dr Stoate: That is true, but, given the thinness of the evidence base which you have mentioned now and in your evidence, do you think that the department might be rushing things a bit in its automatic identification and data capture system recommended in Coding for Success? Do you think we have gone ahead of ourselves, given the lack of knowledge we have on what is currently happening?

Professor Franklin: No, I do not actually. I think a uniform system for identifying products is an essential prerequisite for many of the systems that will come later. Many of those systems that can come later have enormous potential, and we are not going to get anywhere with those unless we have a uniform coding system. I think that is essential. Indeed, the Coding for Success report includes data from several systems that we have introduced and tested at Charing Cross and other hospitals, at Imperial, which do show benefit, so obviously we are in favour of that. We think these systems do have huge potential, but they have to be introduced properly. Mike has given a great example of a system that was piloted, evaluated. You cannot just put these things in and expect to see a magic result overnight.

Dr Stoate: Thank you very much.

Q320 Charlotte Atkins: You state that technology that improves patient safety in other countries might not work in the NHS. Why is that?

Professor Franklin: It is mainly due to the different systems that are used in different countries. One example might be that in the United States in hospitals it would be custom and practice that many of the prescribers are not based within the hospital, they are based within their own offices, and they prescribe for their patients that are in the hospital. They would therefore give verbal orders over the telephone for a prescription to be written that would be written down by a nurse on a piece of paper, it would be faxed to the pharmacy and the pharmacist would enter it into a computer system, so it is perhaps not surprising that a computerised position order entry system, allowing the doctor to enter that prescription directly into a computer from his or her office, would have enormous benefits in terms of efficiency and safety. In the UK, on the other hand, verbal orders are generally banned, doctors are prescribing from the hospital directly on to a specially formatted drug chart at the end of the patient's bed, and the benefits are less clear. There are also, in some situations, practices that we have in the UK that we do not have elsewhere. In the UK it is common practice that when a patient is discharged from hospital we supply several weeks of medication to keep them going until they get to see their GP. That is not the case in many hospitals in the United States or in many European countries, so systems that are designed elsewhere will not have the facility for putting in discharge prescriptions, which is one of the main benefits that we might be able to use in the UK setting. It is to do with differences in culture and differences in practice. In the United States, instead of sending whole boxes of medication to the ward, they have individually dispensed doses, mainly due to the requirement for individualised patient billing and a greater risk of theft, because it is a private healthcare system. That is very low in the UK. There are all sorts of different driving forces that mean their systems are very different, so something that benefits their system will not necessarily benefit our system and vice versa.

Q321 Charlotte Atkins: It is not so much that there is a fault within the NHS.

Professor Franklin: No.

Q322 Charlotte Atkins: It is the fact that we have different cultures and different systems. Presumably the principles are still the same of ensuring that there is verification, that verbal instructions are not written down incorrectly, that the person you are treating is the person you are treating, and that you are not giving them the wrong drug. Certainly, where we have seen electronic patient records work and where the clinicians are willing to accept the system and use it, it does seem to cut down on error and cut down on bureaucracy as well.

Professor Franklin: That is right. The principles are the same, we are all trying to achieve the same endpoint, but something that works in one context will not necessarily have the same benefits in another. You might see a system that has reduced the error rate from 20% to 5% but if the error rate in the UK was currently 4% we might not have anywhere to improve. There are lots of different factors that to take into account.

Q323 Charlotte Atkins: Which countries do you think we can learn the most from in terms of the problems we are facing being similar to what other countries are facing? Which systems are closest to the UK?

Professor Franklin: In terms of the systems for prescribing, dispensing, and administration in hospital, which is my main area of experience, New Zealand and Australia are the nearest to the UK system. Their systems are very similar to ours. The system used within mainland Europe is very different and the system used in America is very different again, so you cannot really have much to learn between those three.

Q324 Charlotte Atkins: Is it for historical reasons that Australia and New Zealand are similar? Why has that developed in a similar way?

Professor Franklin: It is partly historical. The differences in the US have often been driven by the fact that it is a private healthcare system growth. It is much more similar in terms of a national health system in Australia and New Zealand. It is to do with the different driving forces. In the UK we do not need to know how much every single patient costs us in terms of their medication at the moment, whereas that is something that has been needed for a long time in the US - you know, every single paracetamol tablet has to be accounted for and charged.

Charlotte Atkins: Thank you very much.

Q325 Dr Naysmith: Professor, Franklin, in your memorandum you listed various ways in which technology can compromise patient safety. Could you elaborate on this, please.

Professor Franklin: There are several things that I listed in my submission. To elaborate on some of those a bit more, one is over-reliance on technology. For example, if we are all used to reading the barcode on a drug you might then not check with your eyes the strength and details of that drug, and there is always the possibility that the barcode might have been put on incorrectly, particularly if it was done at a later stage of over-labelling, so you remove then that basic check of: "Have I got the right drug and the right dose?" That leads to deskilling. We hear about doctors in primary care who are now very unwilling to prescribe in a patient's home or at a care home because they do not have the computer system in front of them with the interactions. As a pharmacist, when I first qualified, I knew all of the additional instructions - which medicines had to be with food, after food, before food - off the top of my head. Now I know that the computer system puts all those in, so on the rare occasion that a computer goes down and we have to typewrite labels, we all have to look at every single label. In some ways that is inevitable but it is something that we just need to recognise and be aware of. Another example is where one of the systems we have tried, an electronic prescribing system, actually had very little clinical decision support, so it did not include allergy checking, and although we had made that clear to the prescribers, we still found doctors who would make errors due to not checking a patient's allergy status and prescribing something they were allergic to because they kind of assumed, almost subconsciously, that the computer would do it: "The computer must be right. The computer must know this." Actually the computer only does certain things. It cannot replace the human brain and the human cognition. We also see development of workarounds - and I think several people have referred to workarounds this morning. An example we have seen is relating to barcodes, where we have had a system where patients are identified by a barcode, but sometimes a barcode around a wrist will not scan, perhaps because it is curved or it has got damaged in the shower, or whatever else, so we find nurses printing out duplicate barcodes and sticking them to the patient's table because they scan beautifully on a hard, wooden, flat surface, but then, if the patient moves beds, you are aiming to find the table and not the patient. You then have an enormous potential for risk. Even though overall the system, we believe, did improve patient safety, you have the potential for these issues if staff perceive that the system does not work or it is frustrating because they cannot get the barcode to scan, and you then get these workarounds. We also have new types of errors, because sometimes an electronic system or a technology system will require selection from pull-down menus, and I think we all know how easy it is to select the wrong email from your Blackberry or the wrong file name when opening a Word document. In the same way, when presented with an enormous list of drugs it is quite easy to select the wrong one, particularly if a prescriber is presented with a much longer list of drugs than he or she is otherwise familiar with. For example, if you type "beclomet" in a drug list, you get a long list of drugs because there are lots of drugs that start with that prefix. You also get additional steps put into wordprocessors. Again using the example of electronic prescribing, when a doctor prescribes on paper he or she really just has to prescribe the name of the drug, the dose and the route: Aspirin, 150 mg, by mouth, once daily. With many electronic systems, you then have to pick the specific tablet. Aspirin does not come as 150 mg, so you have to pick either two 75 mg or half of a 300 mg, and suddenly the prescriber has to make the choice between those. That is a very straightforward example but you can imagine much more complex situations where the prescriber is having to make a choice they did not previously have to make and that is because the system needs to know the exact tablet because it is perhaps linked to a barcode verification at the point of administration. No longer can you leave the nurse, therefore, to make that decision; you have to specify it, so there are additional steps and any additional step can introduce new errors. Finally, you can remove some steps. An electronically prescribing system that allows you to prescribe without seeing the patient means that you might miss vital information (for example, that the patient has a feeding tube in and cannot swallow a tablet, or that they are bright yellow with jaundice). You miss what is really important clinical information about the patient because you can now do it without seeing the patient. Those are just some examples.

Q326 Dr Naysmith: That is quite a broad range of different types of examples. Is there any broad conclusion you can come to in trying to say how we are going to deal with this new thing that is being introduced into treatment.

Professor Franklin: I think the main thing is awareness of these things. Certainly with the systems that we have studied, both evaluating their impact and also studying their implementation and how they use them, a lot of it is about being aware of these things, so perhaps if a hospital is putting in a system that has been used elsewhere, go along and see what lessons they have learned. Do not assume that you can just put it in and it will work, because it will not. It requires an enormous change in work patterns. There is a growing area called the socio-technical aspects, which is to do with how humans and work systems work together, and how the technology will change people's work patterns and the social interactions, and it is important to study those things. I really believe that, overall, these systems have enormous potential benefit but realising these benefits means paying close attention to all these issues and monitoring it, introducing it slowly, evaluating it, rolling out, as we have heard with Mike's example.

Dr Naysmith: Thank you very much.

Q327 Chairman: Roger, is the barcoding of medicines a potential protection against fake medicines, or is barcoding easy to fake or replicate?

Mr Lamb: You can copy a barcode, certainly. It is quite easy to do but there are other ways that you can protect against that. For instance, you could put a randomised serial number in there, which you can store in a two dimensional barcode more easily than a mini barcode, and then you could check against that randomised serial number with a manufacturer's database to ensure that this was a number that should be available to be used at that time. If it did not match a number that had not already been used or a number that did not exist, then you would know that it was potentially a fake. You can use encryption and those kinds of things in the number to protect against that, so there are ways of staying ahead of the counterfeiters, yes.

Q328 Chairman: Is that used in other walks of barcoding, not just medicines. Is it used elsewhere in the world?

Mr Lamb: It is something that is also being looked at in the aerospace industry. Clearly counterfeiting is not limited to medicines. Bus crashes and plane crashes can have counterfeit parts in them.

Q329 Chairman: When you say it has been looked at, has it been implemented?

Mr Lamb: It has not been implemented yet, no. There is some work that is going on in that area but it does involve putting together an infrastructure. Every manufacturer has to join the club, effectively, and agree to put all of their medicines up on a database, and the links from there to the manufacturer's own database to check serial numbers have to be there and the technology has to be in place, for instance, at the pharmacy or wherever you are having the check being done. We are doing some work around global traceability standards. Our standards are not cast in stone. Our standards have been developing for 35 years. Our standards are designed for our members and the NHS is a member. The standards are designed by members. We get together with manufacturers and distributors, wholesalers and industry associations, the hospitals, the pharmacies, and regulators. We all agree what is best for the supply chain within healthcare, how can we all work together to ensure that these things will work, and that the interoperability and compatibility is there and everybody gets out of it what they want.

Chairman: Could I thank you all very much for coming along and assisting us with this area this morning. Thank you.