UNCORRECTED TRANSCRIPT OF ORAL EVIDENCE To be published as HC 831-i

House of COMMONS

MINUTES OF EVIDENCE

TAKEN BEFORE

THE COMMITTEE OF PUBLIC ACCOUNTS

MONDAY 16 JANUARY 2006

 

A SAFER PLACE FOR PATIENTS:

LEARNING TO IMPROVE PATIENT SAFETY

 

 

DEPARTMENT OF HEALTH

SIR NIGEL CRISP KCB and PROFESSOR SIR LIAM DONALDSON KCB

 

NATIONAL PATIENT SAFETY AGENCY

MS SUSAN WILLIAMS

Evidence heard in Public Questions 1 - 161

 

USE OF THE TRANSCRIPT

1.	This is an uncorrected transcript of evidence taken in public and reported to the House. The transcript has been placed on the internet on the authority of the Committee, and copies have been made available by the Vote Office for the use of Members and others.
2.	Any public use of, or reference to, the contents should make clear that neither witnesses nor Members have had the opportunity to correct the record. The transcript is not yet an approved formal record of these proceedings.
3.	Members who receive this for the purpose of correcting questions addressed by them to witnesses are asked to send corrections to the Committee Assistant.
4.	Prospective witnesses may receive this in preparation for any written or oral evidence they may in due course give to the Committee.

 

Oral evidence

Taken before the Committee of Public Accounts

on Monday 16 January 2006

Members present:

Mr Edward Leigh, in the Chair

Mr Richard Bacon

Greg Clark

Mr Sadiq Khan

Mr Austin Mitchell

Jon Trickett

Kitty Ussher

Mr Alan Williams

________________

Sir John Bourn KCB, Comptroller and Auditor General and Mr Tim Burr, Deputy Comptroller and Auditor General, National Audit Office, gave evidence.

Mr Marius Gallaher, Second Treasury Officer of Accounts, HM Treasury, gave evidence.

REPORT BY THE COMPTROLLER AND AUDITOR GENERAL

A SAFER PLACE FOR PATIENTS:

LEARNING TO IMPROVE PATIENT SAFETY (HC 456)

 

Examination of Witnesses

Witnesses: Sir Nigel Crisp KCB, Permanent Secretary and Chief Executive of the NHS, Professor Sir Liam Donaldson KCB, Chief Medical Officer, the Department of Health; and Ms Susan Williams, Joint Chief Executive, National Patient Safety Agency, gave evidence.

 

Q1 Chairman: Welcome to the Committee of Public Accounts where today we are looking at the Comptroller and Auditor General's report on A Safer Place for Patients: Learning to Improve Patient Safety. We are joined once again by Sir Nigel Crisp, who is the Permanent Secretary and Chief Executive of the NHS. Would you like to introduce your team as usual, Sir Nigel?

Sir Nigel Crisp: Thank you, Chairman. Firstly, the Chief Medical Officer, Sir Liam Donaldson. We are also joined by Susan Williams, who is the Joint Chief Executive of the National Patient Safety Agency.

Q2 Chairman: Thank you very much. Perhaps we can start by getting an idea of the scale of the problem. The relevant figure is figure one, which you can find on page one. There is also reference in paragraphs 2.10 and 2.11 on page 27. You have got rather a wide range. The NAO estimate that at least 2,000 patients died in 2004-05 as a result of a safety incident. The NHS incidents range from 840 to this figure of 72,000 which you can find at paragraph 2.10. How can you expect to address this problem when these estimates range so widely?

Sir Nigel Crisp: Obviously that is an extremely fair point and it is why so much effort has been put into the last few years to get a much better understanding of what is actually happening, both in terms of the number of incidents and the type of incidents. I can say that the current rate is we are getting about 60,000 reports a month now, which is an update on these figures just for the last month. Also, literally now, we have got 100% of NHS organisations in England and Wales as part of the reporting system. We are bringing the reporting system together but as you know, and as the report makes clear, there are cultural issues here about making sure that people report things and how they report them and it is very important that we get the definitions right. What I think I am saying in response to your question is this is the latest stage of knowledge, it is moving quite fast, we are getting more information, and we will be in a better position in due course to get a closer picture on it.

Q3 Chairman: We can pursue that during the afternoon. Can we get an idea of international comparisons. This is dealt with on page 67 of the report, appendix four. We have got a high and rising incidence of unintentional harm to patients. Looking at these figures, if you compare the figure for London, England, particularly with some states in America and in Canada, I wonder whether the NHS is becoming a less safe place than some other systems.

Sir Nigel Crisp: May I just make one or two comments and then perhaps I could turn to Sir Liam who is much more expert on this. One observes from that table the range of different years that relate to that. What we do know from the National Patient Reporting System that we have got is that we are now seeing as many reports proportionately as the best in the world, so we are getting a figure that is as good as the best in the world and maybe slightly ahead of them. We are getting a much better understanding of the issues and we have got the data. Also, we have got a truly national approach to patient safety which I think is not true elsewhere. The architect of much of that was actually Sir Liam and I think it would be useful if he could say something about the international comparisons.

Q4 Chairman: Certainly.

Professor Sir Liam Donaldson: Thank you, Chairman. A lot of the data are derived from surveys rather than from established reporting systems and, on the whole, they have been surveys of medical records carried out in hospitals. They have been undertaken using broadly similar methodologies. A lot of the earlier studies showed lower rates of prevalence of what is usually internationally called medical error. Experts that I have talked to, and I have talked to them very extensively worldwide, use a ballpark figure of 10% of hospital admissions.

Q5 Chairman: So you reckon we are about average?

Professor Sir Liam Donaldson: I do indeed. If you want to go into it in more depth, I think there are some reasons why some of the earlier US figures ----

Q6 Chairman: I think we need to have a note on that so we can get it right in our report. Obviously things have moved on. Sir Nigel, you are spending about £15 million annually on the National Patient Safety Agency, is that right?

Sir Nigel Crisp: Something of that order, yes.

Q7 Chairman: Just give me four concrete achievements of this Agency so that we can have an idea of whether it is providing us with value for money.

Sir Nigel Crisp: I think the first one is the National Reporting System which is now receiving around 60,000 reports a month. The second area would be if I can take two different patient safety solutions that have come from there. One was related to the infusion devices which was identified as a problem but since then safer practice notices were issued and opportunities for high risk identifying of the infusion pumps have been reduced by half and there are some significant, but we cannot precisely quantify yet, cost savings that are coming from that. Another one would be the alert on the safe storage and handling of potassium chloride, which was implemented across the NHS.

Q8 Chairman: You mentioned that but I am told, in fact, that warning was given four years after an alert was issued in the US about that particular issue of potassium chloride.

Sir Nigel Crisp: Can I defer to one of my colleagues to make a response to that.

Professor Sir Liam Donaldson: There have been alerts issued in other countries and some of them did precede the NPSA's alert. The NPSA's alert arose directly from analysis of the data coming from the pilot phase.

Q9 Chairman: Complete your answer, Sir Nigel.

Sir Nigel Crisp: If I mention two others plus one general point. The 'cleanyourhands' safety alert and campaign has begun to have a really significant impact now. That has also come from there. The other point that is here is, as the report itself says, there is an improving safety culture in the NHS. It needs to improve further, we would be absolutely clear about that from the Department, but the NPSA has played a significant role in that.

The Committee suspended from 4.39pm to 4.45pm for a division in the House.

Q10 Chairman: Can we now look at the National Reporting and Learning System, if you could look at paragraph 2.34 on page 35.

Sir Nigel Crisp: Yes.

Q11 Chairman: It took two years longer to set up, did it not, and it cost over £1 million more than planned. Was this value for money?

Sir Nigel Crisp: The reason for the delay, which was two years, as you say, was that it was piloted, so there was a pilot system which was put in, which was a genuine pilot, from which some learning was taken but it proved to be unsatisfactory and, therefore, there was a second procurement made which led to the system we have got now. If I can put it like this, it was actually an attempt to make sure that we were getting quality and value for money that led to the delay, making sure we worked it through. You are right that the original business plan was £9 million and something on a lifetime cost and the revised business plan, which came in as a result of some further review, took it up to £10 million and something, so approximately £1 million. It was worked through very clearly on both of those business plans with Treasury signing off and so on.

Q12 Chairman: There are certain parts that are worrying. For instance, an anonymous e-reporting system was set up, and that is mentioned in paragraph 2.36, but I was told that you have only received 108 emails, is that right?

Sir Nigel Crisp: Could I hand that one to Sue Williams.

Q13 Chairman: 108 emails nationally is nothing, it is what we get as individual Members of Parliament every other day.

Ms Williams: We launched the www service later than the main reporting system as we wanted to try to make sure that as many trusts were connected so we did not undermine trust reporting. We now have got 2,914 reports in from that route. Very interestingly, 9% of those reports are coming in now from medical staff. That is a higher percentage from normal reporting systems. It is clearly attracting a group of staff as a safety net.

Sir Nigel Crisp: Can I just add to that. There are two points. One is this was a safeguard system to make sure that we could pick up whistle blowing, if you like, the people who might not wish to report through the standard system. Secondly, the relatively small number is probably a reasonable indication that people are willing to use the main system. That is the thinking behind that and, as you see, it has now picked up in terms of numbers.

Q14 Chairman: In your first answer to me you said that all the trusts are now using the National Reporting and Learning System but at the time this report was written 35 trusts were still not using it. That is right, is it not?

Sir Nigel Crisp: Yes.

Q15 Chairman: So obviously something has happened. What has been happening? Have you been running around desperately to make sure that all the trusts are using it in recent months?

Sir Nigel Crisp: Again, can I pass that to the NPSA.

Ms Williams: We have a team of 28 patient safety managers who work right across the country and they have been working with local trusts to encourage their connection to the system and the receipt of reports.

Q16 Chairman: I wonder whether that has had anything to do with the fact that this hearing is taking place. Here is this report being published on Thursday 3 November, 35 trusts were not using it, and suddenly here we are meeting in the middle of January and they are all using it. That is very good, we like to feel we have some influence.

Sir Nigel Crisp: May I just say, that does mean it was 95% before, so it was very sensible of us to have wiped up the 5%.

Q17 Chairman: Fair enough. We had a PAC recommendation in December 2003 that the work of the National Clinical Assessment Service be extended to cover nurses and other clinical staff. The Treasury Minutes said that was accepted by the Government but nothing has happened yet. When we make these recommendations and they are accepted by the Government it is nice if they are carried out.

Sir Nigel Crisp: I might ask Sir Liam to step in in a moment. Where we are is this is a much bigger group of staff, about half a million staff, as opposed to the relatively smaller number of doctors. We have a group led by the Chief Nursing Officer which is working on this with the NPSA to try and develop a system which will allow for issues more to be dealt with locally than nationally. What we do not want to do is to draw every problem up nationally but to make sure that we have got the right principles being applied locally that where there are issues of individual staff ----

Q18 Chairman: So this is going to happen?

Sir Nigel Crisp: Something will come out in the spring.

Q19 Chairman: If you look at page three, this is the stakeholders in patient safety. If you look at that, Sir Nigel, at these very large numbers. This is figure two, page three. I just wonder whether you are making enough progress in rationalising matters. We have this very high level of incidents, although there is some doubt about that, we have this extraordinarily complex system, whereas what we really want as patients is simply zero tolerance on this issue, is it not, and a lack of confusion?

Sir Nigel Crisp: I completely agree, what that shows is there are a lot of stakeholders. The biggest issue on that is are they all making demands on the system to get reporting, which is picked up elsewhere in the document later on. What we are trying to do at the moment, and I know that the NPSA has got somebody seconded to our Connected for Health programme, is to create a single portal for people to report so that broadly the same information that somebody may be reporting in a local primary care setting or acute hospital is made available to the relevant stakeholders. If you look at that list clearly there are people with very clear statutory responsibilities, like the coroner and so on. There will not be a complete ability to share it but we think we can do better than we are doing at the moment and that is what we are trying to do.

Q20 Chairman: Lastly, Sir Nigel, did you see the front page of the Daily Telegraph today? "£235,000 lavished on slipper safety advice. A £235,000 scheme advising the elderly on how to wear slippers is among the array of examples of wasteful government spending in this area. They include £40,000 spending by the NHS on a 46-word patient experience definition that required two £8,000 workshops, £4,000 of public meetings, two £1,600 meetings with children, three £600 in-depth interviews. Among the aspirations established by the exercise was that patients wished to be treated with both honesty, respect and dignity", which we would have hoped would be obvious.

Sir Nigel Crisp: I have not seen the Daily Telegraph.

Q21 Chairman: On slipper safety.

Sir Nigel Crisp: I do note from the figures in this report that the biggest area of reports of incidents is people falling. I cannot remember where it is in the document. It is something like 40% are about falls. We do know that in areas where people have set up what are called falls clinics - they are as simple as that - we are seeing some improvement in the reduction of fractured femurs and so on. Whether that particular expenditure was justified or not, reducing falls amongst elder people is undoubtedly justified.

Q22 Mr Khan: Can I say how reassured I am that you do not read the Daily Telegraph. I am pleased to hear that. I am afraid I have got to leave shortly for another meeting, so apologies, no discourtesy is intended. My first question is do you think it is realistic and reasonable for one in ten patients who are admitted to NHS hospitals to be unintentionally harmed?

Sir Nigel Crisp: No, I do not. As a patient, this is something I want to know that we take extremely seriously.

Q23 Mr Khan: What are the experiences of other countries?

Sir Nigel Crisp: Broadly similar. I think it is worth drawing out, as it does in this report, that two-thirds of that 10% do not experience any harm, so we are lumping everything together in that. Again, Sir Liam is the international expert on this and he could talk more about that.

Q24 Mr Khan: What I am interested in is 10% is not realistic and it should be lower - the obvious thing is to say it should be zero of course - if you were to come back here in a year's time or 24 months' time, what would that figure be?

Sir Nigel Crisp: I doubt in that sort of period that it would shift, but I do not like using an argument that we are just as bad as everyone else.

Q25 Mr Khan: How long do patients have to wait before they see an improvement that is noticeable?

Sir Nigel Crisp: You will see certain categories where we have got improvements. I could have said in response to an earlier comment that the National Patient Safety Agency has sent out 15 safety alerts in the last three years about specific devices or use of drugs or whatever; and that compares with ten in America over the same period, for example. We are actively trying to focus down on individual issues and make sure that those particular sorts of incidents never reoccur. We have got the IT system. A lot of these incidents were about slightly wrong levels of medication and when we are not relying on people's handwriting we will get better at that as well, so you will see some changes.

Q26 Mr Khan: How soon before we see changes on the ground?

Professor Sir Liam Donaldson: Could I add a comment? I think in most developed countries the ballpark figure is similar. The comparison should be with other high risk industries, like the airline industry who have systematically improved safety. It has taken them a decade or more to get to where they are, in fact even longer than that in some industries, but we have some of the key ingredients in place that have been shown from evidence from other industries to work. We are seeing cultural change which is probably the most important thing, more safety awareness in local services, and that is acknowledged in the report. We are seeing more reporting and analysis of reports, which was also a way in which the airline industry changed. We are seeing specific solutions coming through to reduce risk. Those are not working as well as they could be yet but, as Sir Nigel said, there are more of them coming through and certainly the advent of the electronic patient record will benefit safety of medication which accounts for 25% of the harm worldwide, medication errors, and it will also reduce some of the problems that result from poor communication and fragmented clinical information.

Q27 Mr Khan: So a noticeable change a decade from now?

Professor Sir Liam Donaldson: Absolutely, yes, but with incremental change over that period.

Q28 Mr Khan: One of the things you referred to was the culture. In the context of enhancing the safety culture within NHS trusts, there are comments made in the report about having an open and fair employer so that staff feel confident coming forward. Can I ask you what further actions you expect the NPSA to take to improve the culture in the NHS so that staff feel they have an open and fair employer?

Professor Sir Liam Donaldson: I think they have already put out guidance to good practice that staff should not be suspended unless there is evidence of negligence or careless conduct. On the majority of occasions when something goes wrong there is an error but it is a failure provoked by weak systems supporting the practitioner concerned, so just by careful monitoring. We do live in a blame culture society, as is the case in many Western countries, where scapegoats are looked for and individuals are blamed for mistakes but as we have seen in other industries, like the airline industry, that blame culture can be rolled back but it requires effort not just within the service concerned but by society as a whole and in particular the media.

Q29 Mr Khan: In particular, in your answer a knock-on effect that will have on clinical negligence cases is if you are admitting your mistakes that may have an impact on the number of cases that are settled.

Ms Williams: The programme we have to support a culture of change is we have trained 8,000 staff in contributory analysis, which is a particular technique which seeks to look at the contributory factors that lie behind an incident which starts to move people away from individual blame. At our conference next week we are launching a cultural assessment tool so that trusts themselves, whether at unit team level or strategic health authority level, can assess the level of maturity against a well recognised and used tool in other industries. We have trained and worked with 113 boards to talk through the issues of open and fair culture. We have issued a chief exec checklist so that chief execs themselves know the role that they can play to promote safety. Also, we have run leadership courses through the lens of patient safety to introduce them to some of these concepts. In terms of your last question, I am sorry I lost the ----

Q30 Mr Khan: The impact on settlements in cases.

Ms Williams: Just recently we issued a being open policy and teaching materials. This involves apologising and giving a full explanation involving the patient and their relatives in working through what actions might prevent harm in the future. We built that policy from experience both in Australia but particularly in veterans' health services in the USA where they have run this policy for a number of years and their negligence bill has not increased during this time.

Q31 Mr Khan: At the beginning of your answer you referred to your 8,000 staff who have been trained on the forms. How will the NPSA be able to identify learning when it says in paragraph 2.38 that trusts "are not required to provide information on contributory factors"?

Ms Williams: A number of trusts are using the form that ----

Q32 Mr Khan: They are not required to, are they?

Ms Williams: Not at the moment because we are reliant on seven or eight commercial vendors and not all of those systems collect contributory factors. The numbers where we are getting this information is increasing. What we would like to see over time as people become familiar with these terms - these are very new ways of looking at incidents - as trusts become more familiar is their internal forms changing.

Q33 Mr Khan: Do you envisage it being a requirement to provide those?

Ms Williams: I think in time, yes, it will be.

Q34 Mr Khan: This is probably a question for Sir Liam. Most countries favour a confidential rather than anonymous service for reporting because it means that you can learn from the information you are given. Why is the National Reporting and Learning System that we have anonymous?

Professor Sir Liam Donaldson: Only one aspect of it is anonymous. The confidentiality code can be broken in circumstances where there is a very serious cluster of cases that needs to be investigated further. By and large, the majority of reports are made through local risk managers, the clinicians giving their reports to the local risk managers. They are being open about it anyway. It is important to emphasise that a lot of learning needs to take place at local level, it is not just a case of looking at reports at national level, those incidents need to be used at local level to introduce safer systems in the hospital.

Q35 Jon Trickett: I want to reflect on patient choice since we are now offering a choice of hospitals to patients. What information is provided to patients about the level of accidents, say per thousand staff, so they choose which hospital they would prefer to go to?

Ms Williams: At the moment there are a number of trust boards who do put papers on their public part of the agenda which show the number of incidents by specialty and what action they are taking as a result. The work that we have done with groups of patients to discuss this issue to feed into the patient choice agenda is the issue for us is that more reports is a sign of a healthy environment in which incidents are raised and action can be taken and, therefore, it seems more counterintuitive but more reports is a good thing for patients to be looking for. In fact, to choose a hospital where there were very few reports might be a concern because it might show that there might be a level of cover-up.

Q36 Jon Trickett: Does the GP provide to the patient the number of accidents in the four hospitals which are being offered to the patients?

Sir Nigel Crisp: No. The two things that the patients will get are the MRSA rate, which is one of the issues here, and that is published by the hospital, and the second thing that is also available to patients is the Health Care Commission's report on the hospital. We do not have a figure of accidents per so many staff.

Q37 Jon Trickett: I will ask you about the two tables in here in a second. What monitoring do we do? If I go to my GP and I am referred to three NHS trusts and a BUPA hospital and I ask what information he or she has about the number of accidents across the four sites, is that information available to him or her?

Sir Nigel Crisp: There is not information that is systematically available about accidents across sites. There is about the things that we collect and publish, and I deliberately say MRSA and I do deliberately pick out the Health Care Commission's report which will be made public. Those are in the public domain. As you know from the earlier discussion and the points that Ms Williams has just been making, there is not a simple definition of what are accidents as opposed to anything else.

Q38 Jon Trickett: You have signed the paper off. Do we collect information about hospitals in the independent sector, the number of accidents?

Sir Nigel Crisp: We do not. What we are doing with the NPSA is first of all starting with the hospitals physically within the NHS with the intention of then moving on to deal with the independent sector of whatever sort.

Q39 Jon Trickett: Do you have statutory powers to receive the information from the independent sector?

Sir Nigel Crisp: The Health Care Commission does.

Q40 Jon Trickett: The NPSA, does your remit run into the independent sector hospitals?

Ms Williams: Our remit extends to wherever NHS care is funded and, therefore, clearly the independent sector is a vast area and ----

Q41 Jon Trickett: Are you collecting information from, say, BUPA hospitals?

Ms Williams: Not at the moment. We are concentrating on getting all the NHS trusts reporting.

Q42 Jon Trickett: Now we are committed, are we not, to 15% of all operations going into the independent sector?

Sir Nigel Crisp: No, we are not actually. That is a quotation from Mr Reid which was up to 15%.

Q43 Jon Trickett: We are committed to a number of NHS operations being commissioned in the independent sector, whatever the figure is. We have no idea at all how many accidents have taken place in the various hospitals which are being offered by the GPs, is that correct?

Sir Nigel Crisp: We do not have the same system in place yet in terms of direct reporting.

Q44 Jon Trickett: Would it not be a good idea for the local health authorities to find out how many accidents take place in independent sector hospitals before they are offered to patients as one of the four choices?

Sir Nigel Crisp: In the contracts with the independent sector agencies we do have requirements about what we call clinical governance, which is about reporting and how they manage patient incidents. It is not the same mechanism.

Q45 Jon Trickett: You have no idea how many accidents are taking place at Methley Park, which is the local BUPA hospital in my patch?

Sir Nigel Crisp: I think that is true. I do not know whether either of my colleagues know.

Q46 Jon Trickett: You do not have any idea at all? On the ISTCs, the independent sector treatment centres, are we monitoring those?

Sir Nigel Crisp: Not through the same system. As I was saying, we have contracts with the individual organisations which have certain requirements about clinical governance in them and about the reporting of incidents, how they are managed and who is overseeing them and so on. That is something we have got under review.

Q47 Jon Trickett: Let me ask you another question because I have just come from a meeting with my local chief executive who tells me he is about to hand a lot of staff over to the PFI partner for the local hospital. They will be working in an NHS hospital but it is a PFI hospital. Those porters, electricians, joiners, domestic cleaners and all those kinds of staff are not going to be employed by the NHS. Do they come under the reporting mechanisms which you have been talking about this afternoon?

Sir Nigel Crisp: Yes, I think they do.

Ms Williams: There are a number of PFI hospitals which are reporting through to us. If I may add that the larger private sector, independent sector hospitals very often do have their own reporting systems and we are in discussion with them about how they might link through to the National Reporting and Learning System.

Q48 Jon Trickett: You have given me two answers. Let me just deal with the second one and come back to the first. In terms of those staff who are employed by the PFI partner, are they obliged to report accidents in exactly the same way as NHS staff are?

Sir Nigel Crisp: In an NHS hospital.

Q49 Jon Trickett: Yes they are?

Ms Williams: Yes.

Q50 Jon Trickett: You then went back to the debate about the ISTCs. My understanding at the moment is that the ISTCs are not required to provide information. Whether or not they are doing, they are not required to, is that right?

Ms Williams: That is right.

Q51 Jon Trickett: Does your remit allow you to go into those hospitals at some point, it is just that you have not got round to it yet?

Ms Williams: We would like to develop an arrangement so that the independent sector, whether it is ISTCs or the larger hospitals, BUPA et cetera, can report incidents so that there is learning that can affect all patients.

Q52 Jon Trickett: So it is an aspiration, "we would like"?

Ms Williams: Yes.

Q53 Jon Trickett: I think that is profoundly unsatisfactory from the point of view of the patient, Sir Nigel. What do you think?

Sir Nigel Crisp: I understand the point, but that is why we have got the contracts with people and we have got what I have described as clinical governance arrangements to make sure that there is reporting to us of incidents so that we can investigate them, and why we have used our clinical governance team to go in and look at where we have had incidents reported.

Q54 Jon Trickett: Basically it is a lack of an even playing field between the NHS and the rest of the medical health sector, is it not? Can I just draw attention to table six on page 25 which shows this remarkable curve in terms of the number of incidents in each acute trust. The problem is it is not comparable to table five, which is the number of incidents per thousand staff. It is a very crude figure indeed, is it not? I might ask the NAO to produce it on the same basis. I will put that to Sir John. The curve would be probably less since the smaller number of accidents might be taking place with only half a dozen staff or something.

Sir John Bourn: Right.

Q55 Jon Trickett: All these curves are all the same. Have you formed a view as to what correlation there is between the ones who have so few incidents and any sort of external factors which might govern a particular trust, or is it simply that self-reporting is not working, as I suspect?

Sir Nigel Crisp: You are quite right on the last point. These are reports of incidents rather than actual incidents, so there is some spread in what people are reporting. Very definitely we pick up on patterns but we need to pick up on patterns firstly at the local level and then nationally.

Ms Williams: In a sense this is not unexpected given where we are in relation to the cultural issues that we were talking about earlier. Over time we would expect to see an increase in reporting rate across all of the NHS.

Q56 Jon Trickett: You would expect the curve to flatten out, would you not, and it is not over the two year period that we are looking at. Have you not asked the question of yourself in a way that you can report to us why there should be some trusts which are reporting almost no accidents at all?

Ms Williams: They have now but some trusts during this period did not have a centralised reporting system. What we found when we were developing the scheme, particularly in primary care and in the ambulance services, was there were some parts of the country where they were very reliant on the paper system and things were at a very early stage. There is nervousness amongst staff groups about reporting. Our role is to try to promote a culture where we see a year-on-year increase in reporting from all trusts.

Q57 Jon Trickett: My time is up but I wonder if I could ask the NAO to produce those figures I asked for and also whether there is a correlation between the number of stars which each trust has so we can see the curve for no star, one star, two star and three star trusts.

Sir John Bourn: I will produce that information.

Q58 Mr Bacon: On page 34, paragraph 2.31, Sir Nigel, there is a reference to the fact that: "Healthcare organisations in other countries, having compared the merits of anonymous and confidential reporting, have generally opted for confidential reporting." This system opted for anonymous. Why do you think that was? Do you see yourselves moving towards a more open system?

Sir Nigel Crisp: I think this was the same point Sir Liam responded to a moment ago. I think we have got it confidential at a local level and anonymous at a national level. That is felt to be the right balance so that confidentiality can be handled and learned about at the local level whereas anonymous is the right level for us to be looking at the big patterns. Is there anything you want to add to that?

Professor Sir Liam Donaldson: I have already responded to part of that.

Ms Williams: We went for that de-identifier so that we do not carry names of clinicians or patients at a national level in the database. That is because what we are looking for is themes and trends, types of incidents, where we might be able to develop a system-wide intervention to prevent harm recurring to those particular groups of patients, therefore we do not need the identifying details about individual people at a national level.

Q59 Mr Bacon: I appreciate that you want to have as open reporting as you can about the facts and the themes and the trends and why people behave in certain ways, but you still want to be able to take corrective and, if necessary, disciplinary action, do you not?

Ms Williams: Yes, and that will take place at the local level.

Q60 Mr Bacon: You are saying the nature of the report is such that there is a local identifier of who it is?

Ms Williams: Yes.

Q61 Mr Bacon: I think the very first committee meeting I attended of this Committee some years ago, certainly if not the first one of the very earliest, was on the NHS Litigation Authority. I remembered that meeting when reading this statement on page 55 where it says in paragraph 3.36: "the prevailing legal system does not encourage health professionals to be open after an adverse patient safety incident..." What have you done, Sir Nigel, to try to suggest policy changes to the prevailing legal system as it relates to medical error and as far as it relates to litigation since October 2001 when we had that hearing?

Sir Nigel Crisp: Some very specific things, but can I ask Sir Liam, who has been the architect of these, to address it.

Professor Sir Liam Donaldson: Apart from the measures that have been taken to encourage reporting and which have been pretty successful given the level of reporting that we have seen over the last couple of years I also produced for the Government a report on reforming the medical litigation system called Making Amends which is about to work its way through the House towards a Bill. That firmly places emphasis on trying to get blame and retribution out of the litigation system, allowing patients, not just with small claims, to have no-fault compensation but also to ensure they have an apology, an explanation of what has happened and a report from the local health service telling them what action will be taken as a result of the incident that harmed them and how it might prevent harm to another patient. That is another strand of action which tries to improve the climate and stop us going down the American path of very confrontational and costly litigation.

Q62 Mr Bacon: This is a question for Ms Williams. It relates to paragraph 2.38 and the follow-up work to the An Organisation with a Memory report, and it says halfway down that paragraph: "Despite the existence of well-developed international incident classification, the National Patient Safety Agency decided to define its own taxonomy for national reporting and produce tailored versions for use in nine different healthcare settings." Why did you not follow the widely used international incident classification?

Ms Williams: We could not find any widely used international incident taxonomies.

Q63 Mr Bacon: You did not find anyone who used them?

Ms Williams: Not an international taxonomy. What we found was a number of state-wide taxonomies in Australia or in the United States, very often uni-fucntional, so only concerned with a particular type of speciality.

Q64 Mr Bacon: So is this paragraph wrong?

Ms Williams: No, it is not wrong. There are a number of taxonomies around the world which we did review and what we found was there was very little for mental health, in fact none, for learning disability, primary care or ambulance services. There was some work done in some states in some countries that looked at acute services.

Q65 Mr Bacon: So you constructed a new classification?

Ms Williams: So we worked with clinicians to construct something that was relevant for the UK.

Q66 Mr Bacon: Can you tell me whether the next sentence is correct. It says: "...reporting fields, which identify the contributory factors to the incident, are optional, and compliance is variable, even though the learning of lessons is most likely to come from this information." Is that sentence in all its particulars correct?

Ms Williams: Yes, it is correct. As I was explaining earlier, not all the commercial risk systems that the trusts have purchased collect contributory factors. We have thousands of reports which do have contributory factors on but this is an area where we want to make changes over the next year or so. We are going to be reviewing our data set, we gave a commitment to review it once we have rolled it out to all the -----

Q67 Mr Bacon: Do you mean the fields will be obligatory rather than optional?

Ms Williams: I think they need to be because that is where the -----

Q68 Mr Bacon: Is it not rather obvious to make them obligatory if you want to have complete data to work with? You do not need thousands of consultants to tell you that.

Ms Williams: The free text in the reports that we currently get reveal an enormous amount and we are able to use that for learning. Certainly ideally we would like the contributory factors but it would mean commercial systems making a change. One of the things that we have been doing is working with Connecting for Health and one of the solutions to this would be a national specification for risk management systems.

Q69 Mr Bacon: Is this yet another bell and whistle added on to the original Connecting for Health specification?

Ms Williams: I think it is something that we could very closely work with them on. It is a specification that would be tendered but it would give that mandatory flavour which I think we are all asking for.

Sir Nigel Crisp: We are holding off the bells and whistles at the moment.

Q70 Mr Bacon: May I ask about the Department of Health's identification of the Australian patient safety system which is called AIMS, Advanced Incident Monitoring System? The Department of Health identified that as a workable system but when the responsibility was transferred to the NPSA you did not go with that, as it were, an off-the-shelf working system. You started from scratch. Can you say why?

Ms Williams: The Department of Health tendered for a system and when we arrived in post there was a consortium in place between the company that operated the AIM System and a UK-based software system. We decided to let the pilot run. We evaluated it in April 2002 and we found that there were a number of problems. There were technical difficulties and we learned a huge number of lessons but it was not a system that at that time we felt could be rolled out to the whole of the UK.

Q71 Mr Bacon: Could I just check that Mr Stuart Emslie, who wrote to this committee with a note about this system and indeed about what he feels was a waste of money which the Department of Health was engaged in on this procurement, is the same Stuart Emslie who was reported on 2 December 2001 in The Sunday Times as having given an internal briefing to the Department of Health the previous month, in November 2001, that 16-20% of your budget disappeared through waste, fraud and mismanagement?

Sir Nigel Crisp: I do not know.

Q72 Chairman: I understand that this is a matter which is sub judice. Under the rules of the House, as our Clerk advises us, it might be difficult to pursue this matter. I understand there is a writ against this man. Is that right?

Sir Nigel Crisp: Yes, that is right.

Q73 Mr Bacon: I am not wishing, Chairman, to stray into anything that might come before the court. I am simply trying to identify if this is the same person who was referred to in the article in The Sunday Times on 2 December 2001.

Sir Nigel Crisp: Frankly, I do not have a memory for everything that has appeared in The Sunday Times in the last five years, but I would request the Chairman that we do not go into this area, for the reasons that the Chairman has stated.

Q74 Mr Bacon: I would just like to know if it is the same person. I think it is right that you can confirm whether it is the same person or not.

Sir Nigel Crisp: I suspect it is.

Q75 Mr Bacon: Can I ask you one more question, and this may be for Ms Williams again? It is about power generators. In the Adjournment Debate the other day there was an answer by the Minister of State concerning medical injury, the Sarah Lynch brain damage case, a very sad case. One of the problems, and it is arguable to this day, 20 years later, whether this was a contributory factor, was that there was a power cut and the back-up generator also did not work. What data do you keep centrally on the state of back-up generators and whether they are all in good condition and maintained regularly? In this particular incident the back-up generator log book was destroyed. Do you keep data centrally on that?

Ms Williams: No, we would not keep data about individual pieces of equipment or estate at the NPSA.

Q76 Mr Bacon: It would be at the trust level, would it?

Ms Williams: That would be at the trust level. There will have been guidance from NHS Estates in the past requiring trusts to make sure that there were suitable back-up arrangements.

Q77 Mr Mitchell: Can I carry on with the National Patient Safety Agency and ask Sir Nigel how he rates the success of the agency in meeting its key target of improving the culture of the NHS? Would you say it was stunning? Would you say it was mediocre? Would you say, in a civil servants' phrase, it was disappointing, or lousy?

Sir Nigel Crisp: I would say good and more to do. I think we have come quite a long way but there is an enormous amount further that needs to be done. Why I say good is because we do have this reporting system that is at the level of other people around the world or where we are perhaps leading the way. We do have the 15 safety alerts that I referred to and so on, so I think it is fair to say good but I do think, as this report reveals, we have got a lot further to go to see improvements, so I am going wider than just the NPSA, but I think they have played a significant part in this.

Q78 Mr Mitchell: But come a long way more slowly than everybody would have hoped?

Sir Nigel Crisp: I think that is true but I do think that it is a very strong point that they took over a piloting system from the Department of Health, evaluated it, found it wanting and then went on for a further procurement. You hope pilots work but if they do not you want proper evaluation. I am pleased that the board and chief executives had a proper evaluation. That is disappointing but I think we have come a long way.

Q79 Mr Mitchell: Can I ask Ms Williams why it took such a long time to get the National Reporting and Learning System off the ground? This was presumably a key task and yet you were messing about for ages.

Ms Williams: We could not proceed with the first pilot and therefore we had to ascertain whether it would be possible to roll out a system with just the Australian company that we mentioned earlier. That was also found not to be possible. We also discovered during that period in 2002 that the scale of the enterprise was such that we needed to seek Treasury approval for a full business case. That was done and from receipt of approval it has taken two years to roll out to 607 organisations, which I think is a major task. Of course we would like to have done it more quickly. However, we did not wait until we had rolled it out before starting work on a range of solutions. We had a number of issues that were raised with us by patients or by members of staff, we worked on those and we issued guidance to try to prevent harm, so we did not wait for the reporting system to get going before we started work on our solutions.

Q80 Mr Mitchell: You talk about the Australian system but Mr Emslie tells us that there was an existing NHS Safecode system which was a Crown product developed and funded in this country. Why did you not consider using that?

Ms Williams: There are a number of risk software systems developed in this country. I do not believe it would be appropriate to comment further on an individual system.

Q81 Chairman: Was it unsuitable?

Ms Williams: I do not believe I can comment.

Sir Nigel Crisp: The organisation carried out a rigorous evaluation, which is what we would want them to do, speaking for the Department of Health, and took the view, having done that evaluation, that they needed to tender again, in part because they needed to in Treasury terms, and accordingly did so. They made a judgment on the evidence available to them at the time and in an appropriate fashion.

Q82 Mr Mitchell: Did you even consider Safecode?

Sir Nigel Crisp: I think this is all part of the issue you talked about earlier, Chairman, and I think we really should not get into this.

Q83 Mr Mitchell: So we take it that it was not considered?

Sir Nigel Crisp: I am happy at some other point if you want to send you a written note or something but I do not think I am prepared to discuss this particular issue further as this is a matter that is sub judice.

Q84 Chairman: Would you be more happy giving evidence in private or would you rather not give any evidence at all?

Sir Nigel Crisp: I would rather not give any evidence at all.

Q85 Chairman: Because it is sub judice?

Sir Nigel Crisp: Because it is sub judice.

Chairman: That is our problem. That is the rule of the House. I am quite happy to take advice but under the rules of this House we are not supposed to take evidence on matters which are sub judice.

Q86 Mr Mitchell: Why, at the end of the day, when you had devised a system, did you come up with one where the trusts questioned the value of sending data to the system, given the lack of feedback and the lack of emphasis on solutions and given its complications?

Ms Williams: In terms of complications we had a choice. We could either create a system whereby trusts reported separately to us, which would have meant reporting once on their own system and separately to a national organisation. We took the view, and it did take longer, that we would integrate our requirements as far as we possibly could into the commercial systems. That has meant at the initial stages a mapping exercise which has caused additional problems for the trusts but once mapped it is very much easier for the trusts to send us information. On the whole question of feedback, we are very aware that unless there is regular feedback this does act as a barrier to reporting. We would argue that we have produced three reports from each of the two pilots that we had, plus the Observatory report last year. As Sir Nigel has said, we have had 15 solutions. We have had conferences, we have run training sessions for 47,000 staff over the period, and therefore in one sense there has been feedback. It is not sufficient and clearly this is something we need to do more of. We have agreed with the Department of Health that there will be quarterly reports coming from the Observatory, which is the term we use where we bring together information from the reporting system together with other information sources - litigation, complaints, et cetera. We will publish more regularly on a quarterly basis and we have been piloting an extra net with trusts so that they can get immediate feedback which will enable them to benchmark themselves against others in the organisations.

Q87 Mr Mitchell: But it seems odd, having moved so slowly and looked at this system and rejected that system, for reasons we are not allowed to be told, that you came up with a system which did not satisfy the needs of learning because it does not tell us about the causes of whatever has happened. How can you identify learning when trusts are not required to provide information on contributory factors?

Ms Williams: As I have previously answered, we would like to see in future more of this information coming in. Many of the systems do provide this. We have many thousands of reports which do include contributory factors. We do believe that ideally it should be a mandatory field and that is something we would be working on with Connecting for Health.

Sir Nigel Crisp: May I make one point on this, which I hope is helpful, which is that the point there is about the feeder systems not requiring that rather than about how they can be collected centrally. That is about using the feeder systems in the trusts.

Q88 Mr Mitchell: What we need at the end of the day is guidance on avoiding accidents and that is not what you are able to provide.

Ms Williams: We have produced a major guidance document. It is called Seven steps for patient safety. It outlines for trusts a comprehensive range of policies which they would need to develop in order to build the infrastructure that would support safety at a local level. On its own it provides the framework within which they develop their systems but on top of that we produce seven or eight other tools and techniques that would support safety at the local level.

Q89 Mr Mitchell: Yes, but at the end of the day patient safety incidents cost, it is estimated, a couple of billion in extra bed days, so why have you not done more to develop guidance on costing the patient safety incidents and listing the solutions to them? This seems to be a key weakness.

Professor Sir Liam Donaldson: The agency has put out 15 alerts on different subjects. Over the same period of time the joint commission in the States, which is probably the international benchmark for putting out these sorts of solutions to reduce risks, put out ten, so they have put out very extensive guidance and several of them are in areas where the cost savings are very big. Sir Nigel mentioned the Clean Hands campaign. There is the whole question of the infusion pumps which cost lives and cause harm, so there are some very significant steps that have been taken in my view, and once the analysis of these nearly a million reports is more fully developed I think the solutions will flow out even more swiftly. They have looked, for example, at misplacement of nasogastric tubes which cost the lives of children and babies. They have looked at wrong side surgery. All of these things are very important measures to reduce risk in specific areas as well as the general Seven steps type of approach which Ms Williams has mentioned.

Q90 Mr Mitchell: Yes, but this is an agency which is our agency which is supposed to develop solutions to our problems in a field where you told us at the start of this that we were world leaders and doing very well, thank you very much, an agency which has already rejected Australian experience. Why should we be reliant on the Americans in this kind of field? Why it is not doing it itself?

Professor Sir Liam Donaldson: If you look around the world we are one of the few countries to have a nationwide system. The Americans have only got systems in certain parts of the country, so have the Australians. Indeed, the number of incident reports we have already I do not think has been surpassed anywhere in the world. Proportionate to the size of the populations, we are level with the Veterans Administration which covers seven million people in the States. We cover 53 million people, so both numerically and proportionately I think we are in the lead. Other countries, and indeed other industries, have shown that you have to get high quality data in before you can start analysing. As Ms Williams has said, even before the data are fully in they have put out 15 alerts, which I think is quite a strong record, certainly in comparison to the other example that I gave.

Q91 Kitty Ussher: As somebody who is not expert in this particular field of policy, reading the brief and the report for the first time, I must admit I was immediately rather scared. I think the number of one in ten is much higher than members of the public would expect and is certainly way higher than they would hope for. The idea that when you go into hospital and your life and welfare is in someone else's hands and in one in ten times it will be made worse in accidents rather than something internal is really quite terrifying. I have heard you say that you think things are improving but obviously there is a long way to go. Could you describe the situation before Ms Williams' agency was established so that we have some kind of benchmark about where it started from? Where were we ten years ago, for example?

Professor Sir Liam Donaldson: Perhaps I could start on that. Ten or 15 years ago most members of the medical profession would say there was no such thing as a bad doctor. Most would say that you could not measure quality so why bother to try and improve it. Over the last five or six years we have put in place a comprehensive quality framework in this country which is admired internationally, with clear national standards, with inspectorates. Safety is the first of those national standards that were issued a year ago on which the Healthcare Commission inspects. At local level every hospital has now a duty of quality, and again that is very unusual compared to other countries, and local programmes of what we call clinical governance (which is a way of ensuring that clinicians are involved in quality assurance), quality improvement and safety are in place, and indeed the NAO report talks in positive terms about our clinical governance programme, so things have moved on a lot. This strand of safety has been added to that overall programme and I think has in place the ingredients necessary to improve safety very considerably: the cultural change, the technical support with reporting and learning systems, the area which we are working on at the moment to improve education and training. If you take the particular element about poor practice and bad doctors, as I was saying earlier, we have also moved forward very substantially on that in identifying bad doctors early, trying to rehabilitate where possible but ensuring that patients are protected at an early stage. There is more to do but the emphasis is very much on quality and safety in the NHS today.

Q92 Kitty Ussher: What was the trigger for the establishment of the agency?

Professor Sir Liam Donaldson: It was a report that I produced called An Organisation with a Memory. I had had a longstanding interest in and had read a lot about the work that was being done in the airline industry and the way that they had managed to improve safety over many years and I thought that there would be a comparable programme that could be launched in healthcare.

Q93 Kitty Ussher: Was there any national budget stream for patient safety in the holistic rational sense prior to the establishment of the agency?

Professor Sir Liam Donaldson: Not specifically, no, and one was created by the implementation of An Organisation with a Memory.

Q94 Kitty Ussher: In your opening remarks in your conversation with the Chairman you suggested that you should write to the committee with the international comparisons to make sure that we had up-to-date information. Since then you have mentioned a couple of other instances. Since members, if you will permit me, Chairman, keep saying, "How does our one in ten stack up when compared internationally?", and you have mentioned that we are a world leader now, could you expand on that answer a little bit more and give us a quick indication of where we now stand compared to other countries?

Professor Sir Liam Donaldson: In developed countries the ball park figure of one in ten hospital admissions resulting in some form of error or mistake is probably comparable across all countries. We do not know what the position is in developing countries. One would assume that, because of their poor infrastructure and resources, the problems there would be more serious, but the World Health Organisation is currently researching that. As far as the scale of the problem is concerned we are probably broadly comparable with other countries and certainly there is a great deal of concern in the US about the level of inadvertent harm caused by their healthcare system. As far as making commitment to action is concerned, we are in the forefront, although commitment and enthusiasm are growing across many countries of the world now and I have talked to people in other countries about what they are doing. The area where we need to achieve more is in this area of implementing risk reduction measures because, aside from some limited evaluations of the benefits of introducing, for example, electronic medicine prescribing into some parts of the world, there are very few examples of where reductions in risk can be quantified and attributed to particular interventions. We are trying to learn as much as we can from the researchers and from what is happening elsewhere as well as implementing our own programmes.

Q95 Kitty Ussher: Given that we had not done much until very recently and given, as you seem to imply, that many other countries are in the same situation, if we are all on a ratio of one to ten what do you think potentially, hypothetically, that ratio can be reduced to once we all operate at the maximum of our potential? I am not talking about the timescale but if that could become in theory one in 15 or one in 20 do you have a sense of the potential improvement there?

Professor Sir Liam Donaldson: Given the experience of other industries, those scales of reduction are achievable and I do not think there is any reason why healthcare could not achieve the same sort of record of year on year improvement.

Q96 Kitty Ussher: One in 20 then would you consider?

Professor Sir Liam Donaldson: It is difficult to put an exact figure on it.

Q97 Kitty Ussher: But that order of magnitude?

Professor Sir Liam Donaldson: Yes.

Q98 Kitty Ussher: You mentioned the implementation mechanism for the things that you have found out, which was going to be my next question, and I presume Ms Williams is the most appropriate person to answer. You have described how incidents are reported and you obviously now have a large quantity of data. You have said that you have issued 15 alerts and have this training agenda and the Clean Hands campaign. I launched our own Clean Hands campaign in my constituency so I know it is there, but what kind of enforcement powers do you have and what kind of checking or accountability powers do you have to make sure changes are made?

Ms Williams: The NPSA itself does not have enforcement powers. There are three ways in which it is possible, if you like, to find out whether anyone is taking any notice of what we are putting out. The first thing is that we do have our own evaluation programme where we look to see what are the barriers to implementing our suggestions because we ourselves want to learn for future products how we can make it easy for people to implement our solutions. Secondly, and very importantly, the alerts go out through an alert system which is monitored by strategic health authorities. It is called the SAD system, so when something goes out trusts are required to indicate whether they are taking action, whether it is appropriate to take action, and when they will take action, so there is that performance monitoring. The third important strand is that the Healthcare Commission, which, as well as self-assessment, will be undertaking random inspections, have agreed to include in their criteria a check on a random sample of alerts that we put out, so they will be absolutely able to see a demonstration that something has happened at the other end.

Q99 Kitty Ussher: That is reassuring. Has any of that actually happened yet? Do you have any data as to how effective you are being?

Ms Williams: We have some examples. One of the topics we took was the standardisation of the crash call number. Standardisation is a common safety solution. What we found before we standardised to four twos - this is where you call for help from a team if somebody is having a cardiac arrest - was that there were 27 different telephone numbers across the Health Service, so staff clearly moved from one to another site and very often agency staff would be working across different locations. As at the end of 2005 all trusts have standardised to the four twos, 2222, so this is the standard number that can be used right across. Another example would be one of our very early alerts. Potassium chloride has already been mentioned. We asked trusts to withdraw it from general ward areas. This is a very toxic drug in its undiluted form. Before we put out our alert we did a base line. 34% of ward areas had not got this drug on their shelves, as it were. We checked two years later and in fact we are up to 98% of wards that have now removed it from their ward areas. This reduces the likelihood of a member of staff in a hurry reaching for this drug and giving it in an undiluted form.

Q100 Greg Clark: Ms Williams, what is the annual budget of your organisation?

Ms Williams: The original NPSA budget was £15 million a year.

Q101 Greg Clark: The original one?

Ms Williams: From 1 April 2005 we took on a range of additional responsibilities following the Government review of arm's length bodies and that added an extra £20 million or so to our budget.

Q102 Greg Clark: So the budget for next year is what?

Ms Williams: It is of the order of £44 million.

Q103 Greg Clark: How many employees do you have working for you?

Ms Williams: With our new responsibilities we have 315 whole time employment.

Q104 Greg Clark: How long have you been Chief Executive of the organisation?

Ms Williams: Since its inception in October 2001.

Q105 Greg Clark: During that time have you identified any principles that would tend to make the hospital more safe and less prone to these patient safety incidents?

Ms Williams: What we have identified, having looked at other industries, is aspects of what we would call a safety management system. There are certain things that need to be in place that would tend to lead towards a safer environment.

Q106 Greg Clark: Can you give me some examples?

Ms Williams: They are a reporting system, an open and fair culture (which can be tested), a root cause analysis of serious incidents when they occur, feedback to staff, multidisciplinary teamwork, communications, work on those particular areas, handovers, and then time for learning. We found in some trusts that they had very well developed systems where on a regular basis the multidisciplinary team sit down - and it can happen in a GP practice or in a specialty - and discuss, "What has gone well, what has not gone so well, what can we put in place to prevent risk occurring?". It is that sort of drive that we want to see extended right across.

Q107 Greg Clark: That seems very sensible and I would expect that, but it strikes me, just reading the report, that some of those principles do not seem to apply to the organisation itself. Take learning, for example. Mr Bacon has already raised a point about international comparisons and I detect a note of criticism in the NAO report when it says, "... and, despite the existence of well developed international incident classification, the National Patient Safety Organisation decided to define its own taxonomy and national reporting and produce tailored versions for use in nine different healthcare settings". For an organisation that is there to promote learning it seems strange that you would ignore the international examples and go for something that is entirely unique.

Ms Williams: We did not ignore them. We went through a process of reviewing the classifications that we could find internationally and, as I have previously said, we could not find any taxonomies that related to mental health, learning disability, the ambulance services, primary care. There was some work done on acute services in some parts of the world.

Q108 Greg Clark: But just on acute services is it not possible to take in that, which is, I would imagine, a large portion of the incidents?

Ms Williams: Again, when you look at any other national system, there is always a need to customise it for the local language -----

Q109 Greg Clark: Customise, yes, but to start from scratch seems extreme.

Ms Williams: We felt that we needed a system that had local clinician support and we went through a process which involved several hundred clinical staff to reach a taxonomy that learnt from others but actually met the requirements -----

Q110 Greg Clark: It strikes me that as an organisation you advocate learning but do not seem to have done much learning yourself when it comes to this.

Sir Nigel Crisp: Can I just say something there? I think that paragraph does say that there are well developed international incident classifications, but the point that Ms Williams and colleagues are making is that they are not comprehensive and we wanted a comprehensive system that covered all patients. Whilst it is appropriate to learn from other people, we now actually have a system that does cover the whole patient population.

Q111 Greg Clark: I would have thought in the context of this organisation that learning implied sharing best practice rather than adopting a unique approach. Let me move on to another principle. It strikes me that it would be reasonable to suppose in any discussion of safety that mistakes happen when procedures are complex and unclear. I assume that would be a common sense assessment. Would that be reasonable, that complexity is an area of safety?

Ms Williams: Yes, complexity is plainly going to create a more difficult environment.

Q112 Greg Clark: But then we see on page 34 of the report, paragraph 2.30, that the NAO concludes that trusts "face an extremely complex system of reporting and investigation". For an organisation that is, one would hope, aiming to promote simplicity and clarity that is a dreadful conclusion from the National Audit Office surely.

Ms Williams: It is true to say, as the report makes clear, that there are a number of bodies to which trusts should report and some of these are for very good statutory reasons, whether it is for overdose of radiation or whether it is the Health and Safety Executive, or whether it is the Health Protection Agency which offers surveillance for -----

Q113 Greg Clark: Surely, for an organisation to be about simplicity and clarity, to have a conclusion from the National Audit Office that described not just a complex system but an extremely complex system of reporting and investigation, that seems to be dysfunctional.

Sir Nigel Crisp: To be fair, if I may come in, if I am reading paragraph 2.30 properly, I do not think that is purely about the NPSA. I think that is about the fact that trusts do have to report to a lot of people.

Q114 Greg Clark: Shall we read it out: "... the National Reporting and Learning System added to the list of organisations to which trusts were already required to report and trusts still face an extremely complex system of reporting and investigation. Figure 14 overleaf", it goes on to say, "shows the main national reporting systems, but around 30 routes still remain." That seems extraordinary.

Sir Nigel Crisp: The point I was making was that I am not sure that is entirely fair to lay that at the door of the NPSA because if you look at those if you look at those other agencies, police and coroners and other people, they require information as well, and it is not surprising that they do. What we have discussed is -----

Q115 Greg Clark: The problem is, is it not, that the way the NPSA has gone about its work has duplicated the systems in place rather than added to them? For example, if we take Appendix 5, page 71, the final bullet point says that "trusts that were visited felt that the local systems were more important for learning lessons". As far as I understand it, one of the objectives of this organisation, the National Patient Safety Agency, is to promote learning and yet we find that their own systems are not, practitioners find, the best place to promote learning; it is the local systems that have been added to. That again is not a happy conclusion, Ms Williams.

Sir Nigel Crisp: I think there are two points here.

Q116 Greg Clark: Ms Williams' perspective is the one we would like to have.

Ms Williams: Of course, local learning is absolutely vital and it is the building block on which any national system will sit. However, what a local system cannot do necessarily is pick up themes and trends that are applying across systems. They will only know of the incidents that they report locally. They will not know that actually it is part of a trend that is quite widespread. Only a national system can do that.

Q117 Greg Clark: Just on that point, Ms Williams, paragraph 2.37 says that the National Patient Safety Agency "could have collected aggregate information on commonly occurring incidents that trusts knew about and used it to promulgate learning nationally".

Ms Williams: We did look as part of our business case at a range of options and one of them was aggregate collection of data. This would not have allowed us to pick up the individual reports on a particular issue. It would have provided summaries of information, statistical information. It would not have yielded the richness of the reports that we have received and that we are able to take action on. On our very first alert, if I could give you that example, we had 40 individual reports on a particular drug which meant that we were able to take action. If it was aggregated information all we would be able to receive would be something like something like 2,000 medication incidents. That is not a basis on which we could have taken action.

Q118 Greg Clark: Your organisation had an objective to be a leader in this field, to promote the profile of patient safety. You were established in 2001. In 2005 77% of junior doctors said that they needed more information on what your agency was about and 60% have never heard of you. Is that a good performance?

Ms Williams: We know that doctors internationally, not just in this country, are a particularly hard group to reach in relation to patient safety and reporting. That is precisely why we mounted a campaign this year to increase the knowledge. We have, through doctors.net.uk, which is a web-based organisation which many thousands of junior doctors in training are members of, found that now 29,000 doctors in training have been through that programme.

Q119 Greg Clark: I was interested in the results of that. I agree this seems a commendable thing but, having gone through this programme of creating awareness, initially 13% thought the organisation would improve patient safety. As a result of people going through the process it rose to 34%. In other words, having been made aware, having been briefed and having gone through a course, 66% of doctors still thought that this was not going to make a difference.

Professor Sir Liam Donaldson: If I could just add on that, we have put into the training programme for all junior doctors from now on a competency on patient safety, so I think that situation will dramatically improve over the next few years.

Q120 Chairman: Why?

Professor Sir Liam Donaldson: Because they will not be able to ignore it. They are going to be tested in examinations of the system.

Q121 Chairman: As I understand these questions that Mr Clark was putting to you, there is too little feedback. People just do not feel that the work of this agency is making a great deal of difference. That is as I understand his ten minutes of questioning.

Professor Sir Liam Donaldson: Ms Williams has mentioned the feedback that they have done. I have regularly, in my Chief Medical Officer's newsletter to all doctors, mentioned patient safety, I have mentioned the role of the agency, I have emphasised the importance of reporting. As I say, it is a world first to have in the training curriculum a competency for all doctors on patient safety. We have a lot more still to do but we are taking some quite strong steps.

Q122 Mr Williams: Sir Nigel, you found succour in table 23, page 67, on the international comparisons?

Sir Nigel Crisp: I do not know if I found succour in it.

Q123 Mr Williams: You were quoting it as showing not unreasonable results as far as the country was concerned.

Sir Nigel Crisp: These were results over a number of years and I think what I said was that I was not quite sure what this showed, and then we went on to Sir Liam talking about international comparisons that he was aware of. I was not particularly taking succour from that. It seems to me that that shows really quite a mixed picture.

Q124 Mr Williams: It is meaningless, is it not, as a table? It has only got two London hospitals representing the whole of the UK and there, when it comes to preventable adverse events, we are second worst out of the whole list in the table.

Sir Nigel Crisp: I am not sure that table offers us much insight.

Q125 Mr Williams: That is all right, as long as you are not shouting behind. I probably misunderstood.

Sir Nigel Crisp: That is what I was intending to say.

Q126 Mr Williams: I just wanted to make sure no-one laid any confidence on that because it is statistical gobbledygook. It is utterly meaningless. The reality is better reflected, is it not, on page 25, and very worryingly reflected, where you deal with the number of incidents?

Sir Nigel Crisp: Yes.

Q127 Mr Williams: And in table 6. I know you must be as worried about this as I would be, but it seems to me when you get a range from almost nil at one end to 13,000 at the other end between trusts, that someone is not telling the truth, are they?

Sir Nigel Crisp: I think the table on page 6 actually reflects reporting practice rather than number of incidents, and that is partly because of the timetable.

Q128 Mr Williams: Exactly, that is what I mean, someone is not telling the truth because they are not reporting, they are not recording.

Sir Nigel Crisp: I have to say some of these trusts only came on to the reporting system during the course of the year in question I think.

Q129 Mr Williams: If you look at the table, the median figure, because you cannot get an average, comes out at 3,700. The worst is three times worse than that at 13,000 in that year.

Sir Nigel Crisp: Yes.

Q130 Mr Williams: Then, at the other extreme, you find people having recorded nothing at all. Either they are unbelievably competent or they are just concealing the truth, or not interested in finding the truth.

Sir Nigel Crisp: Or possibly they are not connected to the system or they became connected to the system during part of the year.

Q131 Mr Williams: It is the median, so you cannot really say that.

Sir Nigel Crisp: I beg your pardon. Yes.

Q132 Mr Williams: You referred to a figure of 50,000 per month and I missed what you were talking about there.

Sir Nigel Crisp: What we are getting at the moment is 60,000 incidents being reported to the NPSA a month. In this document it was about 40,000, which was the figure quoted in this report, which shows how fast it is increasing.

Q133 Mr Williams: So that is 720,000 a year?

Sir Nigel Crisp: Something of that sort, yes.

Q134 Mr Williams: We are told by the NAO and it is reflected in this table that only 24% of the trusts bother to routinely tell patients when they have been involved in an incident. It could happen under an anaesthetic, it could mean you were given drugs which you should not have been given. How on earth can they justify 1:4? Or, put the other way, how can they justify 3:4 not telling the patient?

Sir Nigel Crisp: I agree, I would not want to justify it.

Q135 Mr Williams: So what are you going to do about those? It does have consequences, does it not? If you have not been informed it could well be that it has had a medical effect which is serious to you and you do not even know the hospital was responsible, but also if it is not reported then the GP does not know about it, so in any subsequent diagnosis, looking at this patient, he is unaware something happened in the hospital which could have been the cause or contributed to the new situation. That is very, very worrying indeed, is it not?

Sir Nigel Crisp: Indeed, I think it is, and that is precisely why we are paying so much attention to this, because these incidents have not just happened because we are starting to report them. It is actually important we are starting to report them so we do something about them and pick them up in the ways you are talking about.

Q136 Mr Williams: If we look at the number of incidents - I assume these are the incidents recorded in that table on page 25 - the top number reported is 13,000.

Sir Nigel Crisp: Yes.

Q137 Mr Williams: Are we to believe that three-quarters of those, say 9,000, were not actually reported to the patients? Is that not what follows from the 24% figure?

Ms Williams: We know that because of a range of reasons - fear or retribution, the general atmosphere in which reporting takes place - there are some places where staff are more nervous of speaking up than others and to tell patients, but we have issued a policy in September ---

Q138 Mr Williams: But there are an awful lot of patients out there who have been denied the information they should have had, and in some cases needed to have, and this could have an effect on their future health and also on their rights, because if it had been a preventable incident then they do have a right to take action. It is not antisocial to take action if you have suffered a serious health injury as a result of something someone else could have prevented. What are you doing about all these people who are wandering around unaware they have been the victims of failures by medics? If you have a constituent who complains against a consultant and you take it up with the trust, you are likely to find that the consultant might no longer want to see the individual who dares to complain. It is stacked, is it not, against the patient?

Sir Nigel Crisp: I think there are a number of points here. We are now starting to get this information so we know what is going on. You can see in various places in this report, including the stuff for example on the North East Strategic Health Authority, how they are trying to change the whole system in the North East to make sure actually it is the norm that people report, that you do have a no blame culture and you do get into the position which you are precisely describing of where we want to be. We have also got, as Sir Liam has said, the new NHS Redress Scheme which will make it easier for people to deal with the more minor incidents without getting tied up into legal issues.

Q139 Mr Williams: If they are ever told about them.

Sir Nigel Crisp: I agree with the point you are making. Even though two thirds of these are things which do not actually harm the patient, they should nevertheless in principle know what has happened.

Ms Williams: We launched a policy called Being Open in September, and this year we are running training programmes for trusts using a variety of techniques for staff to be aware of the policy where we are saying very clearly as an agency that you should tell the patient or their relatives, you should offer an apology and you should involve them in the investigation and discuss with them what action should be taken to prevent harm to others. That training programme involves trusts developing their own policy as to who should tell the patient, in what circumstances, and actually what we have found is that staff themselves need training and practice in telling patients. It is a very traumatic thing for example to tell a family that maybe an overdose of medication has been given, and what we have discovered in our work is that clinical staff themselves need support from their organisation and training in how to do that and how to do it well.

Q140 Mr Williams: I am glad that you are trying to address the problem you have identified. Why, though, were you two years late in setting up your system to collect all incidents? What had been the target time for you setting it up?

Ms Williams: The target in Building a safer NHS was to have the system up and running and connected by December 2002.

Q141 Mr Williams: How many months was that?

Ms Williams: From when we arrived in post that would have given us about a year.

Q142 Mr Williams: And you were two years late on that, so that was three years?

Ms Williams: We delivered the system in two years from the date of the Treasury approval.

Q143 Mr Williams: The report says to us it was set up two years late. Is that correct or incorrect?

Ms Williams: As we discussed, we did have a pilot scheme which was evaluated and we found it was not something we could roll out. The Health Service has often been criticised for rolling out pilots and we were very careful to make sure that when we evaluated it we looked very closely to see whether it was a system which would work in all 607 trusts.

Q144 Mr Williams: Did the Department set the target?

Sir Nigel Crisp: We would have set the original target.

Q145 Mr Williams: You got it wrong, did you not?

Sir Nigel Crisp: We did?

Q146 Mr Williams: Either you did or they did. I do not mind which one puts their hand up.

Sir Nigel Crisp: We set a target which was by a certain date this should be delivered. We then set up the NPSA to do it. In the event what happened was we had started a pilot process before we set it up with the NPSA, the pilot was not successful and we evaluated it - and I think I have been criticised by this Committee before for not evaluating things properly - people here did evaluate it and as a result they bit the bullet that they needed to have another scheme that would work better and that is what they have done. That is why it is two years late.

Q147 Mr Williams: This probably looks good in a Parliamentary Answer to say it is up and running in one year but then you sit back and hope everyone will forget how you got to that.

Sir Nigel Crisp: That is not how we handle targets. We have a good record on targets and this particular target was not met by the health system, and we are part of the health system. I am content to be very clear with you on that, just as on other occasions I have been very clear when we have hit targets.

Chairman: We still have a few supplementary questions.

Q148 Mr Bacon: Ms Williams, you said your budget is now £44 million, can you say how much of that is spent on employing staff, on staff salaries?

Ms Williams: Staff at the headquarters, I would say is about £30 million or so.

Q149 Mr Bacon: Do you have people deployed regionally as well?

Ms Williams: Yes, we do.

Q150 Mr Bacon: I was talking about the total. Of your £44 million and your 300 or so staff, how much goes on salaries?

Ms Williams: I would say it must be £40 million or so.

Q151 Mr Bacon: The majority of it?

Ms Williams: Yes.

Q152 Mr Bacon: I did a little sum and 40 million divided by 300 people gives an average salary of £133,000, or an average cost of employing somebody including everything else of £133,000. Is it possible you could write to the Committee with a more detailed breakdown of your budget and, in particular as far as salaries are concerned, stratify them as would happen in a company annual report within each strata of £10,000, so we know the number of people above £30,000 and so on all the way up? Is that possible?

Ms Williams: Yes. I have just been corrected, the staff total is £34 million.

Q153 Mr Bacon: If you could write to the Committee on that, I would be most grateful. The other thing is I noticed in your biography it appears you have had five posts as joint chief executive with Sue Osborn. Is that right?

Ms Williams: That is right.

Q154 Mr Bacon: Since 1990?

Ms Williams: That is right. She is sitting behind me.

Q155 Mr Bacon: Why have you had all these posts jointly together? Do you work-share?

Ms Williams: It is a job share.

Q156 Mr Bacon: So you have consistently throughout your career moved with the other Sue because you job-shared a series of jobs?

Ms Williams: Yes.

Mr Bacon: Okay.

Chairman: You do not want to job-share with me then, Mr Bacon!

Q157 Jon Trickett: Here is a case which I heard of lately. A young woman giving birth was in labour for 32 hours, and for the last 24 hours she was asking for a caesarean which the clinicians told her was not necessary. Eventually the heart beat weakened but still the caesarean did not take place until they discovered that inadequate oxygen was getting to the infant's brain and then an emergency caesarean took place. It is possible to hypothesise three possibilities. One is that the guidance to those clinicians was, "Don't do a caesarean unless it is an emergency". Second, that the appropriate level of decision-maker was not present for the last part of the labour, in which case it is probably not negligence but it is still an accident. The third one is that somebody was negligent. I wonder about this no blame culture, frankly, because the child is now having brain scans to see what damage has been done by the failure of the hospital to do what was right by the mother and child. Supposing it was negligence, how does this no blame culture work? Let me say that the mother is not a very educated person and she just thinks it was an incident which happened because nobody has gone to her and said, "We have made a mistake here, we should have done something else", and no one is tying the child's passive character to the fact this happened during the birth process.

Professor Sir Liam Donaldson: On the specifics of the case, I do not know the detail but we would be happy to investigate if you sent them to us. Speaking generally, having a no blame culture does not mean that nobody can ever be held to an account as an individual. If their conduct has been negligent, careless, incompetent, then clearly there are circumstances in which an individual could be held to account. The problem in the past has been that understanding safety has been posed too much around the individual's role in it and has not acknowledged all these wider system things. Speaking again in general terms about the kind of example you give, you are quite right to say that it could be factors to do with the organisation of services, and so there is a need to acknowledge that individual accountability is not removed from a general no blame culture; it is a question of balance.

Q158 Jon Trickett: So how would the management become aware of this?

Professor Sir Liam Donaldson: Such an incident would be reported as an adverse outcome of care.

Q159 Jon Trickett: But nobody has said this is an adverse outcome of care.

Professor Sir Liam Donaldson: I do not know the detail of the individual case but ---

Q160 Jon Trickett: I was hypothesising three possibilities on the facts of the case. This has happened and it does happen from time to time, but how would the incident be reported? At what point would blame be attributed, if blame was to be attributed? It seems to me it has not been regarded as an incident really.

Professor Sir Liam Donaldson: Such an incident - again not talking about the specifics of this one - should be identified locally, reported locally and it should then be investigated and the causation of it analysed and conclusions drawn and changes implemented to prevent such an incident in the future.

Q161 Greg Clark: A question to Sir John. The National Patient Safety Agency now costs £44 million and some of us have been a bit concerned about some of the practices we have heard about. Have you been able to come to a view through this study whether as an organisation it offers value for money?

Sir John Bourn: I think the view we have come to so far is that you can certainly identify some trusts where there is what I might call a fearless yet sensitive analysis of accidents and improvements have been made, so you can find individual cases of value for money. It is also true that in terms of the Agency we have not yet reached a point where you can say value for money is being secured by it because we do not yet have a national system of analysis and sharing lessons which is fully used, so we have not yet got to the point where we are getting full value for money from the money that is going on the system.

Greg Clark: Thank you very much.

Chairman: I think that is an appropriate place to end. I am afraid I have to say to you, Ms Williams and Ms Osborn, that it is very likely when we produce our report we will be issuing a question mark about whether your organisation does provide value for money, given it was set up in 2001 and the delays there have been in bringing in the system. Thank you very much.