Public Audit Committee

Meeting date: Wednesday, March 13, 2013

Official Report 681KB pdf

---

Section 23 Report

---

“Management of patients on NHS waiting lists”

The first substantive item is continuing consideration of the Auditor General’s report into the management of patients on national health service waiting lists. We have two panels of witnesses. I welcome the first, which is the biggest panel since I have convened the committee. Ian Crichton is chief executive and Susan Burney is director of the Information Services Division of NHS National Services Scotland; Professor Fiona Mackenzie is chief executive and Andy Rankin is head of patient access at NHS Forth Valley; Robert Calderwood is chief executive and Jane Grant is chief operating officer for NHS Greater Glasgow and Clyde; and Gerry Marr is chief executive and Dr Alan Cook is associate medical director at NHS Tayside. The clerks asked the witnesses whether they wanted to make opening statements. Given the size of the panel, the witnesses said that they were happy to move directly to questions from committee members.

I will kick things off. Although the Auditor General was clear that, with one minor exception in NHS Tayside and of course the exception of NHS Lothian, she did not find evidence of manipulation of waiting time information, a key finding of her report was a trend that she could not explain. The trend was that the use of social unavailability codes increased significantly—in fact, it tripled—between 2008 and 2010-11. However, late in 2011, when concerns were raised about the inappropriate use of social unavailability codes in Lothian, their use suddenly dropped in most other boards.

The Auditor General was clear that she had no explanation for what she considered to be a quite noticeable trend. I ask the boards that are represented today what their explanation is for the increase in the use of social unavailability codes and—perhaps even more important—for the sudden drop in their use in 2011.

We have been asking ourselves that question, which is a relevant and pertinent question to focus on and start with. In relation to the rise, we must go back to the introduction of the new ways approach in 2010—it came in from 2008 to 2010. That introduced a new set of rules that had not been in place before.

When that approach was introduced in Tayside, we put quite a lot of effort into training our staff. We held a number of workshops, to which more than 136 staff came. The staff’s understanding improved considerably when the new rule set came in about what the social unavailability rules were. With that training, people had a spread of knowledge about how to apply the rule set. We think that that was the main reason for the rise that occurred.

Because, as the Auditor General said, the reason was uncertain, I asked some of our booking staff—the people who did the coding at the time—about it. They told me that quite a lot of education had been done—certainly locally in Tayside—and they were aware of the rule set. They were starting to apply that, which they felt accounted for the rise. That is directly from the booking staff involved.

I also asked the booking staff what they thought the fall related to. In Tayside, we are now in a different scenario—a post-Lothian scenario in which all the issues have been raised and we have become sighted on things that we probably were not sighted on before. Locally, we have done a lot of training and education about the rule set, to address any misunderstandings that got built into the system before.

I asked the staff why they thought that the numbers had fallen. They said that the rule set was complex. The new ways document contains about 99 pages of guidance over the different sets. Misunderstandings built up in relation to that but, as matters have been clarified through the training and education that we put in place post-Lothian, those have fallen away. That was the staff’s understanding of why the figures had fallen.

I might come back on that, but perhaps the other two boards would like to speak first.

From NHS Greater Glasgow and Clyde’s perspective, I will build on the points that Dr Cook has amplified. There was a general understanding throughout 2008 and early 2009. The true extent of the waiting list as per the new definitions was being understood and arrived at.

Parts of the explanation in Glasgow will be common across the national health service in Scotland.

Going into the beginning of 2011, we had a significantly bad winter and a number of operations were cancelled because people could not get to healthcare premises. In some instances, operational problems meant that theatres were out of action, so we had to relocate and refer patients to different services. However, the underlying trend in Glasgow in the first quarter of 2011 was a significant rise in the number of patients who sought to be treated in a hospital of their choice or by a surgeon of their choice. Therefore, when patients were offered early access to an appointment across the common waiting list, a significant number of them declined and agreed to wait for the hospital or surgeon of their choice. However, as Audit Scotland’s report intimates, there were no data collection systems in those days to capture the data and record the information in real time.

The second part of the question was about why the use of social unavailability codes came down. That was fairly straightforward. In the early part of 2011, the board received a report from the acute operating division intimating the movement in the overall waiting list. With a combination of funding that was released by the Scottish Government and additional funds from the health board, over the period of 2011, we increased our acute elective capacity. Therefore, by the end of 2011, our ability to deal with demand had increased compared to that ability in December 2010.

The figures for NHS Forth Valley were actually relatively steady over that period. The reduction in our figures coincided with the work towards the 12-week treatment time guarantee. We did not really start to have reductions in the use of unavailability codes until May or June and in the lead-up to October.

One reason why the committee was interested in speaking to NHS Forth Valley was that the trend that I referred to was not as noticeable in that board. The Auditor General said that the recorded information that Audit Scotland found in NHS Forth Valley was of a better quality than that in other health boards. I wonder whether Professor Mackenzie is puzzled about why the pressures that in other boards led to a sudden drop in the use of social unavailability codes—whether that was targeting of resources or a better understanding of how it worked—did not lead to that in NHS Forth Valley? Why did the board have a steady use of social unavailability codes?

That relates to our use of unavailability codes. The internal audit report and the Auditor General’s evidence show that a number of factors affected how we used those codes. We had systems limitations, so we were not able to put clocks off and on. Therefore, people were kept on an unavailability code to keep them visible. We did not use medical unavailability to any great extent; in fact, the internal audit report brings out that there was a preference for keeping people on the list when, actually, to be technically right, we probably ought to have returned them to their general practitioners. Issues of that sort meant that, for us, unavailability was used as a proxy for managing a group of patients. Far from being not patient focused, staff were trying to be very patient focused and ensure that people were kept in the loop and communicated with, and that they knew when their treatment would be available. Staff also tried to accommodate people’s preferences—Robert Calderwood talked about that—which are usually to be treated in their local area.

You are saying that, if the new ways system was managed properly with patients at the centre, there was no reason why the discontinuity in the use of social unavailability codes should appear.

In preparation for the 12-week treatment time guarantee, all the systems had to change, anyway. That is what I am saying drove the major change in our use of unavailability as we moved towards October. I am also saying that our local use of unavailability reflected the practical issues that we had, but people recorded that use fully and well, which was the point that was brought out in the exchange with the Auditor General.

Dr Cook, you said that the drop in the use of social unavailability in 2011 was because you were entering a “post-Lothian” period. It is hard not to take that as a suggestion that the unacceptable practices in Lothian were revealed at that point and then you in Tayside changed the way in which you were using social unavailability.

09:45

That is not a fair reflection of what was said or what was happening at the time. This is all related to the degree of scrutiny and to the training and effort that is put in at different times over a cycle. Issues were not raised with us around social unavailability in the lead-up to the Lothian position. You can see from the information that came to us from ISD that the matter was never raised with us at meetings. Social unavailability was never raised with us as an issue at meetings with the Scottish Government access support team. Social unavailability was never raised at any of the national waiting times meetings pre-Lothian, certainly to my knowledge. It is the post-Lothian position that I am referring to as a different scenario—when people became acutely aware of the issues of social unavailability. That is what I am trying to describe, rather than there being a difference around the matter. The issue became very apparent to us, in a way that it had not pre-Lothian.

I appreciate that—and, believe you me, I am trying to be fair.

Yes—I am sure that you are.

However, I think that you are saying that, when the unacceptable practices in Lothian were discovered, you then trained your staff not to use those practices, and that is why your use of social unavailability codes dropped, as in Lothian.

No. That would not be a valid conclusion, because TTG legislation was coming in and more training was being offered around that time, linked to the new rule set.

The training programme has been running since 2010. When the situation in Scotland emerged through Lothian, it would have been wrong not to focus on it more strongly in the local systems. It was the sensible thing to ask whether there were lessons to be learned in any system in Scotland from what emerged in Lothian.

After the report on the situation came into the public domain, we did a lot of work with our staff. We undertook a major culture survey through our organisational development people. We engaged with all our coding staff and told them that we needed to hear what things were like, where they thought training was deficient and where the challenges lay around the complexity of the system. I commissioned a report on that, and I received it from my team. According to the report, the focus groups among our team most often said that they were proud, that they recognised that the job was tough, with some pressure, but that they unequivocally did not generally feel any pressure from managers to do anything that would be considered to be something other than within the rules.

To me, that reveals a sensible and accountable approach as a response to a problem that was exposed in the Scottish health service. I am sure that all the boards satisfied themselves that there was nothing untoward from that point of view.

The application of social unavailability codes in Tayside pre-Lothian was closer to the position at Forth Valley. The September 2011 unavailability figures were about 26 per cent across Scotland; we were running in the high teens. We did not have the rise that was mirrored across Scotland.

The key point is that NHS Forth Valley did not have the drop that Tayside did. I appreciate your point, Mr Marr, about staff not feeling pressured, and my committee colleagues might wish to discuss some issues in that regard. However, I cannot get away from the fact that you appear to be saying that you were using social unavailability codes in one way in 2010, and that in 2011, after training and consideration of the Lothian experience, you were using them better and differently, as a result of which far fewer patients were considered to be socially unavailable. I cannot see what conclusion we can reach except that, prior to Lothian’s practices being discovered, those same practices were taking place—not as manipulation but as an inappropriate use of the codes—so more patients were on that code. I cannot see what other conclusion we can reach from your evidence.

I do not know the numbers for other boards, but Audit Scotland and the internal auditors examined 50,000 transactions over a total of 170 audit days, which is 25 per cent of the total audit days assigned to internal audit. That was a deep, forensic dive into data. In our case, the audit raised a suspicion on 63 transactions, which is 0.2 per cent.

I think that it is reasonable to say that, while we were populating the system, training people to use it and developing and improving it all the time, that is a natural feature of a system that is implementing the new ways approach. We can put that against 50,000 transactions and the fact that Audit Scotland did not draw attention to NHS Tayside in its report. That is the evidence that suggests that, over 50,000 transactions, there was not an apparent problem in the use of social unavailability codes.

No; that might be evidence that there was no manipulation. However, what Audit Scotland drew our attention to was the discontinuity—the sudden drop in the use of social unavailability codes in 2011. I am trying to get at why that happened in Tayside. Dr Cook’s evidence was that it was because we had moved into a “post-Lothian” situation—in other words, an understanding had become available of the inappropriate practices going on in Lothian. Surely the only conclusion to reach is that the misapplication of codes was also happening in Tayside but that it was corrected at that point and the use of social unavailability codes fell.

I think that what was said about Tayside in the Audit Scotland report was that there were areas of misunderstanding of the rules. The rule set is 99 pages long and the rules are complex; staff must get their heads around what the different parts of the rules mean. I think that there was misunderstanding of what the rules were. Like you, convener, Audit Scotland said that manipulation was not identified, and I agree with that. There was a tiny little pocket in Tayside but, at a whole-system level, I think that some misunderstanding built up around the rule set. That was clarified in a series of education and training events that we put on to ensure that staff understood the rules.

Early this year we had a series of talks for our waiting time staff, of whom 219 out of 250 attended. We had two-hour long workshops that all the staff were able to come to. We developed a LearnPro electronic module—it is one of the electronic teaching aids—that goes through the rule set. It is interactive so that staff can answer questions and get feedback at the end. That has been completed by 632 staff. The understanding of the rule set is much clearer now than it was before. I think that there had been some misunderstanding previously. However, like you, convener, I do not think that there was widespread manipulation of the system—that is not what we found.

Just to reiterate the findings, Audit Scotland looked at the nine-month period from April to December 2011 and had no concerns about any of our data. We were not one of the boards that showed any unusual trends or one of the boards that was looked at. Audit Scotland did not come to speak to us in that regard, and it looked at 30,000-odd of our transactions.

Okay. Thanks. While we are on unusual trends, I want to ask a quick question of Mr Calderwood. Audit Scotland identifies that at one stage in a particular specialism—orthopaedics—in a particular hospital, 70 per cent of patients were deemed socially unavailable. I think that your argument in your first answer was that that was because patients were exercising a choice about where they would be treated. Why do patients in Glasgow express such a remarkably high level of choice about where they would like to be treated? It is orders of magnitude different from anywhere else in Scotland.

I could spend a long time explaining how Glasgow citizens regard access to their local hospital as critical. I am sure that Ms Baillie knows as many of those arguments as I do.

The point that I made, which I want to evidence, is that in the period in question, patient choice was part of the rise in social unavailability. Let me take the specific reference to Glasgow in the Auditor General’s report. It was highlighted that in April 2011, 924 patients were on the waiting list at the Western infirmary under the term “socially unavailable”. Had Audit Scotland picked July 2011, it would have found that the number was 343. There was a very selective approach, whereby one waiting list was picked out of eight, in one month.

The movement between April and July 2011 is consistent with the point that we made about the board investing additional resources and seeking to address the choice issue, which was becoming stark in the first quarter of 2011.

Let me try to put the Glasgow position in the wider context of the Auditor General’s report. In 2010-11, NHS Greater Glasgow and Clyde added 467,000 patients to our out-patients waiting lists, and almost 146,000 in-patient day case treatments were performed. In 2011-12, out-patient attendances rose to 475,000 and actual surgical procedures rose to 148,000. I contend that that is consistent with the board recognising the growing demand and putting in additional capacity.

On the point about lessons from Lothian, the Lothian experience became public knowledge in the NHS in February and March 2012. If we look at the Glasgow figures, we can see that in March 2010 social unavailability across all our waiting lists was 25.6 per cent. That rose to 36.9 per cent in March 2011, at which point we became concerned about the issue. By March 2012 social unavailability had dropped to 17.3 per cent. Therefore, long before the Lothian information became available and long before the Auditor General came to speak to us about 2011, Glasgow had identified a problem and addressed it.

That is fundamentally our position. There was an issue, which we believe was driven by patient choice. We have gone out of our way to try to address it. It remains a challenge.

I am sure that members from Glasgow will want to pursue some of those points, but I want to bring in Mary Scanlon.

First, I want to make a point about the evidence that we heard at our most recent meeting, because I would not want anyone on the panel to think that Audit Scotland carried out a minuscule study that involved one or two patients. Audit Scotland told us:

“this has been the most data-rich, data-intensive exercise that we have done.”—[Official Report, Public Audit Committee, 27 February 2013; c 1226.]

Therefore, please do not minimise what was done. Barbara Hurst also said that the patterns looked unusual in 3 million transactions, so I have no patience with anyone who says, “Well, it just went a little bit wrong here and there.”

I want to ask about how the problem was found out. As the convener said, the use of social unavailability codes rose from 11 per cent in 2008 to 30 per cent. Also, according to Audit Scotland:

“23 per cent of inpatients across Scotland had an actual wait ... of over nine weeks, compared to three per cent with a reported wait of over nine weeks.”

Why was not the issue picked up by internal auditors, Audit Scotland, ISD, which has a quality assurance role, health boards, the Scottish Government or non-executive directors of health boards? Why did no one pick it up? We are here today because someone was brave enough to be a whistleblower.

Who wants to start?

Perhaps ISD will start, given its quality assurance role, which, according to Audit Scotland, includes raising

“any concerns about their data, such as differences from what would be expected based on previous quarters”.

Why did you not issue warnings?

I think that one needs to go back to 2008 and the beginning of new ways. That is the context. The starting point for NHS Scotland was in managing waiting times in a totally new way—hence the name, “new ways”. Prior to that, social unavailability, as you know, was not tracked and put into reports in the way that we currently track and report it.

Therefore, we took the view from the start that, clearly, social unavailability would rise because it was a new thing that we were measuring. We believed that there was a logic that, over time, as board waiting times reduced, essentially the pressure on unavailability would rise. As well as the exercising of choice, which was mentioned earlier, we expected that various dynamics would come into play, so we were not surprised when we saw social unavailability rising. We were aware of the social unavailability rises through the period. We were in touch with, I think, three boards about the social unavailability increases that we saw. There was nothing in any of that that we regarded as untoward, or anything that we had as a concern.

10:00

We have gone back—hindsight always makes us all very smart—and looked at our risk registers at the time. The organisation’s focus was on ensuring that the statistics were as good as they could be. The focus of the health service was on pulling off a pretty significant change in patient waits. At the time, a lot of the focus was on ensuring that the reporting systems that were being deployed to support the information were working in the way that they should.

It is easy with hindsight to look back and say, “Why didn’t you make a big deal about social unavailability?” I have no evidence that we thought that social unavailability was a big deal. There is no evidence of discussions to say that there was anything wrong there. There was no reason to have any belief about impropriety.

At the time of the change to new ways, we were told in the Parliament loudly and clearly that new ways was the end of hidden waiting lists. However, new ways found a new way of hiding waiting lists, because the use of social unavailability codes increased from 11 to 30 per cent, with 23 per cent of patients having an actual wait above the target when only 3 per cent were reported as doing so. Despite politicians in the Parliament being told that there would be no more hidden waiting lists, the social unavailability codes were being used—by everyone round this table—to further hide patients to make the figures look good. Do you agree that it was a new way of hiding waiting lists?

No, I absolutely do not agree. The major difference was that, in the reporting and in the information that we have published, those social unavailability figures were very clear, so they were there for all to see. Our waiting list publications website gets hits from 55,000 people a year, so 55,000 people have been able to see the growth in this thing over time and the reduction in this thing over time. That has been remarkably transparent; I would say that you would be pushed to find another health system that would give you that degree of transparency.

Do you think that it is transparent that 23 per cent of patients had an actual wait of more than nine weeks but only 3 per cent were reported as doing so? Is that transparent?

What is transparent is that the percentage of patients over time who were socially unavailable was visible as opposed to having been put somewhere else, which is what the former system did.

No warning bells rang with you when social unavailability increased threefold. Do you think that that is transparent as well?

Let me say two things. First, it is easier to see that looking back. The retrospective adjustment that was made as we went through the period masked the issue to some extent, so I do not think that we would have seen the issue as starkly at the time. Secondly, when we asked several boards why social unavailability was increasing, technical and operational reasons were usually given for that.

Audit Scotland said that alarm bells should have rung. Given the figures that have been quoted, should alarm bells have rung in any health board? Although NHS Lothian was found guilty of falsifying and manipulating its figures, there was not sufficient evidence of that in the rest of the health boards—it is not that they were not guilty, but that there was insufficient evidence. Did alarm bells ring in any health boards or in ISD Scotland?

I can speak only for us. In terms of alarm bells that someone was doing anything dishonest or untoward, there were none. There was no reason to believe that.

Not even with NHS Lothian, which was found guilty of manipulating the figures?

Not even with NHS Lothian. We had discussions with NHS Lothian about where it was. We were told that a lot of its issues revolved around systems.

It is important to understand that every health board in Scotland, at chief executive level, had to sign off on its submission to ISD. From where we were standing, we had absolute assurance from each board that the figures that it was reporting were correct.

Returning to your point about transparency, the Statistics Authority did a review of our waiting times practices. I think that that was in 2010. The review told us that we had done a decent job of reporting waiting times and making what was being done publicly clear. The Statistics Authority was keen to use us as an exemplar elsewhere in the United Kingdom. We have to be able to separate activity going on in boards, and the way in which boards govern themselves, from ISD’s role as a keeper of national statistics.

In hindsight, you are quite content with the role that you have played in quality assurance and that people can have trust in your figures. All I can say is thank God for whistleblowers.

I move on to my second point. When the waiting time target moved from 18 weeks to 12 weeks, I am not aware that huge resources were given to health boards. I appreciate that you were undergoing efficiency savings, which were going back into front-line care. Was the reduction from 18 weeks to 12 weeks an impossible task that led to looking at how figures could be presented using social unavailability codes? Was the manipulation, and indeed the muddle of the waiting lists, a consequence of Government policy or should you have asked for more money to achieve the targets?

From a health board perspective, if we look at the support that we got in NHS Tayside, we would see reducing waiting times as important. It is important for patients to be seen as quickly as they can be.

We all agree with that.

It is one of the elements of a quality service. A quality service is person centred, safe and timely—that is waiting times. It is about equitable services, efficiency and being effective. It is a core component of a quality service. Intrinsically, we would like to have waiting times as low as they possibly can be.

On resourcing, we have had support from the Scottish Government. NHS Tayside has had recurrently about £5.3 million specifically to look at waiting times and address related issues. In addition, we have allocated £3.9 million recurrently from our budget to look at waiting times. In resourcing terms, from our finance department, about £9.2 million has been allocated.

We consider that appropriate for the reasons that Dr Cook has just demonstrated, in terms of equity in access. I would not describe the challenge as impossible. It will always be difficult. It should be difficult, because we should always strive to get waiting times down as low as we can. The figures from NHS Tayside that we quote are no doubt replicated in my colleagues’ boards as an appropriate use of resources to achieve the 12-week target.

If I take it at a higher level, ISD’s published statistics include the target level—the performance after clock stops are deducted—versus the total patient journey, both of which were very evident to the board. In NHS Greater Glasgow and Clyde, 92 per cent of all patients were treated in under 12 weeks in 2010-11. That includes all clock stops. That figure is for in-patient day cases; for out-patients, it was 93 per cent of patients, irrespective of clock stops. Their total journey was less than 12 weeks.

That figure of the total experience of the patient was available and was monitored. Under new ways, we are allowed to adjust that total performance to reflect our performance against the target; in other words, where patients decline the opportunity or are medically unfit to receive the treatment, they are deducted from the targets. Hence, we get to the board’s performance against the incrementally improving waiting time targets that have existed for the past 10 years.

From a financial perspective, I suspect that over the past 10 years, between new, targeted Government money and reinvestment of efficiency savings, we have invested tens of millions of pounds—although the figure is probably closer to £100 million—in increasing our acute capacity so that people can get treatment quicker.

I genuinely believe that during the time that we have been moving to introduce the targets, the high-level visible performance of the NHS in Scotland has improved and moved towards the targets. As I tried to explain, that is set against a backdrop of rising demand, which has had to be factored in, and inevitable service disruption, which has to be caught up on so that patients are not disadvantaged.

The other point that has been highlighted but which did not come out so much from the Audit Scotland report is that a considerable number of patients who were socially unavailable when Audit Scotland was looking could, under the rules, have been returned to their GP. It was a conscious decision—certainly in NHS Greater Glasgow and Clyde and, I believe, a number of boards across Scotland—not to disadvantage those patients in recognition of their choice but to keep them on the waiting list so that they could then be treated. However, under the technical rules, we in NHS Greater Glasgow and Clyde could have discharged a significant proportion of our patients who were coded socially unavailable back to their GP.

I would like to come in on the activity change and the resourcing that Robert Calderwood said had been put in. Back in 2008, across Tayside we carried out about 25,000 in-patient procedures. By 2012, that figure had gone up to about 30,000, which is about a 20 per cent increase. The resourcing that came in was put into additional staff. Figures from our human resources department on consultant staffing in Tayside show that back in 2008 we had 352 whole-time equivalent consultants. In 2012, we were up to 411, which was an increase of 59 whole-time equivalents—about 17 per cent. That matches the sort of activity increases in day cases that we have seen from 25,000 in 2008 to about 30,000 in 2012. The resourcing was there and it has been allocated appropriately, as Gerry Marr said. We are delivering increased activity to get through additional operations.

Quite a bit is said about NHS Tayside in the case study on page 22 of the Audit Scotland report, but I will not go into that.

My next question is about staff being under pressure, which is highlighted in the NHS Lothian case study on page 21 of the report. Audit Scotland refers to boards not having the

“capacity to treat patients within waiting time targets ... and a management culture of not wanting to report bad news.”

Paragraph 60 mentions

“accusations of a ... bullying culture”.

We know that that affected the staff. Where did the fear and pressure come from? Why was management unable to report bad news? Why did Audit Scotland say that staff were under pressure and that boards did not have the capacity to treat patients? Where did the bullying and fear come from and why were staff scared to report bad news, as reported by Audit Scotland?

I am happy to kick off on that. Our audit report made clear mention that the culture was open and that staff did not feel that they were under any pressure. I want to convey that staff were trying very hard. Robert Calderwood gave a good example of the way in which people were being kept on waiting lists when they might technically have been sent back to their GP. Staff were trying very hard to accommodate people and were trying to take into account people’s wishes to be treated locally, as opposed to having to travel. In our case, people generally have that preference.

Staff were under pressure because they were busy, but certainly no pressure was being brought to bear on them to behave in an untoward way. In fact, people were supported to deal with any of the challenges that they had.

To be fair, Forth Valley was mentioned time and again in the Audit Scotland report as an exemplar of good practice. We wanted you to come along so that we had an exemplar of good practice, but we also wanted to drill down to where staff were being affected by that culture, so thank you for that.

I would like to hear from some of the others.

10:15

I do not recognise those issues of a bullying culture and a fear of reporting an inability to deliver a target. NHS Greater Glasgow and Clyde can demonstrate that waiting list information, along with all other patient quality information, is debated publicly every month at its public meetings. Those reports are provided throughout the system. I do not recognise where the Auditor General and Audit Scotland detected those issues within Greater Glasgow and Clyde. Indeed, the report does not make any specific mention of an example of such a bullying culture.

A bullying culture is mentioned in paragraph 60 in relation to NHS Lothian:

“Evidence that staff in NHS Lothian were put under pressure to falsify patients’ waiting times has resulted in accusations of a more widespread bullying culture in the NHS.”

You all seem to be in denial about the Audit Scotland report. Is the report accurate?

On your point about a bullying culture, I will expand on what we did when the Lothian report was published. In conjunction with our employee director and our HR department, we set up a series of workshops with those of our staff who worked in the departments that were responsible for waiting times management. We did that in a way that we think was appropriate, given the number of people. There were a whole series of workshops with 340 staff in total, of which about 50 people were accompanied into the workshops.

The findings from that series of workshops were the opposite of what you describe. I can quote from them. One question was, “How does it feel to be working in Tayside?” and 13 out of the 26 people in one workshop said, “Proud.” Another 13 said, “Busy.” The other information that we got from that report, which we made available to our internal auditors, was that staff appreciated the fact that the chief executive, the chief operating officer and the senior managers took the personal time to inquire about the wellbeing of the staff and to ask whether they felt pressured and whether they felt bullied to do something that was inappropriate. The answer was an unequivocal no, so I challenge that point about a bullying culture.

The way in which Audit Scotland has extrapolated the Lothian experience to other boards is not for me to comment on—I have no knowledge of how it might have done that. All I can say in good faith is that we very properly conducted an exercise with our staff and our staff gave us the opposite message to the one that you raised.

My final question to the health board chief executives is whether this is an accurate report. Has Audit Scotland recorded the information accurately within the report? Do you agree with the contents of the report?

Audit Scotland has reflected a situation that it found at a moment in time in the NHS in Scotland by looking at data retrospectively. It has tried to extrapolate that into a situation that was identified more recently in Lothian. The way in which the information has been portrayed—Audit Scotland is unable to determine whether social unavailability was properly applied, when there was no requirement in real time for boards to record that information—is one interpretation.

It is clear that Audit Scotland has not identified manipulation across the NHS in Scotland. As I showed with the orthopaedics example, if the report had used the July 2011 figure, the convener would not have quoted the example of orthopaedics at the Western infirmary. Indeed, the other example in Glasgow that Audit Scotland chose to quote is ophthalmology. The report quoted the month of May 2011, saying that 40 per cent of the waiting list was socially unavailable—145 patients. If it had quoted the month of July, that number would have dropped to 42. It is not for me to determine why Audit Scotland picked particular months and did not show an extrapolation.

Mr Calderwood, I want to be clear on the fairly central issue of Audit Scotland’s inability to find out why social unavailability had been used in specific cases. You are saying that that is because you were not required to keep that information, so you saw no reason to do so. You think that that is an unreasonable—

At that particular time, none of the information technology systems that NHS Greater Glasgow and Clyde used allowed staff to capture information about the application of the code.

That was not what I asked. You were not required to keep information about why patients had been deemed to be socially unavailable. Was that the gist of your evidence?

I said that we did not routinely capture those data in 2011, so to look for them in 2012 and expect to find them was always going to be challenging.

But Audit Scotland’s criticism is that you did not routinely capture those data.

That is correct, but there was no requirement for me to collect them, so why the criticism?

In your view, the failure is one of the Scottish Government’s management of the NHS because it did not require you to collect that information.

I have to be clear, Mr Gray. During the period in question, I do not believe that the NHS in Scotland, particularly NHS Greater Glasgow and Clyde, was failing to provide speedy and equitable access to services. I have sought to demonstrate the total journey times of patients as published and retrospectively audited by ISD. We were seeing in excess of 92 per cent of people within 12 weeks for in-patient, day case treatment, including the clock stops.

The purpose of any audit is to identify system weaknesses. If you ask me about our internal audits plus those of Audit Scotland, the answer is yes—Audit Scotland identified weaknesses in the system, and that is self-evident. An action plan containing 22 actions was published, and I know mine because our audit committee and board accepted the plan and that those weaknesses existed. Those action points will be completed by the end of March. That is the purpose of any audit and what I expect to emerge from a forensic audit of any system. We welcome it, because it leads to improvements to the systems that we have in place.

Whether the results of those audits should be extrapolated to cover issues such as harassment, bullying and manipulation of figures, or whether a certain level of unavailability means de facto that the unavailability code is being applied inappropriately is much more complex. I accept the veracity and accuracy of our audit and that our system had weaknesses, but I offer the evidence that I have found in our own system and say that we do not concur with what has been articulated this morning. We have sought evidence in order to come to that view.

I only quoted from the report.

I understand.

I seek clarification from ISD, because I want to be really clear about this. Mr Crichton, you gave evidence that the extent of your quality assurance of the waiting times figures that ISD provides regularly is that the figures that are given to ISD have been signed off by the NHS board chief executives. In this case, over the period of time that we are considering, the chief executives signed off those figures, so you are entirely content with the extent of ISD’s quality assurance of those figures.

It is an element, but it is only one element. I will ask Susan Burney to say a bit about how we validate, verify and quality assure our figures, because that goes beyond what you describe. I said that the signing off was definitely an extra element that we asked for, and it was quite unusual.

In response to Mrs Scanlon, you said that it was the fact that those figures were signed off by the NHS board chief executives that meant that you were content that you had checked them.

Then I have not been clear. The point that I was trying to make is that that requirement goes above and beyond all the normal checks and balances that we have in place. We would not rely purely on the chief executive’s letter for the purposes of deciding whether, in our view, the statistics were accurate.

Mrs Scanlon was asking about what ISD did to find out that they were accurate.

It will be better if Susan Burney takes you through the technicalities.

We routinely look retrospectively at the data that come to us for any unusual patterns. For example, if one board’s figures are an outlier, we will contact it and ask it not so much for an explanation, but to confirm that the data are correct. What we are really looking to do is to ensure that what we have been given is correct and there has not been a mistake somewhere in the submission of the data. We have a list of data quality things for each board, some of which are minor. Over the years, as the data have improved, the number of data quality questions has reduced.

In general, our process is to look for anything unusual and go back and say, “This looks unusual to us. Can you tell us why it might be and confirm that it is okay?” If there is something unusual and there is a data quality issue, you will find that our website has a section on data quality, and for each board there is a list of things that people ought to keep in mind when using the data. Because data are never perfect, we have a duty to help people to understand it where we know that there is something that they will need to keep in mind, whatever use they might be making of it.

I think that Mrs Scanlon’s point was that Audit Scotland said that there was an identifiable unusual trend that should have rung warning bells. The question was: why did you not notice that?

We noticed that Lothian was unusual and we talked to it about the fact that, compared with other boards, it was different. The explanation that we were given was one of an operational nature. Around that specific issue, we had no reason to question further. The board gave us an explanation, and that was straightforward.

But Audit Scotland specifically says about the trend in the reducing use of social unavailability in Lothian that the same pattern was demonstrated in other boards around Scotland.

There seemed to be a general consensus at that point that rising unavailability was due to a number of reasons, which people have rehearsed here today, around the introduction of the new ways approach and the reducing waiting times putting more pressure on patient availability, including their early availability. There was a general consensus in the health service and in Government around there being plausible reasons for that.

It is important to look at the context. At the time, social unavailability was a small element of the overall waiting list transition that we were managing. Most of the focus was definitely around achievement of the waiting time targets, progress towards future targets and so on, and not around social unavailability, which was a small part. It was a part that we were looking at, but it did not figure on any of the risk registers as something that people were extremely worried about. The term was not in widespread use.

One thing that you majored on earlier was this business of “since Lothian” and social unavailability. Since Lothian, the whole social unavailability thing has taken a much higher position than it previously had in terms of general consciousness and terminology. Again, I say to you that, as well as our not being concerned about it, more than 50,000 people go through the website every year and they can see the list of things that we have highlighted as being worthy of keeping an eye on, which Susan Burney mentioned. It was there for people to see. We did not know—

It was there for you to see.

Sorry?

It was there for you to see.

We saw it, and—

You have just said that it was all about achieving the targets, but what we want to know is why the figures that were presented to us were not about achieving the targets. It is not just about a top-line figure. You have a responsibility and a quality assurance role in monitoring the waiting list. I think that I have said enough, but it is not only about achieving the targets. It is important to look at how they were achieved.

I want to be clear. My organisation is not responsible for achieving the targets. There is a letter of understanding between me and the director general for health, and it is quite clear that matters of statistics are at arm’s length and we are neutral. Our role is around providing clarity on performance against targets and giving the public confidence that the numbers that people look at are solid. That is where we spend our time.

If we look at the questions that were coming in from politicians and the press and at where everybody was looking, the focus was not on social unavailability but on waiting times and how the system was performing.

10:30

I call Mr Doris. Sorry—you have been waiting a long time.

I will ask some specific questions of the ISD witnesses, but I first want to ask Mr Calderwood some questions about NHS Greater Glasgow and Clyde. If in the period during which Audit Scotland was reporting, someone from Greater Glasgow and Clyde was deemed to be socially unavailable, did they remain on the waiting list?

Yes.

Do you have any figures that you can give me on whether, irrespective of whether they were socially or medically unavailable, their 18-week waiting time guarantee was fulfilled? What percentage of all patients, including those who were socially or medically unavailable, still had their 18-week guarantee met?

In relation to the stage-of-treatment targets that applied during 2011, if you took the absolute backstops and recorded all the clock stops—this is the point that I was trying to make in response to an earlier question—92 per cent of the 146,000 in-patients and day cases that we treated in 2010-11 were treated within the 12-week timeframe. At the time, that was the stage-of-treatment target. Irrespective of clock stops, 92 per cent of people were treated within the 12-week guarantee. Of the 467,000 out-patients, 93 per cent had their appointment within 12 weeks, which at that point was the stage-of-treatment target.

Does that include the numbers in the Audit Scotland report, which refers to 900 orthopaedic patients at the Western infirmary and 145 ophthalmology patients at the Southern general hospital?

It does indeed. A percentage of the ophthalmology patients, who were out-patients, and a percentage of the in-patients may have fallen into the 7 per cent who were not seen within 12 weeks. However, if I were to move that data set up to the 18 weeks, which was the backstop guarantee, the position in Glasgow is that 97 per cent of all in-patients and day cases were treated in less than 18 weeks throughout 2010-11, including clock stops. For out-patients, the monthly figure is that something like 98 per cent of all patients who were referred to Glasgow were seen within the 18-week backstop.

The rise or fall in social unavailability must be seen in the context of how it affected the eventual journey of the individual patients. I contend that the performance of the NHS in Scotland, when looked at in absolute terms, is to be commended. I certainly think that the work that is done in NHS Greater Glasgow and Clyde by the clinical and managerial staff is something that they recognise and are quite proud of.

We will shortly come on to how you deem someone to be socially unavailable. Whether there is a good-news story or a bad-news story in Glasgow, the problem seems to be that it was not reported properly at the time, for whatever reason. We will come back to that. Were all the statistics and data that you have given me published?

Yes. Those statistics come from the validated ISD data sets.

Right. So, they are not hidden.

No, they are not hidden at all. The point that Mr Crichton was trying to make is that the data are publicly available through ISD and the website.

NHS Greater Glasgow and Clyde definitely has to improve, but I want to look at that snapshot in time. What would have happened before the new ways system was implemented? Would patients have remained on the waiting lists, and would that have been reported publicly?

Before the introduction of the new ways system in 2008, there were what were referred to as hidden waiting lists—people were not on the active waiting lists but were on another waiting list. From 2008, the patient either has been on the waiting list—which is very publicly available—or has been returned to their general practitioner. There is now no hidden waiting list. Whether someone is medically or socially unavailable, those data are collected and published.

Okay. If I am an orthopaedic patient in Glasgow and I am offered a procedure at the Southern general hospital but I want to go to the Western infirmary, do you tell me that that is fine or that it is not fine? What do you tell me at that point?

I ask Mrs Grant to answer that.

At that point, the patient will generally have been seen as an out-patient by a particular person and may have been listed for surgery. If they were at the Western and there was no space within the guarantee at the Western, we would generally phone them to say, “We have a reasonable offer for you”—in essence, seven days’ notice—“at the Southern general.” The patient may then say yes or no to that offer.

If the patient says no, they receive a second reasonable offer within seven days. That may be on the same site or on a different one. If they decline those two reasonable offers and say that they want to stay with the consultant they have already seen at the out-patient clinic—that is generally what happens if they have already seen the consultant—it is explained to them that we would be unable to accommodate that within the guarantee. Therefore, they have a choice: they can choose to wait, or they can choose to go elsewhere, given that, at that point there would be a slot elsewhere, because we run eight orthopaedic sites across Glasgow and Clyde where orthopaedic in-patient and day case surgery takes place. That is how we work.

Okay. That sounds like patient choice; it sounds fine. At that point, whose responsibility is it to record that information? Is it done right away or in batch form? Do you wait until there are 100 patients to put into the system at the same time, or is it an individual’s responsibility at the time to enter those data into the system, so that they can be audited? Who does that job?

The person who makes the call to the patient would record that. There has been no uniform recording on the IT systems. As Mr Calderwood outlined, our systems did not accommodate that well. The information was usually recorded manually and those records are not as robust as they might have been—that is a fair point. As we move forward with TrakCare we will be implementing the system in Glasgow and Clyde in a more robust fashion.

When you say that it was recorded manually, do you mean that it was written on a bit of paper and put in a file rather than on a computer system?

Generally, the waiting list co-ordinators would keep a spreadsheet. However, that was not in the IT systems and therefore available for scrutiny in the way that we are now trying to make information available.

When was it eventually put into the IT system for that period?

The outcome would have been input in terms of the date, but the evidence has proved that the actual manual recording of, for example, “I had a conversation with Jane Grant on 24 June,” was not in the IT systems, because they were not capable of doing that uniformly at that time.

Okay. What training would the staff who were entering that information have had on what they should have noted in the hard-copy files before looking to see what they could put into an IT system? What training was available?

Our staff certainly have been trained in new ways—the green book, as we refer to it, which was the original guidance. As Tayside has indicated, we have gone back and reinforced the training on the issues, particularly in new ways. Also, as we move forward in the process to TTG, compliance and the new circulars, we are about to start in April an electronic version of the IT training programme so that all our staff can routinely and regularly access it. We can keep that up to date as we go forward. We have put in a lot of effort to ensure that the training is there and can be evidenced and accessed more appropriately, particularly for new people and the large number of people who have to manage waiting lists.

That is one of the other issues: there are a significant number of people in the system who are required to manage waiting lists. The rules are complex and lengthy and it is therefore important that we have uniformity. It is a complex scene: even the Audit Scotland reports—the current report but also the 2010 report—describe the rules as complex. They are complex, and for a large number of people who are accessing and managing waiting lists, that is a challenge.

I understand that. I ask the question because I want to know whether the person who was recording the information at the time had been trained and told, “When you write down ‘socially unavailable’, you should give a reason at that point.” Have they been trained and told to do that? Were the people recording that information also recording information before the new ways waiting times came along, when the routine culture was just to put “unavailable” as the status code? Was a culture change needed within Greater Glasgow and Clyde in how information was recorded? People were working under a system where one in three patients was deemed socially unavailable and on a hidden waiting list, and then new ways comes in and the same staff are asked to do something different. At that point, were they explicitly told that they had to give a reason, whether on a bit of paper or in an IT system? Was there a cultural or training issue? I will come on to what is happening now, but at the moment I am asking about that point in time.

Five years ago, when the new ways system came in, there was undoubtedly a transition. People were trained on new ways but, to be perfectly honest, it would be hard to be absolutely explicit about who recorded what in 2008, which was five years ago. Undoubtedly, there was a requirement to apply periods of social unavailability, but at that point there was no requirement to describe in detail why those periods were applied—it was just recorded that somebody was socially or medically unavailable.

My constituents in Glasgow will be asking about what will happen if they need a surgical procedure today in NHS Greater Glasgow and Clyde. Let us assume that the same discussion takes place with someone who wants to go to the Western but who has two offers at the Southern, and it is explained to them that, although the waiting time clock will stop, they will still be seen at the Western. What happens to that information? How is it recorded today?

Today, we have the TrakCare system in parts of NHS Greater Glasgow and Clyde, which can record that. We are implementing that system throughout the health board, and it will be available uniformly by the early summer. As required by the TTG circular, we send letters to patients to describe the situation. For example, if we agree a period of unavailability with a patient, they now get a letter that describes that. If they are unhappy with that or do not understand it, there is a process by which they can contact us. Patients now receive confirmation in writing, which did not happen previously.

Are you confident that every person in NHS Greater Glasgow and Clyde who has responsibility for recording such information knows that, with the TrakCare system, they have a duty to record more than just that the patient is socially unavailable? Are you confident that staff are trained and are aware of that?

We have certainly put a big emphasis on that. I have described the current training, and we will do more in future, which will be comprehensive and will involve an e-learning package. In addition, we are reinforcing the rules. From 1 April, we have plans to implement a full audit process under which every month a number of records in the board area will be validated, independently from the directorate teams. In addition, the directorates will be asked to scrutinise and audit another cohort of patients. We will therefore have a transparent monthly audit report that describes the situation. That process shall be in place from 1 April.

It is incredible that what could be a good-news story for NHS Greater Glasgow and Clyde—it has worked towards meeting waiting time targets and extending patient choice—has turned out to be presented as a negative, because of the inability to record information appropriately. Two weeks ago, the Auditor General said that one advantage of targets is that, when they are not met, rather than ring alarm bells, that should inform a health board that it needs to consider its resource allocation and direct resources towards the area where the targets are not being met.

We have heard that NHS Greater Glasgow and Clyde has done that with surgical procedures at the Western, the Southern and elsewhere. Mr Calderwood, can you give an assurance that, in future, where targets are not met, you will continue to redeploy resources properly to ensure that they are met?

I have another question for you, on patient choice. For some time, everyone has been aware that the Western infirmary will close. How will you ensure that patients know what choices will be open to them once it closes and that we have an informed patient group who can make that choice?

On the issue of resources, the board monitors its delivery of all the Government targets and seeks to deploy the available resources to meet individual patient needs and to deliver Government objectives, as set by Parliament.

Over recent years, I think that we have demonstrated a balance in identifying and moving resources to meet those targets. I do not think that anyone could say that, in response to future Government decisions on resources and potential changes in demand, everything will flow automatically. There is a requirement on the board to look across the whole healthcare spectrum. Elective and emergency targets within the acute sector are indeed important, but equally the board must look at a whole raft of other community primary care and health inequalities issues. We constantly need to balance that spectrum against the available resource.

10:45

At the moment, up to and including the current financial year 2012-13, I believe that NHS Greater Glasgow and Clyde has identified demand and need and has sought to meet those by increasing the resources available. Equally, I have to say that we have not always achieved every target—there are some targets that we have struggled to deliver against more recently.

The situation regarding the future changes in acute services within NHS Greater Glasgow and Clyde was well trailed a number of years ago, but we will embark on a communications exercise in the second half of 2014 in relation to the impending changes that will come in 2015. To take your specific example, the in-patient base of the orthopaedic service at the Western infirmary will move to the new Southern general. The out-patient clinics at the Western infirmary, including the diagnostics and the day surgery that are currently provided there, will go to Gartnavel general. Therefore, if you live in the west or north-west of the city and you currently choose to go to the Western infirmary orthopaedic department, you will continue to access that service as an out-patient or as a day case in the west of the city at Gartnavel general. Tonight, if you were under the care of that team, you would have your in-patient surgery at Gartnavel general. Under the new model, in all probability you will have your in-patient surgery at the Southern general.

Therefore, when the changes are complete, we will not have eight discrete orthopaedic teams but a lesser number. Three teams—namely, those at the Victoria infirmary, the Southern general and the Western infirmary—will move to become one team that will support the south and north-west of the city.

Could I come and see how the TrakCare system works in Glasgow to see whether it is as sensitive as you say that it will be once it rolls out across the city?

We would be delighted to facilitate that visit.

I say to my good friend Bob Doris that I envy his position in having all that choice, including the list of consultants that Mr Calderwood described earlier. As far as I am aware, such choice is not open to people in Shetland, Orkney, the Highlands or most other parts of Scotland. I can only envy people in Glasgow. However, that is not what I want to ask about.

Professor Mackenzie, can you help me with some practical details that I want to get to the bottom of? Am I right in thinking that you and your colleagues, as chief executives, meet regularly together?

Yes, we do.

How often is that?

Monthly.

Who do you meet with, in addition to the chief executives?

Generally, in the morning we have a meeting among ourselves and in the afternoon we have a meeting with the NHS chief executive, Derek Feeley, and his team.

Are there any standing items on that agenda as a matter of course?

We generally discuss an overview of performance and the key issues that you would expect us to talk about. We may also discuss things that are particularly topical at the time and, obviously, our forward plans.

Was the issue that we have now been discussing for a couple of hours this morning a topical issue at the time when, post the NHS Lothian example, it first became public knowledge?

Waiting times would always be something that we look at together in terms of general performance. Yes, there would usually be a very high-level discussion highlighting any particular issues that we needed to focus on. That would be the general tone of it.

As a neutral observer—that is why I am asking you these questions—in the context of the post-NHS Lothian scenario, do you recall the issue coming up? Was the issue of unavailability codes specifically mentioned in the chief executives’ meetings?

I could not say that it was mentioned particularly. I think that the issues in NHS Lothian became clear over quite a prolonged period, from the first point at which NHS Lothian was mentioned as a matter of interest until it became clear what the issues were. I do not particularly remember unavailability being a major point of discussion at that point.

I do not want to ask you unfair questions about your recollection of meetings some time back. When the NHS Lothian issue erupted—because of a whistleblower, as we heard earlier, rather than because of anything that came from within the system—did that then become an issue that was of note and of importance to all chief executives across the country?

Obviously, everybody was aware of the issue as it became known, and everybody would have thought about their system and how things worked. Robert Calderwood’s point is important. If you do not mind me making a point, I say that in overall terms—that is often what we looked at as a group—the performance, disregarding the stops and wherever anybody was on the list, was relatively good, so we would not have been worried by the global perspective on performance.

I chair the chief executives group and I co-chair the afternoon meeting with Derek Feeley. Such issues are discussed regularly. It would have been irresponsible of the chief executives and the Scottish Government not to discuss the implications of the Lothian situation. Of course we had a discussion, but it was in the context of the NHS’s overall performance on waiting times and the Government’s appropriate decision to invite auditors to look at the issue, because the public require assurance. If we had not given the subject proper attention, that would have been irresponsible. That proper attention has been given over the past number of months.

Waiting times are not exclusively about unavailability. That has become a focal point because of the issues that were raised in the Audit Scotland report. In the past 10 years, the Scottish Government has made extraordinary effort with the health service and front-line staff. As Alan Cook has demonstrated, that investment has involved a 19 per cent growth in activity and a 20 per cent increase in consulting staffing. A huge effort to do that—

We are here to discuss a specific report, but you are giving me an overview.

I understand that, but it is important to set the discussion in the context of the NHS’s overall performance. In relation to the report, I have already said that there is no magic number for appropriate unavailability. The auditors have pointed out system weaknesses, and every board has an action plan, which is to be completed by the end of March. We have accepted the findings, and every board is implementing measures to deal with the system weaknesses, so that we can improve how unavailability is dealt with, which we are discussing this morning.

Since you have taken over and answered lots of questions that I did not ask, I ask whether the chief executives decided to initiate all that work themselves or whether Mr Feeley asked you to do so.

There was a coming together of the executive team along with Government officials to decide the priorities.

Were you asked?

Yes—of course we were.

You were asked. When were you asked?

I do not recall the specific date.

Was it in the post-Lothian period?

We were not asked in the context of post-Lothian. When something erupts in the health service—

You were not asked in that context—

Let me finish. If something comes to light that is a concern to the Government, the vehicle for discussing that is the monthly joint meeting. That is the context in which the Lothian issue was raised—in my view, appropriately.

So you were not asked—

Asked what?

You were not asked by Mr Feeley to take up the issue of unavailability codes when it burst into the public domain because of the Lothian scandal.

We discussed the consequences of the Lothian report. As the accountable officer for the NHS in Scotland, Mr Feeley—rightly—sought assurances that we were giving the issue due attention. We discussed the possibility of an internal audit and an Audit Scotland process. That is the normal conduct of business that I recognise.

In response to the convener, you said that you commissioned a report. Was that in the context of what you have described?

No—that was my local report.

Who commissioned that?

Me.

Not your board?

No—it was on the board’s behalf.

What was your board doing?

We were reporting to our board on a monthly basis, in the normal way. At a board meeting, I advised the board that I had commissioned work. That was not to do with any national request; that was me as the accountable officer in the local system seeking to satisfy myself, on my board’s behalf, that we had the issues dealt with in Tayside. The decision was entirely local. I informed the board that I was taking forward the decision, and we subsequently reported.

Your board did not initiate any of that. In the “post-Lothian” period that Dr Cook described—that was his term, not ours—did your board not say, “This is something we need to be aware of, Mr Marr. Is this going on here? Should we be assured that this isn’t happening here?”

In fact, we undertook a very comprehensive review of our data, including those from the organisational development exercise that I described. That was comprehensively reported to our board.

As I said—and I repeat—that report did not flag up to us that there was a serious problem with social unavailability. If the auditors subsequently point out systems weaknesses to us, we have to accept that. I have never in any way criticised or sought not to accept the findings of the internal audit. We did a huge amount of work after the Lothian report was published to assure our board that we did not have a difficulty with social unavailability.

You have made the point about the wider context and so on, but do you feel that the overall target has become more important than anything else?

Sorry—is that question for me?

No—I am asking Mr Marr. He has the floor at the moment, but I will let you in as well if you want, Mr Calderwood.

No, that is all right—I am quite happy.

I would be happy to ask you, too.

No, I do not think that it has become the preoccupation of the health service. It is one of many targets that I believe are justified. The general public deserve to know that they can gain access, given that a number of years ago, we had waiting times that we were all not particularly proud of.

There is a debate for the future about what the irreducible minimum waiting time is when we have 500,000 transactions.

That is a fair point.

I think that 12 weeks is a pretty good place to be, but transacting 500,000 patients will take a number of weeks.

I accept that—that is a very fair point. It is the consequence of the target that concerns many of us, however—and that probably goes much wider than those of us in the room. Did your board, and did you, as a chief executive, address the consequence of that target and what it meant for people in your area?

Yes, absolutely. Our view is that the target is achievable with a great deal of hard work and effort. Balanced against that—and I can only speak from my own context—we put as much emphasis on other aspects of quality in our 2020 vision for the health service of Scotland. I would not want to distort our effort on accident and emergency waiting times, unscheduled care and all the other things that we have to do in order to fulfil our local delivery plan for Government. I do not believe that the target is a distortion of that effort; it is part and parcel of the challenge that we face on a daily basis.

You do not feel that your staff were put under unnecessary pressure because of the target with regard to how they had to perform.

They feel pressure—I refer back to the local report that I commissioned, in which the staff acknowledged that they felt pressure. They also felt supported, which is a very different conclusion from their feeling pressured and consequently feeling bullied. The staff said that they felt pressured but also supported, and it is our responsibility to support our staff as much as we can.

So you do not feel that the target is an impediment to what you are trying to achieve—and, more important, you do not feel that the target puts any undue pressure on staff.

I would take out the word undue: it creates pressure, but our performance in Tayside shows that we are meeting all the other targets that we have to meet while delivering the 12-week TTG.

I also have questions about ISD. I wish to clarify this—I tried to write down the words that were used earlier. Susan Burney said that the Government accepted that there were “plausible reasons”—I think that that was the phrase that she used. Do you meet Government representatives on a monthly basis to discuss a range of performance indicators?

We meet regularly to discuss data and data development. Much of the work that we do on data development is in partnership with and in support of boards and the Scottish Government. ISD people will typically be involved in many meetings and groups, examining data development, including that relating to waiting times.

In terms of data quality, which I discussed before, we will have spoken to boards about anything that we see as unusual. The boards will come back to us and either acknowledge that there is something wrong—and tell us that they will change that or that there is something that we need to know and therefore include in our publication—or they will confirm that the data are correct. If they say that the data are correct, we do not take the matter further. Those conversations will be among many conversations about a range of data quality questions.

11:00

I am struggling to understand what you do that flags up problems—please do not think that I mean you, personally; I mean ISD. What does ISD do that helps the system?

We bring the data in and validate them. We look across the data from quarter to quarter, ahead of publication, looking for anything that looks unusual, which might lead us to go back and check that the data are okay. We have a range of processes for that. The result is a list of questions that we want to ask each board, just to check that the data are correct—

You are talking about an internal challenge function, rather than—

Yes, we look for unusual things in the data, which might mean that we want to reassure ourselves that there has been no mistake in the submission, before we publish.

Do you accept that none of us would have heard anything about the issue were it not for a whistleblower? It was not due to any part of the NHS that the situation—for want of a more pejorative word—emerged.

I come back to the fact that the data are published, so anything unusual is also published.

Some of the data were not published. That is the point.

May I take us back to roles? Roles are fundamental in this context. The Scottish Government is responsible for performance management on a national level; ISD’s responsibility is to ensure that the statistics that we publish have a quality to them; and boards have accountability and responsibility to ensure that the figures that they submit to us are correct and that patient care is delivered. Those are the clear roles that we are transacting.

Part of what you get from us, in terms of value add, is some comfort that the timing of our releases is not politically driven and is independent. The numbers that you are looking at are what we believe the numbers actually are, as opposed to anything that people would like you to believe. We do quality control to a statistically acceptable level, and I think that you can take comfort from the external work that was done in 2010 by the UK Statistics Authority.

What we are not designed to be is a proxy for audit. We cannot compensate for what board systems themselves should be capable of regulating. Any board will have a board of governance—members asked about that—which has a role. Sitting underneath that is a clinical governance committee, which has a role in seeing what goes on with waiting times. It is important to understand that in relation to scrutiny, challenge of probity and so on, a board has far more information than ISD has on what is happening, and will remain the best place in which to gauge the probity of numbers. We can provide you with an external counterpoint, but not much more than that.

I take your point about roles. However, Audit Scotland said on page 34 of its report:

“ISD Scotland was not clear about what issues to escalate to the Scottish Government.”

Do you recognise that?

No, I do not accept that.

Do you not understand why Audit Scotland came to that conclusion?

I do not understand that, and I do not understand why Audit Scotland did not discuss making the point with us before it made it.

Do you think that there is any basis for the comment?

No, I do not. The report contains good suggestions for improvement, and we are keen to improve—

But not that suggestion.

We have always been quite clear about our role. It is really important for us to be clear, because if we were less clear—if we were too close to Government or if we were confused about our role in supporting boards—that would be unfortunate.

Okay. Did you at any stage raise with the Government the statistical imbalances that you mentioned were coming through in relation to social unavailability codes?

It comes down to what you mean by “raise”. If the question is whether the Government was aware of the statistics, in the way that anyone else who was looking at our website would have been aware of them, the answer is yes—

No, I was not asking that. Did you have formal meetings with the Government, at which you could say, “Look, there’s something going on here. We don’t exactly know what it is, but we think that the Government should look into it”?

No.

You did not have such a meeting at any time during the whole episode.

No.

There has been a great deal of discussion about how the figures are extracted and the limitations on them. On page 7 of its report, Audit Scotland said:

“The systems used to manage waiting lists have inadequate controls and audit trails, and the information recorded in patient records is limited.”

The appendix, on audit methodology, shows that a huge number of detailed questions were asked to which the systems could not produce answers. The multiplicity of systems, which I know is being worked on, especially in Glasgow, caused problems. What reassurance is there that those systems will be brought together, that we will have uniformity of reporting and that we will not have the different boards using different criteria in the future?

NHS Greater Glasgow and Clyde is well on its way with a project plan to reach a situation in June 2013 in which we will have only one information technology system—TrakCare—which, when completely rolled out in Glasgow, will represent the principal IT system in NHS Scotland. The system will be operational in NHS Lothian, NHS Lanarkshire, NHS Ayrshire and Arran, NHS Grampian, NHS Borders and NHS Greater Glasgow and Clyde, which together represent about 70 per cent of NHS activity.

The consortia are working with TrakCare to ensure that we develop an audit program in the suite of software modules that will give us complete transparency and auditability of every individual patient’s journey. We are very confident that, from this summer, that system will be in place and there will be a degree of consistency across a significant proportion of the NHS in Scotland.

I accept that there will be a uniform system, but will every board have the same criteria in using that system? As we all know, the approach can be different on a number of things.

The new ways approach and the treatment time guarantee have a series of protocols that must be applied uniformly across NHS Scotland. Each board published its own access policy, which determines how it will address the needs of its patients—in other words, where they will obtain that activity from. Each board has slightly different access criteria, so what represents a reasonable and fair offer to an individual patient will be determined by reference to the access policy of their own resident board, and those are clearly quite different across Scotland.

Who will have the overarching responsibility for ensuring that there is uniformity?

At the moment, it remains the case that it will be for each board to demonstrate that it is applying all the rules and regulations and that it is consistent with its access policy in its own certification of its data. Clearly, if there are concerns or there are significant variations, that will usually be picked up in the normal management interactions between the boards and the Scottish Government health department.

So there is a process in which the different boards confirm that they are using the same criteria.

There is increasing regularity as we move to the treatment time guarantee. A number of aspects of the treatment time guarantee are enshrined in parliamentary legislation, in relation to how it must be applied and how citizens of Scotland should get equitable access to it.

I realise that there are a lot of exchanges between the different boards and that uniform systems are coming in, but I still struggle to see who will ensure that the criteria are the same in each board, so that we are comparing like with like.

The application, reasonableness and audit trail of clock stops will be consistent across Scotland, because we now have to enter into written communication with the patient to confirm that their clock—their treatment time guarantee—has been altered by discussion and agreement with them. As I understand, that transparency will be there across the whole of the NHS in Scotland. What individual patients deem to be an appropriate choice and why they seek to exercise that choice will vary across the system because, as Tavish Scott pointed out, certain resident populations may not have a lot of choice and others may have significant choice. Therefore, there will be fluctuations based on how populations act on the choice that they have.

There will not be a single system in Scotland. For example, our system will not be TrakCare; we will continue with Topas. In our system, Topas is being rewritten to meet the criteria of the new rules. The criteria of the rules in Robert Calderwood’s board will be the same as the criteria of the rules that we must write to in order to meet the requirements of the new guidance.

Does that guarantee that someone who is dealing with waiting times in Glasgow is doing the same thing on the same day as someone in Tayside or Shetland? That is a different question, but the systems will have been written to apply the rules. The practice is an issue for training, development, audit and checking in the future to make sure that the rules are applied, and it is the responsibility of each local health board to give that guarantee.

All our different systems refer back to the guidance that came out last year in chief executive letter 32, which was on the treatment time guarantee, and CEL 33, which was about the national access policy. Whatever system we have, whether it be TrakCare in Glasgow or Topas with us, we make sure that we refer back to the national guidance so that we can answer questions. As has been said, the Topas system has been reconfigured and linked back to the new ways and treatment time guarantee rules so that it answers the core set of questions. There will be a core data set that all the different systems will answer.

Does ISD have a role in ensuring that criteria are maintained at the same level throughout the NHS?

We do in the sense that Susan Burney talked about when she mentioned looking for variation. As we move away from one social unavailability code, we move to the more patient-driven range of different codes that has been put in place. The improvements that are now being made are twofold. First, the Government is being very clear about the definitions of the different codes and what should be entered, so definition is much better than it has been in the past. Secondly, we will look at those different codes to see whether anyone is not using any of them, or whether there are outliers. That would mean that we would be in a position to take a view on that, as would you.

In a practical sense from our board’s perspective, eight different reasons are accepted for patient-advised unavailability. Topas has been configured to use a drop-down box that shows the choices. It can show that the patient advised that they are on holiday, that they have a personal, work or carer commitment, or that they will be on jury duty, and the other choices are to do with patient choice, such as the patient wishing to see a particular consultant or to go to a particular location. In Topas, each of those accepted patient-advised reasons for unavailability is in a drop-down box that is used after discussion with the patient to make sure that we are capturing the information that is put out in the national access policy.

I talked earlier about Audit Scotland and its methodology, and the large number of key questions that the systems could not answer at that time. As was said when evidence was given previously, the result was that Audit Scotland looked at 3 million patient transactions, did a trend analysis, and eventually extracted 310 patient records, out of which it found 20 inappropriate uses of the unavailability code. Is that a reasonable approach, and is that a reasonable figure? I am asking for an opinion.

Audit Scotland’s methodology in the construction of the audit and the ability of the NHS systems to provide data that would assist or clarify the point have proved challenging. Glasgow has had its IT systems for 16 years, and we knew what they could and could not do, so when Audit Scotland came to us with its methodology, we knew that we could not answer what it asked, and we had always known that. Short of going at it in a different way, I am not sure how we at NHS Greater Glasgow and Clyde could have shone any more light on the issue. As Mr Marr and others have said, the attempt to make sure that there was no manipulation of the figures and to reassure the people of Scotland was highly desirable, but my comment is about how we could exhibit information that we were not collecting at that point.

Mr Doris asked about the process of coding a patient as socially unavailable. That has been helpful. Since the situation at NHS Lothian, there has been a much greater focus on coding. Have there been any changes to that process since the situation at NHS Lothian came to light? Have there been any changes in the seniority of staff who clear that coding, or has the only change been the one that has already been described, which is that staff have been better trained in use of the codes?

11:15

I made the point earlier that, at the time that the NHS Lothian investigation and, ultimately, the PricewaterhouseCoopers report were made public, NHS Greater Glasgow and Clyde’s social unavailability as a percentage of waiting lists had reduced to 17.3 per cent. As I said earlier, when I looked back retrospectively, I saw that it had been significantly higher. As at the end of January 2013, our social unavailability is 17.6 per cent. It is difficult to say what is an acceptable level of social unavailability. However, we are seeing a fairly consistent position now, in which about 17 to 18 per cent of people have reasons for seeking to defer their access.

So there is no change in the process.

I do not think that we have made any changes locally.

With the advent of the new TTG, the process has completely changed in the sense that there is a requirement to communicate with patients and have patient-driven unavailability as opposed to social unavailability. The fact that we have to confirm in writing is also a key part of this. We now need to write a letter to the patient saying, “This is what we’ve agreed and, if you don’t agree, let us know.” That is quite different from how it was before. Before, there was a phone call and we had a discussion. Now, we confirm in writing. If the patient receives a letter that says something that they are not clear about or do not agree with, they have the opportunity to challenge that immediately. From that perspective, the process is much clearer.

Going back to my previous point, I think that the new ways rules are complex. When we phoned patients and talked about things that we all use as normal currency, some patients might not have been clear what we were saying, whereas now they have a letter saying, “This is what we’ve agreed with you and, if you’re unclear, please let us know.” The TTG circular and the guidance now require that to happen. I assume that that is happening everywhere. It is certainly happening in NHS Greater Glasgow and Clyde. That is quite different.

It is happening with us as well. In addition to the letter that goes out to reflect back on the conversation with the patient and gives the reason for the patient-advised unavailability, we send out a letter that clearly states the implications of that. If a patient has said that they are unavailable for two weeks for jury duty, for example, we will recalculate their wait statement for them so that they know their new guarantee date. It is not just acknowledging the situation; it is putting it into context for the patient, with a new date.

On your point about training, I have been over some of the training in which we have been involved. In addition to training, there has been a lot of work on opportunities for waiting times staff to discuss issues. We have set up a generic email drop-box on NHS Tayside waiting times. If staff have any uncertainty about what is going on, they can email and it will be answered. We collect all that information and put it on to a frequently asked questions page on our staff intranet page so that everybody is aware of the correct process to follow.

If there is uncertainty from our perspective about what the rules are, we escalate issues to the Scottish Government access support team, which is very good about coming back with its interpretation. We have put in all the educational resources and we have a much clearer environment in which all the issues can be discussed in open forum, and we can get the correct answers and the correct interpretation. Staff are using that. There have been more than 50 emails asking for clarification of issues.

A far greater number of telephone calls are coming into senior members of the waiting times teams, asking them for clarification of issues. The environment is very open and transparent. We are seeing that from a patient perspective at the start.

We developed the letter that we send out to patients in Tayside at our public partners event in discussion with our public partners and patient representatives. We looked at the wording to ensure that it was clear and that the wait statement was clear. They fed back to us from a public and patient perspective what they wished to see changed in our drafts and we changed them accordingly, so that is what is written into our access policy as the information that we give out. In addition, the telephone conversations that take place are all structured now in a way that is set out in our access policy so that we are clear in what we say.

It has been said that the audit process has picked up weaknesses in the system, but panel members may or may not be aware that this is not the first such report by Audit Scotland. Previously, Audit Scotland recommended that a specific code be set up for patients who want to choose a local hospital or consultant, which applies more to Glasgow than anywhere else. Were panel members made aware of that previous recommendation? If not, would you have welcomed and implemented it had you been made aware of it?

We were aware of the Audit Scotland report that was published in March 2010 and the action plan that was attached to it, and we were aware that one of the recommendations was about amplifying the codes that would be available. The debate went on for a period of time and agreement was eventually reached across NHS Scotland on changes to the codes, which I believe were issued to the service in late 2011. I hope that I am correct about the dates when the new codes were introduced and there were new definitions for unavailability.

The ISD guidance states:

“While the patient has rights of choice, NHS Scotland has a responsibility to present the best possible options ... to ensure that they are seen within waiting time guarantees. There are 2 options for dealing with these situations”—

that is, when the patient wants to be seen by a specific clinician. The guidance goes on:

“One option is to make a reasonable offer of appointment, whether for the specified clinician or not and, if declined the guidance relating to declining an reasonable offer would apply”—

that is, the patient would be sent back to the GP or whatever. The guidance continues:

“The second option would be to give a period of Social Unavailability from the date of the request until an appointment on admission with a specific Clinician becomes available.”

The ISD guidance also states:

“patients will be considered as transferable within the clinical network in their health board area”.

Some of that guidance is perhaps less specific than Mark Griffin is describing, but it is there.

Yes. The Audit Scotland report states that that updated guidance was published in August 2012, which was some time after the 2010 Audit Scotland report. Perhaps ISD can comment on why there was such a delay in publishing the guidance, given that the recommendation on it was directed at ISD.

Yes. As somebody has already said, there was a great deal of discussion about the code by the Scottish Government, boards and ISD over some time. It was decided to incorporate the new code into the general new guidance on the treatment time guarantee, which was going to be a lot more specific about patient availability codes. It was incorporated in that work and therefore came out around the time that you have stated.

You said that there was a level of discussion. There was also a follow-up letter from Audit Scotland to ask when the guidance would be introduced. I want to drill down to why it took over two years for the guidance to be adopted. What were the discussions? Was there resistance?

Not that I am aware of.

If there was no resistance, why did it take longer than two years?

I think that the reasons are complex. That is more a question for the Scottish Government than for ISD. People are saying that it is ISD guidance, but it is actually Scottish Government guidance.

The 2010 report contained a desire for there to be more clarity around patient choice. At the time, the health service generally accepted that. I think that Susan Burney is making the point that the fact that the treatment time guarantee is enshrined in legislation has slowed down the passage to a point at which the new guidance would be implemented.

The Audit Scotland report in 2010 was reasonably positive about a lot of the achievements that had been made. There was not a rush to immediately go out and fix something. There was a plan to continue on a path and continue to make improvements. That is where we are today.

Although I do not share the deputy convener’s suspicion that every board is a hotbed of manipulation and bullying, a couple of issues have come up recently—

To be fair, the deputy convener made no such accusation.

I quoted from the report.

I suspect that what we should do is read the Official Report later.

Does the panel agree with the report that the long-established NHS whistleblowing policy did not operate effectively in helping to address concerns at NHS Lothian and NHS Tayside?

I cannot comment on Lothian, but I will talk about the process that unfolded in Tayside.

The internal audit report contains a great deal about transparency and co-operation. There is much in the report that commends Tayside’s internal audit report and acknowledges that we had a system that was geared towards transparency and encouraged staff to engage when there were problems. The internal auditor interrupted his audit exercise to come to see me to say that he was surprised to have come across one small area of our waiting times practice in which inappropriate codes appeared to be being used. That was after we had done the major exercise—before the audit was announced—in relation to our staff’s views on whether there was any bullying or harassment. That report unequivocally said that staff were proud to be working in Tayside and were busy, but recognised that the pressure was considerable. There were no red flags around the issue of bullying and harassment. When the auditor advised me of the potential for that, I took immediate action to investigate the matter. The auditor continued with his audit. I used our conduct policy to investigate the matter. Regrettably, some of that, which involved individuals, came into the public domain—I do not want to go into that. That investigation concluded that, in the context of the conduct of those managers, there was no case to answer, and both managers have been back at work for two or three months now.

I offer contrary evidence to the suggestion that bullying and harassment were a feature of the system in NHS Tayside. As Audit Scotland says in its report, the codes were applied inappropriately in a small number of cases. We have publicly apologised for that, and have reassured ourselves, by reviewing all 63 cases, that there was no detriment to those patients’ pathways of care.

We have been thorough and forensic in tackling this problem. We refute any suggestion that bullying or harassment were a feature of what happened in Tayside.

Given the actions that were taken, did you think that it was appropriate to take any additional actions to support staff in relation to the areas of concern?

Alan Cook has already set out the fact that, apart from training, we created the helpline for staff who work on a day-to-day basis and we put in place an escalation policy so that more senior members of staff can meet staff to help them to interpret genuinely complex rules in the course of what is often a pressurised working day. We have put in place mechanisms of support since we received the internal audit report and have improved the system by implementing that report’s 22 recommendations.

Have you done much to publicise that to your staff?

Yes. We have put a great deal of effort into engaging with our staff and publicising it. The questions are on our website, which staff can access. We are being open and transparent about the fallout from our own internal audit inquiry and our conduct investigation.

11:30

Will the Scottish Government’s new whistleblowing helpline be of any assistance?

That will have to be tested. Our attitude—I am sure that my colleagues share this view—is that the whistleblowing helpline would be most successful if it was never used because people had the confidence to expose issues through mechanisms in their own organisations. Clearly, there is a lack of confidence in that at present, but we have a responsibility to work towards that end.

Do the other witnesses want to comment?

I have already commented on that. I do not recognise the issue of bullying and harassment in relation to waiting lists. In NHS Greater Glasgow and Clyde, almost 5,000 people interact with waiting-list management daily, of which 2,000 are consultants. It has never been my experience that consultants in NHS Greater Glasgow and Clyde can be bullied or harassed into dealing with the management of their patients or waiting lists. The articles in the local papers are a regular testament to that.

When we received our internal audit report, the concluding comment was:

“Internal Audit did not identify any instances of inappropriate behaviour or pressure on staff to hide breaches or manipulate data and our review found no critical matters of concern.”

Clearly, that was a comfortable message to get back, but we cannot be complacent—we must always work hard to ensure that staff who are working in a pressured situation are supported if they have concerns. That is done through a number of routine things similar to those mentioned by my colleagues. I have personally met the staff who make the bookings on a number of occasions, and I have been assured that they are receiving the right support. They are keen to do a good job and keep the patient at the centre of what they are doing.

The report says that there was no evidence of wide-scale manipulation or bullying. I take it from the panel’s responses that that is because there was not any.

Yes, that is my view, but I can talk only about my own system.

I echo that.

Yes.

I am glad that Mr Crichton mentioned the 2010 Audit Scotland report. That came slap bang in the middle of the rise and fall period of social unavailability that we have heard about. The report, which came to us at a time when social unavailability peaked, was very positive. Our guests round the table have explained that that peak was as a result of the introduction of new systems and so on.

I want to remind everyone that the scope of the Auditor General’s examination was whether the Lothian situation was widespread across Scotland. As members have said, after looking at 3 million transactions, she did not report any evidence that that was the case, and that appears to have been borne out in the contributions that I have heard this morning.

First, I have a question for ISD. A message throughout the Auditor General’s report is that the data made it difficult to interpret what was happening. I have difficulty reconciling such a comment with the comments made by ISD about the data that it presents. Will Ian Crichton please explain why the Auditor General would say that the data, across the board, are not particularly clear when on a number of occasions this morning he has said that, as far as he is concerned, the data were good and met quality assurance requirements?

ISD and Audit Scotland come from different places in, if you like, assuring the data. As Susan Burney mentioned, we are looking at trends. Every month and quarter, we go through involved processes to go back through data because, for example, there are usually cut-off issues for different boards on timings.

We look at what the different systems are generating, we do a reasonableness assessment of what is coming out, we make various technical adjustments to the information and we ensure that we publish a list of the different areas that we are not 100 per cent comfortable with in terms of dotting all the i’s and crossing all the t’s. That is where we have come from. The Audit Scotland auditors went down into the bowels of the system and did their transaction testing. However, I understand that they then struggled to link what the transaction was telling them to what actually happened to the patient, because it was not recorded. My source for that is the same as yours—the report.

We are comfortable with our statistics. We have talked about the rise in social unavailability, but that was fairly common across the piece. We asked questions when the occasional board looked odd. We have spent a lot of time talking about social unavailability, but that is only a small percentage of everything that the system measures. To me, the most important part of what we have been providing you with assurance of—and the Audit Scotland report bears it out—is that there has been an improvement in waiting times, over time, and it appears to be what we have told you it is and what we have published to the Scottish people. For each individual case, I can do no more than Audit Scotland in telling you what the accuracy level is of its being transcribed properly and moved on up the system. However, I can tell you that, from a reasonableness perspective, at a high level, we think that the statistics are good.

A good question was asked earlier about our view of the percentage of variance—the error rate—that the auditors picked up as they compiled their report. It is hard for me to comment on that, not having seen what they looked at. However, human beings record these things, which are very complicated. An error ratio at the level that you were talking about is probably to be expected.

Thank you for that. Is it not your role to close that gap? We could be sitting here in two years’ time having another look at the issue. God forbid that we would be in the same position of having to make all sorts of inferences and assumptions about what the data are telling us. Is it not your role to intervene in some way to encourage different or better practice in capturing data manually and embedding them in the system so that they can be properly analysed and conclusions can be drawn?

That is a fair point. There is no question but that we can do better there. However, I would not want to sell short what we currently do. Towards the end of the reporting periods, an awful lot of effort goes into liaising with boards and agreeing what the numbers should look like. We help boards to clean up their information as they get to publishing points.

When we talk at this level, the amount of effort, pressure and administration that is required of people on the front line in order to record the information does not always come through. The starting point for them is the patient, not reporting to us for the purposes of information on waiting times. It is sometimes a challenge for me to remind my guys that we are a by-product, not the point, of what the system is there to do. We can work on doing more, but it would be disingenuous if I told you that we would be able to get right down inside the system alongside the person recording the information to ensure that it was done properly. We will help the boards to look at variations and to understand any outlying statistics that we see. We will continue to support the boards as best we can—that is our role.

Committee members have spoken not just at this meeting but at previous meetings about the recording of certain information about the patient that was very helpful in allowing Audit Scotland to come to a view in relation to NHS Forth Valley. Mr Calderwood said that some information was not recorded that might, ultimately, have helped the committee and others to come to more accurate and informed conclusions about what is happening. I hope that the embedding of that kind of information about real people and real reasons for things happening becomes part of the data-capture methodology that is used across the boards so that, when we have another go at this in a year or two, we will be much clearer about what the data are telling us.

Can I come back to you on that?

If you are brief. There is another question about the sort of information that we are given.

I will keep my comments brief. Dr Cook mentioned the move from a single box for social unavailability to a range of availability codes. That is your assurance that, when you come to ask the question in a year or two, you will be able to see a range of reasons for unavailability and will have a much better understanding than we can give you today.

I am really pleased about that.

I turn to a question for the boards. In the absence of clarity when data cannot be completely relied on, they become open to interpretation and, sadly in my view, that is what has being going on here. In the past period, certain members and politicians have been saying things about the data that are clearly unfounded. I would like to ask the board representatives a direct and specific question about that. In your view, were hidden waiting lists widespread across health boards, and particularly in your boards? That has been said—that accusation has been levelled against you. I would like to give the three boards an opportunity to refute that and explain their positions.

I would refute the accusation that there were hidden waiting lists in NHS Greater Glasgow and Clyde. I have made the point a number of times this morning that, if we look at the total waits of all patients in the total journey irrespective of clock stops, 93 per cent of people were treated within the targets. That is an auditable, externally validated figure.

I accept and have said more than once this morning that we were not recording the data in such a way that they could retrospectively be audited, and therefore people could infer that it could be open to interpretation whether we applied the code appropriately. We have offered views on how we used the code, but we cannot demonstrate and validate that our actions were right in 100 per cent of cases. However, if we look at the totality of the patient journey, I believe that, in 2011, the NHS in Scotland was a success. We were moving forward and improving waiting times.

Mr Marr?

Can I come back in, convener?

Sorry. I thought that you wanted to hear from each of the witnesses.

If Glasgow had been operating hidden waiting lists, as has been suggested in the accusation that has been levelled against you, could you possibly have delivered that 93 per cent performance?

No.

Thank you.

Mr Marr?

I am on record as saying that, between the two audits, 50,000 transactions were examined. Audit Scotland examined 30,000 and raised no issue with us. The other 20,000 were examined by our internal audit, which brought it down to 63 transactions that proved to be inappropriate. That represents 0.2 per cent of all the transactions that were examined. That indicates to me that the central causes of the challenges that we faced were weaknesses in the system and the complexity of supporting our staff and training them appropriately.

I am confident that our system has no hidden waiting lists and has never attempted to have such a thing.

Professor Mackenzie?

I agree with that. There is no hidden waiting list. All of the waiting list is visible.

What is your performance compared with Glasgow?

It is similar percentage-wise.

It is up in the 90s.

Yes. It might be slightly lower as we had some other issues at the time, but in overall terms it is pretty good, and the Scottish performance is good.

What was yours, Mr Marr?

Ninety.

Thank you, convener.

I have a quick and simple question as my colleagues have covered most of the technical stuff in their questions. Is it fair to say that the main problem that we have had is an old IT system that did not collect the facts and could not be interrogated? Perhaps some people are making a bit more out of the problems than there is in them.

My contention is that the absence of routine collection of data on the IT system made it extremely challenging for Audit Scotland to ascertain whether everything was appropriately applied. The absence, certainly in the Greater Glasgow and Clyde context, of the retention of the waiting list and the notes that the various parties made on why they arrived at decisions to apply social unavailability codes has compromised our ability to explain unequivocally that they were appropriately applied.

The other point is that, as we have said, there now has to be written communication, so it will be possible to audit tomorrow what we say is fact tonight. The rules now require us to have that clear audit trail both on paper and ultimately in IT.

Over the next 12 months, the test will be whether we are able to demonstrate in an orderly fashion, as my colleague Robert Calderwood has said, that the reasons for unavailability are well documented and are within the rules. The figures must also be reliable, given the systems that we put in place. The figure—whether it is 5, 15 or 22 per cent—becomes arbitrary if we can demonstrate to the public that the rule making has been abided by and that we have coded people appropriately. We should be the subject of that public test over the coming years.

11:45

As I think one of the member’s colleagues mentioned earlier, the systems were generally quite old and were not meant to do what we now need them to do. Certainly, I know that we have two systems in place and that one issue that the audit picked up, which we were already aware of, was that the systems were not talking to each other. Actually, one reason why we kept quite good notes was that we almost needed a back-up for that. In any event, our view is that we need to move quickly to one system and then look at moving to one of the other national options. The systems have not been helpful, because the need has changed over the period of time.

The final questions will come from Jackie Baillie.

Convener, I thank the committee for allowing me to pose some questions.

Let me start with IT systems, which is not where I was going to start. In and of itself, IT is not a reason for the rise, or indeed the sharp decline, in social unavailability. NHS Tayside has said that there was misunderstanding and a need for education, NHS Greater Glasgow and Clyde has rightly raised the capacity issues that are noted in the Audit Scotland report and others have referred to the introduction of the new treatment time guarantee. Were those the reasons for the rise and subsequent fall in social unavailability, based on the evidence that you have already given us? I just need a yes or no.

Sorry, I cannot give you a yes or no, so I might not be very helpful. For us, one issue that has also come out in the discussion is that, in some cases, we were using unavailability codes to hold people rather than send them back to their GP. That was motivated by a fairly patient-centred requirement or motive. From my point of view, when we move into the new guidance, people will need to work absolutely by the rules so that everyone is clear where they stand. That is reinforced by the letter.

However, that is about a person-centred system rather than IT.

Yes, absolutely.

I echo that point. You are absolutely right that IT has nothing to do with the rise, but it has everything to do with our inability to answer the exam question retrospectively. That has been the challenge.

The drivers for the rise in social unavailability, as we have endeavoured to explain this morning and as Mr Feeley’s letter to the committee also tries to explain, need to be tracked back from 2008 to the peak in 2010. A point that we have all tried to make this morning is that the use of social unavailability codes peaked early in 2011 and almost every board in Scotland set about doing something about the issue. All of that was long before the apparent, and now proven, set of issues about manipulation in NHS Lothian came into the public domain. It is slightly disappointing that that nuance is not in the Audit Scotland report as opposed to the nuance about the lack of evidence for or against.

I will give a short answer with a tiny caveat. The answer is yes, as that is part of the patchwork of complexity. Interestingly, our review uncovered patients who should have been coded who were not coded, but our staff said that they had been working hard not to code patients. There are two sides to the issue.

Absolutely. However, I think that the Scottish Government was claiming the “Little Britain” defence of “Computer says no.” Clearly, that has not been the case.

Let me turn to Ian Crichton, whose evidence I found interesting in trying to understand the process. I think that he said words to the effect that the boards treat the patients and operate the systems, the chief executives sign off the figures, ISD Scotland does the numbers—forgive me for paraphrasing—and the Scottish Government’s role is to monitor performance. Does that encapsulate what you said?

Yes.

That is helpful. I am clear that ISD Scotland has discussions with health boards and with the Scottish Government. I am also clear that, in turn, the Scottish Government discusses waiting times with NHS chief executives, as we have heard in evidence today. That happened prior to the NHS Lothian issue as well as post the NHS Lothian issue.

However, I suppose that my credulity is slightly strained because, between the Scottish Government officials and the NHS chief executives, there must be upwards of £2 million in salaries sitting in the room, yet not one of you, who are clever people all—I know that Robert Calderwood will treasure that remark—picked this up. That just tests my credulity. Were you in the room, too?

When you talk about picking this up, what do you mean by “this”?

Sorry—the rise in social unavailability.

Given our role in the meetings with the boards, we would have brought up the boards that were outliers around social unavailability, and we did. We raised the issue with Lothian, I think that we had discussions with Forth Valley, and we had discussions with Grampian, because those three boards seemed a bit odd. Those discussions were in the vein of, “What is going on?”

In the case of Grampian, I think that it was implementing its new patient management system. In the case of Lothian, I think that it felt that it had systemic issues as well. With Forth Valley, the issues were a little bit different—for the reasons that the committee has been discussing this morning. Those are the discussions that we had. My people are extremely smart and if I was not in the room, they were and they are very good at what they do, so those conversations were had.

On conversations with the Government, our relationship with the Government is always interesting because we need to maintain a degree of objectivity. The meetings that we tend to attend, particularly as they relate to statistics, would be meetings that were called by the analytical services division of the Government. The Government performance people are there and our role is to answer any questions.

Susan Burney can correct me if my recollection is wrong, but I do not think that social unavailability factored into those discussions. Typically, discussions around published statistics on different subjects never last more than about 20 minutes. Some of the discussions on waiting times may have gone on for as long as an hour, but the discussions were around how the system was doing and how patients were getting looked after, not around the small social unavailability number that was creeping up.

But those figures were there—they were available; it is just that none of these really bright people managed to pick them up.

We saw them and we persuaded ourselves that there were good reasons for them. The flipside of the report would be that we were right because, with the exception of Lothian, it would appear that the social unavailability growth and reduction were perfectly legitimate.

Okay. Which is your sponsoring department? Do you report to the sponsoring department? Beyond the day-to-day stuff that goes on, who at senior management level do you communicate with?

We are always a little interesting. I am the chief executive of National Services Scotland, so ISD is one of a range of services. If I talk about NSS sponsorship arrangements, I communicate up through John Matheson in the finance department; I have a direct line of accountability to the director of general health; and I have a chair as any other board chief executive would. If you go down a level to ISD, typically, the nominated sponsor for ISD is the Government analytical services division.

Okay. When you provide all these statistics, who do they go to? Does the director-general see them? Does—I forget all the different titles—the person who has responsibility for waiting times see all the statistics as a matter of course?

Under the pre-release access rules, we release through the analytical services division the publication for pre-release view. The analytical services division passes it on to the people who have an interest—policy colleagues and so on. Then there is a meeting, which can be of varying lengths, as Ian Crichton explained. ISD would sit with policy colleagues and others and talk through anything that they might want to ask about the publication of the figures.

Would that be at director level? I am trying to establish the lines of accountability.

It would probably be the head of a division—it is rarely at director level. Of course, it varies a great deal across the different data sets.

It is important to go back to what such meetings are for. Such meetings are primarily for the Government to prepare its lines on things ahead of the statistics being published. We make technical experts available to provide clarity; we are not there to negotiate a line or anything like that.

I understand that. Just to correct something that you said earlier, Mr Crichton, you said that these things are exposed sometimes by questions from politicians. Would it surprise you to know that we were raising the rise in social unavailability in 2009, long before Audit Scotland’s first report, never mind its second one?

Back in 2008, after the first year of the new ways system, a report was produced by the Scottish Government, health boards and ISD, which mentioned social unavailability. The question is about the amount of concern about that compared with the amount of concern about other areas. You might have raised it, but I am clear that from where we were, prior to the events in Lothian, it seemed that people were not concerned about social unavailability as a problem or about the kind of misrepresentation that the audit report talks about.

We certainly were, but there you go.

Finally—because I am conscious of time—I will ask about bullying. I was surprised by how emphatic the chief executives were in claiming that there is not a problem. I urge a degree of caution, given that even the British Medical Association in a submission for the parliamentary debate on the issue said:

“feedback from our members would suggest that this aggressive management style is not isolated to Lothian.”

The chief executives might claim that it is not in their boards but in somebody else’s. However—I put this to Robert Calderwood in particular, as carefully as I can—are you really saying that there is no bullying and no stress placed on any staff member in NHS Greater Glasgow and Clyde in relation to waiting times and social unavailability?

I need to break that question down into two. This morning’s debate has been about whether the social unavailability code has been applied inappropriately to allow the board to claim that it has achieved a target that it clearly has not achieved. I am adamant in my view that we have not done that, and that is Audit Scotland’s view. This morning, we have offered evidence that we are not doing so.

On the day-to-day issues between managers, including medical managers, and clinical staff generally about the pressure on the system to handle elective and emergency patients, there are many fraught conversations on a daily basis; indeed, during the past winter, those conversations between my medical managers and other medical staff have probably been on an hourly basis. Do all my staff believe that all the things that we are doing are in their interests or in the best interests of some aspects of the service? No, they do not. However, the issue for the board is to provide those services for the population and to strive to make our best endeavours to provide them within the resources that are available to us.

I make no apology for the fact that there has been a marked improvement in productivity in the NHS, which in part has been driven by using the resource differently. There has been significant service redesign. When I meet my area medical colleagues monthly, am I told about their disappointment about what they think of as another redesign or another change or target? Yes—that is a fairly regular Monday night with GPs. Fortunately, those meetings are on a Monday night and my meetings with secondary care consultants are on a Friday afternoon, so there is a gap between them. There is a series of pressures on the system. However, is it bullying? No, it is not. There is a debate between managers, including medical managers, and clinical staff about the need to meet patients’ needs. It is all driven by the patient.

I am talking about colleagues and NHS staff who are indeed hard working and who are below the level of the clinician, such as nurses and junior managers in the system. Are you telling me that there is no bullying there?

We need to be clear about what we mean by “bullying”. There is assertive debate throughout the system on a daily basis. Does the board recognise that we have gone from an occupancy percentage in acute medical receiving wards that was in the mid-80s up to one that is in the mid-90s and that that puts significant pressure on staff? Yes—we recognise that and we are working with our best endeavours to address that. Do people feel hard pressed now compared with, say, 10 years ago? Yes, they do. However, that is not bullying and harassment.

For the record, it is important to point out that I am emphatic that, in the context of the audit on waiting times, we have done a huge amount of work and have not been able to uncover any evidence of bullying or harassment. However, I cannot say emphatically that I am confident that that would never be the case in the whole of the complex healthcare system, because that would be complacent. This is not an issue for today but, at the beginning of the year, we launched a comprehensive culture programme in NHS Tayside. We take the issue seriously and we are determined that if we have any evidence of it, we will deal with it appropriately.

Thank you, Mr Marr. Is that okay, Jackie Baillie?

Thank you very much, convener.

I thank the panel for their forbearance in taking part in a ridiculously long evidence session. Your answers are much appreciated, although I know that we do not always give that appearance. The committee is running very late, but I am going to take a 10-minute interval, for the sake of humanity and the rest of us, and to allow witnesses to change over.

12:00 Meeting suspended.

12:09 On resuming—

Let us reconvene. We are running late, but I hope that members will bear with me. It is an important report and I want to make sure that everyone has the time to ask the questions that they want to ask. I give the committee’s apologies to our next set of witnesses, from the Scottish Government, who have been waiting some time. They are Derek Feeley, who is director of general health and social care and chief executive of the NHS in Scotland; John Connaghan, who is director for health workforce and performance; and Richard Copland, who is head of the access, workforce and performance directorate. Derek will make brief introductory remarks.

Thank you, convener. I was grateful for the opportunity to make a written submission, so I will confine my opening remarks to a few points only.

First, the Government accepts Audit Scotland’s recommendations and will act on them; improvements in systems and evidence are required and will be made.

Secondly, I want to put on record that deliberate manipulation of waiting times, such as was uncovered in NHS Lothian, is completely unacceptable. We have now had substantial scrutiny from internal auditors and Audit Scotland, who have looked at hundreds of thousands of records and interviewed hundreds of staff. It is important to note that evidence of further deliberate manipulation has not been found.

Thirdly, I offer two observations on social unavailability. First, it was designed, as part of the new ways approach, to be fairer to patients, which Audit Scotland recognised in its 2010 report. It is intended to offer convenience and choice to patients and families without their having to leave the waiting list if the appointment that has been offered to them is unsuitable. Apart from what went on in NHS Lothian, there is no evidence that any board is manipulating social unavailability or using it for anything other than its intended purpose.

It has also been suggested that what happened in NHS Lothian caused a fall in social unavailability, but there is no evidence of that connection. Indeed, as the committee heard from Robert Calderwood, the rate of social unavailability started to fall from a peak in 2010. Exhibit 6 in Audit Scotland’s report shows that.

Finally, but most important, I am aware that in its most recent meeting the committee expressed particular concern about the impact on patients. In my letter to you I have provided data on what has been happening to patients as they have waited for inpatient and day-case care over the period since the new ways approach was introduced in 2008. The annex to my letter contains data on the median wait for all patients, including those who are deemed to be medically or socially unavailable. It shows steady progress over the period, which has been to the benefit of people who have been waiting. In the quarter ending September 2012, the median wait had reduced to 32 days for all patients.

Thank you very much. A lot of the questioning of the previous panel was about when the general issues of how waiting times were recorded and reported were flagged up to different people who have responsibility in the system, and to what degree those concerns were acted on. Indeed, at the centre of Audit Scotland’s report there is a suggestion that, as the Auditor General said, alarm bells should have rung in 2010-11 and those concerns should have been indicated.

At what point did you believe that there were concerns about the recording of social unavailability and the figures that were being reported through ISD’s regular reports on waiting times statistics? When did you think that there might be problems that needed to be addressed?

The growth trend in social availability numbers really happened in 2008-09. As you heard from the previous panel, that was not entirely unexpected. When you introduce a new system such as that, education and familiarisation happen. I agree that tighter waiting time targets and a narrower window in which to place patients might have contributed to that.

Once you get to 2010, the numbers are relatively steady until December 2010, at which point there was a big peak. That peak was down to the very bad winter, in which we could not offer all the services that we would normally offer, and patients could not get to them anyway. Nothing in that period gave us any concerns or—in the words that you used, convener—rang alarm bells for us.

We also got comfort from the fact that Audit Scotland reported in 2010. As the committee has heard, it made recommendations for us about guidance for patients who cannot attend and patient choice codes, but it did not raise the level of social unavailability as being a big issue for us. We did not hear about it as an issue until well into the investigation of NHS Lothian. I hope that that answers your question.

12:15

That answers my question to a degree, but I am not sure whether your answer is that there was no problem with recording and reporting of waiting list information, or whether you accept—as Audit Scotland believes—that there were problems, but nobody had told you that, so you did not know. I am not sure which it is.

I am saying that I do not think that it is necessarily the case that the level of social unavailability at the time was a problem.

I did not ask about that; I asked about the systems for managing, recording and reporting waiting list information.

We knew that our recording systems had limitations. That is why we are implementing the TrakCare system and why we are investing in new systems.

All the way through the period that the Audit Scotland report covers—or the period from the introduction of new ways in 2008 until relatively recently, which we have discussed—the Scottish Government regularly issued press releases and notices about waiting lists. There were loads of them. I have some here—they say, “NHS on track to deliver waiting times”, “Best waiting times ever” and “Waiting time targets achieved”. As chief executive of the NHS, do you sign off such press releases?

I do not necessarily sign off every press release, but it remains the case that what we said then is accurate.

Did you ever indicate to those who sign off the press releases—I presume that they include the cabinet secretary, who is quoted in a number of them—that you knew that there were problems with the systems that generated the statistics on which the reports were based?

The problems with the systems did not relate to generating the statistics; they related to recording social unavailability.

I am sorry—that is a fine distinction that I am not sure that I get.

With respect, it is not a fine distinction. ISD is the well-respected and well-regarded certified authority for publishing such statistics. There has been a range of independent commentaries on the accuracy of the statistics, which include commentary from people such as the Office for National Statistics—a quotation in my letter to the committee records that.

We were comfortable that the statistics that were being presented were accurate; they had been validated. The figures were also entirely consistent with the numbers from other sources, which I referred to in my letter. In a recent report on waiting times, which made comparisons across Europe, the Organisation for Economic Co-operation and Development commented that Scotland publishes quite a lot of statistics, but it did not comment that there is a lack of accuracy.

The statistics as published were accurate. They included statistics on social unavailability, but we could not tell from the systems that we had what the reasons were for recording social unavailability.

Your point, which is in many ways pretty fair, is that the statistics were produced for you by ISD, which is very good at what it does, so you had confidence in the accuracy of the statistics.

However, earlier, ISD said that it had confidence in the accuracy of the statistics because the numbers that it used had been signed off by the chief executives of the NHS boards. Its sense of the accuracy of the statistics and the confidence that we could have in them was based on the chief executives signing them off. If I put myself in the place of the cabinet secretary, my question would be, “When I put out my press release saying that the waiting times are a great success story, who do I look to to give me confidence that, in six months, a year or two years, I will not be facing an Audit Scotland report that says that it cannot have confidence in the numbers?” Would that be your responsibility, as the chief executive of the NHS? Would it be ISD’s responsibility? Would it be the responsibility of the chief executives of the health boards? Where does responsibility for accuracy lie?

I will invite Richard Copland to speak about that, because one of his previous jobs was head of ISD, so he has expertise in this area.

The responsibility lies with everyone whom you mentioned; we are all accountable in some way. The boards are responsible for ensuring that they get accurate data, ISD is responsible for doing the work that Ian Crichton and Susan Burney described to you, to ensure that the data are valid and make sense, and the Government is responsible for having a sensible conversation with ISD to try to ensure that the data become intelligence that can be used to inform our public reporting.

The other important thing to bear in mind is that we take some comfort from the fact that, if you cross-refer the data to other data, including data that come from Scottish morbidity records, the same kind of story is told.

I was director of ISD from 1996 to 2005, so I was involved when we moved from waiting lists to waiting times. I was the one who instituted the chief executive sign-off. At that time, we got from boards a simple spreadsheet setting out the numbers against the nine-month target, or whatever the measurement happened to be.

New ways brought a different arrangement, which I will explain as briefly as possible. That approach involves the board sending an extract of its live system to ISD. Some boards do that every week and some do it every month; by the end of the quarter, a set of statistics is produced. That is a huge body of information, which comes directly from the system. There are discussions about corrections, of course. I should explain a little about the systems before I deal with them, though.

The systems are not waiting times systems; they are the main enterprise systems in hospitals. In recent years, there has been a big push to get clinicians to use the systems more—for example, for ordering diagnostic tests electronically and getting the results back, rather than using a bit of paper, as well as for recording all the information about a patient’s appointment or a patient coming to attend a clinic. That helps in understanding the context in which corrections take place.

An operator who has made an appointment and is entering information into the system uses drop-down boxes that have dates in them. All that needs to happen for a mistake to be made is for the wrong box to be clicked—they might click 2012 instead of today, and that would register as a breach, because it would look like the patient had already been waiting a year. That sort of thing is the subject of discussions about corrections between ISD and boards. We are talking about a massive amount of data collection. It is far removed from the situation in which ISD received bits of paper from boards.

The chief executive sign-off remains because ISD has a cut-off for publication, and the sign-off shows that the chief executive is happy that ISD has all the figures. The arrangement is extremely complex, and is much more comprehensive than what exists elsewhere in the UK. That should give us some assurance about ISD’s role and the role of the boards, which supply ISD with the information.

I will just finish where I started and ask Mr Feeley a direct question. At any time since 2008-09, have you, as chief executive of the NHS, ever indicated to health ministers that they should be aware that although waiting list statistics are at one level, the use of the social unavailability code is steadily climbing or peaking, as it did in 2010, or dropping off, as it did in 2011? Did you ever make ministers aware of that trend?

I will add a brief caveat before I answer. I was not the chief executive or the director general until November 2010.

At that time, we had a discussion with the cabinet secretary about the winter pressures, but we did not have a specific conversation about social unavailability. I am not aware that we ever raised the social unavailability issues with the cabinet secretary. John Connaghan predated me, and I suspect that he will say the same thing, so I will invite him to do so.

Thank you. I guess that the cabinet secretary would have been briefed as part of the normal general introduction of the new ways approach. I cannot remember the date of that briefing, but it would have been between 2007 and 2008. In fact, she would have been pretty central in having sight of the initial guidance.

Derek Feeley is right to have mentioned winter 2010. In that quarter there was a jump of between 4,000 and 5,000 in the social unavailability figures. That was the peak; after that, we took action to work with boards to reduce those figures and to put in extra capacity. That has already been explained by some of my colleagues from the health boards.

I just want to be clear in my own mind. You say that the new ways approach was introduced in 2008 and that the cabinet secretary was central to that introduction. That makes perfect sense to me, because it was important. At no time during the next two or three years did anyone feel that it was worth drawing the cabinet secretary’s attention to the fact that, as the new ways approach played out, there was a tripling in the use of the social unavailability codes. No one thought that that was worth mentioning, sending an e-mail about, or briefing the cabinet secretary about.

You are right. You have heard extensive evidence today already that there was no cause for concern during that period. We became concerned during the latter part of 2011, which was when the issue of NHS Lothian and its offers to send patients for treatment in England first arose. We asked NHS Lothian to investigate that, and the investigation took place in the latter half of December 2011, as far as I recall. In the early part of January 2012, the Government requested a further, much more detailed internal audit, which reported in March 2012. It was in that period that we briefed the cabinet secretary on social unavailability codes being misapplied in NHS Lothian. That is the timescale.

It is not just politicians who are interested in this. I will quote the Auditor General.

“How the national health service manages waiting lists is very important to patients and the public”

and

“public trust was put at risk following evidence that NHS Lothian manipulated waiting lists and disadvantaged patients”.—[Official Report, Public Audit Committee, 27 February 2012; c 1214.]

Do you agree that the episode has led to putting at risk the public’s trust in the way in which we manage our waiting lists, as stated by the Auditor General.

In my introductory remarks, I said that what NHS Lothian did was totally unacceptable. Once we are clear that there was no repeat of that elsewhere in Scotland, I hope that any general concerns that people have about the NHS will be addressed.

12:30

I hope so, too.

The point was made—by Caroline Gardner, in case anyone accuses me of making this up—that the focus of attention of the Scottish Government and NHS boards during 2011 was on whether the 18-week treatment time target was being achieved, rather than how it was being achieved. Is that accurate?

Caroline Gardner went on to say that if you had looked at other information, such as the information that was available on increased use of the social unavailability code, you would have realised that there were warning signs, which merited further investigation. Do you regret looking only at the achievement of the waiting times target? Do you regret not drilling down to ascertain how it was being achieved? With hindsight, do you think that you should have done more?

We were interested in how the target was being achieved and we did a lot of work with boards on how they were going about that. I will ask John Connaghan to take you through some of the things that we did that were as much about how targets were being achieved as they were about whether targets were being achieved, but first I want to say two things.

First, there is a bit of a sense that social unavailability is always a bad thing. It is not. Audit Scotland recognised in 2010, as did a number of parliamentarians, that social unavailability is a way of trying to be fair to patients. Secondly, the key thing for us was that it was about getting quicker access to care for patients. Regardless of whether we consider patients who were reported against the waiting times target or the whole patient population—everybody—we can reliably say that, since 2008, people have been getting quicker access to care.

Right. Social unavailability codes are no bad thing and are fairer to patients, as you said in your opening remarks. As a patient of NHS Highland, I am socially unavailable during the week. However, in paragraph 42 Audit Scotland said:

“NHS Highland had patients with periods of social unavailability with no end date.”

Is being socially unavailable for the rest of one’s life no bad thing? Is that fairer to patients?

With respect, that is not what I said—

No, it is what I am saying. I was reading from the report.

Patients should not be on a waiting list with no end date. A lot of work has gone on to remove patients from waiting lists in such circumstances.

Were you aware that patients were deemed to be socially unavailable for the rest of their lives?

I was not aware of that until I read about it in some of the fieldwork that went around as Audit Scotland collected its information.

At least in part, what I said about new ways being a fairer way of managing waiting times for patients is a reflection of what Audit Scotland said in its 2010 report.

Yes. A social unavailability code is not always fairer to patients and can be a bad thing, if someone is on it until—

It was clearly a bad thing for some patients in NHS Lothian.

Of course—and in NHS Highland. There was no evidence for other boards.

One of Audit Scotland’s observations was that patients should not be held on lists without any prospect of treatment. The committee has had extensive evidence today from chief executives about how patients were perhaps held unnecessarily on a social unavailability code when they should have been in front of their GPs.

An examination of the drop in social unavailability that occurred is interesting. If there was a fiddle, we would expect the affected patients automatically to be transferred to a waiting list. What happened to the waiting list? The waiting list number fell at the same time as social unavailability numbers fell. How do we explain that? We explain quite clearly in the director general’s covering letter that boards are paying due care and attention to that very recommendation from Audit Scotland.

The other factor that plays into this is that if we examine the preceding two years—in other words, 2009-10 and 2010-11—we will find that the number of patients being returned to their GP averaged about 3,800 to 3,900 per quarter; however, the last two quarters of statistics show that on average 4,900 patients were being returned to GPs. That has led to a drop in social unavailability to tackle the very issue that you have raised.

So do you think that Audit Scotland has got anything inaccurate in the report or do you agree fully with the accuracy of a report that, in Audit Scotland’s own words, contains the most data and is the most extensive that it has ever done?

I should not forget that Richard Copland has an answer to your NHS Highland question.

I get an opportunity to exchange views with Audit Scotland on reports’ accuracy before they are published, so I have been able to make comments in that respect. However, there are still some views that are inaccurate. For example, paragraph 35 on page 20 says:

“Towards the end of 2011, around the time concerns were raised about ... Lothian, the use of unavailability codes began to reduce”

but exhibit 6, which sits below that paragraph, shows that their use was beginning to reduce in 2010.

That said, I am very comfortable with the recommendations that Audit Scotland has made on the back of its analysis. I agree that NHS Scotland needs to do the kinds of things that Audit Scotland has asked us to do and that we should be monitoring the use of waiting times better. That is why we are introducing the changes that we are introducing and why we feel that the use of patient-advised unavailability is going to be a significantly better means of doing this than what we had before.

To pick up on the convener’s earlier comment, I think that it is in everyone’s best interests to ensure that in two or three years’ time we are not sitting around this table asking ourselves exactly the same questions. I certainly have no desire to do so, and the best way of avoiding that is for us to get on and do the things Audit Scotland has said we should do, and deliver more transparent and better-recorded information about why patients are choosing not to be seen at a particular time and how the NHS is responding to their choices.

I am glad that you used the word “transparent”, because that brings me to my final question. One of the key messages on page 34 of the report says:

“Better use of the available information could have”

identified

“concerns about the use of unavailability codes.”

However, I am concerned about the following sentence, which says:

“It could have also identified wider pressures that were building up in the system around the capacity within NHS boards to meet waiting time targets.”

What discussions have taken place between the Scottish Government and board chiefs or chairs about the capacity pressures that boards might have been facing in meeting the waiting time targets, which of course we welcome?

John Connaghan will pick up that question, because it is directly relevant to your earlier question about the “how” rather than the “what”. However, before he does so, I ask Richard Copland to give you some comfort about the NHS Highland numbers.

I took up my role just around the turn of the year and found this issue to be very visible. When these things emerged in Audit Scotland, I began to ask some boards whether any specific issues had come up. I must apologise—I cannot remember the exact number—but in NHS Highland, which I believe will have responded to Audit Scotland, the number of people who did not have any end date was a high single digit or a small double digit. It was also entirely an administrative error. When I got on to this—which was perhaps not surprising, given my background in figures—I began to look for what might or might not give us comfort.

We have heard about the unadjusted wait, where social and medical unavailability are added back in. If we look at the median or the 90th percentile for the number of days either of those applied, we will see that the number of days that are applied for medical unavailability is always higher. That also adds in any clock resets; if a patient CNA’d—could not attend—a number of times and the board resets the start of the waiting time to zero, it does not take the patient off the list, although it could in certain circumstances send them back to their GP.

I have looked at the median unadjusted wait across Scotland. It has come down, and it was well below the target of 12 weeks at the time. However, I took things a bit further and looked at the big surgical specialties of trauma in orthopaedics, urology, general surgery and so on. I looked at the median only in preparation for coming to this meeting, and considered mostly the boards whose representatives have been here today and Lothian NHS Board. In relation to the median unadjusted wait—where everything is added back in—they still met the target in the vast majority of instances. Around 90-odd per cent were doing so, although there were one or two aberrations around some specialties.

I then looked at the 90th percentile, which is the long waiters. There will genuinely be some long waiters in there, because some folk will be medically unavailable for a while. They should be reviewed every 13 weeks. However, it will perhaps give the committee some comfort that unacceptable behaviour in a number of the specialties that I looked at—such as urology, plastics and general surgery—was restricted to NHS Lothian, which was a significant outlier in the 90th percentile. We can see that some of the other boards, particularly earlier on in the 2008-12 timeframe, had some challenges with long waiters, but from 2010 on NHS Lothian was very much the outlier.

I am sorry for adding that.

Okay.

The question about capacity issues has not been answered.

That was a very good question, as it really is at the heart of the question whether the Government’s and the NHS’s performance in the area is believable.

I want to introduce the concept of triangulation, which involves looking at not only one piece of data but several other pieces of data to back up the picture. We have already heard from my colleague Richard Copland about median waits as one aspect of tracking performance, but—perhaps this answers the point directly—we also paid attention to an old measure that is still valid now: waiting list numbers.

I think that we can all remember when waiting list numbers were the national target. We changed the target to waiting times in 2005-06, but waiting list numbers are also important because, if we see a build-up in them, it indicates pressure and capacity issues.

We were very alive to that issue, and we have been so over the past 10 years. We have introduced various support mechanisms to boards on demand, capacity and queue methodology. It is quite interesting to look at waiting list numbers today compared with those in 2005-06. In 2005-06, there were more than 100,000—110,000 or 112,000—people on the waiting list; now, there are 52,000 or 53,000. The numbers are the lowest that we have ever had. Colleagues have already expressed views on the additional growth in capacity and consultant numbers that have led to that.

There is another factor that is also worth mentioning. If we track hip, knee and cataract procedures, which are the most common procedures by volume and cost, we will see that, last year, the highest ever levels of those operations were carried out in Scotland. That is another factor to explain drops in social unavailability and waiting times.

I could say more about our engagement with boards, but I am mindful of the time.

Before we move on, I want to clarify a point with Mr Feeley about exhibit 6 on page 20 of the Audit Scotland report. I am looking at the graphs. Please forgive an old maths teacher, but it looks to me as though the unavailability rates in both the out-patient graph and the in-patient and day cases graph are still rising in December 2010 and that the discontinuous precipitate drop in both rates starts in December 2011, which I think is essentially what Audit Scotland said.

No, I think that what Audit Scotland said was that the use of unavailability codes began to reduce. My contention is that the use of those codes began to reduce in December 2010.

12:45

No, the use of those codes dropped a bit in December 2010 and then went up again. Surely the reduction trend happens in December 2011—if we are looking at the same graphs.

We are looking at the same graphs.

At best, you might say what Audit Scotland said in paragraph 35—that the reduction in social unavailability codes happened as a result of what was going on in Lothian. We should remember that it was in October 2011 that we first heard about Lothian, largely in relation to people going to England. The PWC report did not come out until March 2012.

To be fair, I think that Audit Scotland said that the reduction happened at the same time, not that the reduction was because of what happened in Lothian. The issue is whether it happened at the same time. We are looking at the same graph and we will have to agree to disagree. It looks to me as though the trend drop is in December 2011.

That relates to Mary Scanlon’s question on whether there is anything that we disagree with Audit Scotland about.

Okay. We will now have questions from Mr Scott.

I will leave the maths teachers to fight that one out.

We heard earlier that the new ways approach introduced 99 pages of regulations that health boards had to implement. We heard that that introduced considerable pressure into the system on staff at all levels. Was that pressure reflected in the monthly meetings that health board chief executives had between themselves and with you?

At many of those monthly meetings we talked about waiting times and where we were in relation to the trajectories for achieving improvement in waiting times. We talked about what we were achieving and how we were achieving it. That is the kind of environment in which some of the things that John Connaghan talked about were raised.

We talked about the work that we were doing on better understanding demand and capacity and understanding how people were queuing for care, as well as the work in task and finish groups that involved, for example, specialist clinicians going to look at orthopaedics to tell us how that might work better. We discussed how we were doing on progress towards the target but also how we were seeking to achieve the improvements.

I completely accept that we are talking about a big new system that you were putting in place. Ninety-nine pages of regulations must be pretty daunting for anyone—I speak as an ex-farmer who used to get that from the Government all the time.

This is really heavy stuff, and I want to establish whether, at any stage during the monthly meetings, health board chief executives said “Look, you are putting an enormous amount of pressure on our staff to implement this big new system.” There was a huge amount of political pressure at the top because, as the convener said, the health secretary issued press release after press release to say that things were getting much better. Were chief executives not reflecting what was going on in their health boards to you in respect of those pressures?

It is probably better for me to comment on the detail of that. As part of the introduction of any new system, such as the move towards 18-week referral to treatment standards, we set up an architecture of support with boards, in which issues are considered such as how to balance available resources against the task that is required.

It is quite clear from all of the support work that we did and the literature that we published that we had concern for the redesign and transformation of the systems to make them sustainable within the available resources that we had. There is clear evidence of that, and we can provide it to the committee if required.

On the point about staffing, I have just come across the document “18 Weeks and Workforce Planning: a Resource Pack” in my papers. That is a guidance pack for boards to address the very issues that you have just raised. We were certainly live to those issues—dealing with them is part of good risk management in the introduction of any new standard or target.

Okay. I think that you said earlier, in relation to social unavailability codes in particular, that you asked NHS Lothian to investigate when the issue first arose—I will try to avoid using pejorative terms, so that I do not get jumped on. Is that right?

Yes. Let me go back over the sequence of events. In October 2011, we were alerted—by the press—to the fact that offers were being made to patients to travel to England at relatively short notice. We did not think that that was quite right, so we asked NHS Lothian to investigate the issue. It took a month or so before we got a report on that investigation. I cannot remember the date of the report, but it was certainly produced before the end of the calendar year; it probably came out in the early part of December 2011.

That report—which I am sure is available for committee members—raised certain questions in our minds about the use of unavailability codes in Lothian. At the time, we were not particularly clear about the level of adjustment or the mechanism by which it was achieved. On 6 January 2012, after reflecting on matters over Christmas and the new year period, we wrote to NHS Lothian to tell it that we thought that its report needed to be investigated further and that we would like the process to be quasi-independent. We said that we wanted NHS Lothian to have a thorough independent audit carried out and to report back within a month or two. We received the independent PWC report in March 2012. At that point, it was apparent that the social unavailability code had been misapplied in Lothian. That is the sequence of events.

That is fair—thank you for that clarity.

Did health ministers instruct you to do all that, or was the process driven by your team at senior official level?

I suppose that the answer is that we communicated our concerns to ministers. Ministers were aware of the fact that we instructed a further detailed audit to be carried out in January 2012. In fact, shortly after that request was made to NHS Lothian, ministers decided that, rather than reporting back to NHS Lothian, the independent audit report should report to the Scottish Government, so the report came to us in March 2012 rather than to NHS Lothian.

Thank you; I quite understand what you have said.

Following all that work with NHS Lothian, did you ask the other 11 health boards to instigate their own investigations, using external or internal auditors, into what was going on?

Yes. I cannot remember the exact timing, but around that time we decided that we would ask every other health board in Scotland to carry out the same, detailed internal audit, which, if anything, was even more extensive than the audit that was carried out by Audit Scotland, albeit that it involved a different time period. The number of transactions that were looked at and the number of staff who were interviewed as part of that further internal audit across every other board in Scotland ran to many hundreds. The audit was published at the back end of last year and is available for the committee to look at.

Earlier, the chief executive of NHS Tayside told me that his work had been initiated off his own bat. That does not strike me as being consistent with what you have just said.

It is entirely consistent with what I have said, because the chief executive of NHS Tayside was referring to an exercise that he carried out on culture and behaviour in NHS Tayside, whereas the exercise that I have just talked about was the internal audit that looked at the application of codes.

So the work that he described—he is no longer here, so this is a bit unfair—was not about social unavailability codes; he was talking about something completely different.

You should remember that, at that time, he already had the internal audit report. In looking at that report, he wanted to assure himself that there was not a widespread issue with culture in Tayside. I am not au fait with the precise nature of the exercises that he carried out, but that is what I took from the explanation that he gave earlier.

Yes, I think that he made those comments in response to Ms Scanlon’s question about bullying.

Okay.

Once the audits, whether internal or otherwise, had been initiated by your team—or rather, by the director general and the centre of the NHS—what was the ministerial role in the process? Did ministers ask for those reports to be presented to the Government so that they could see them, or were they content for them to go to the health boards? What was the reporting line as far as those reports were concerned?

I think that the reporting line was that the reports would be made public to both ministers and boards. We should remember that boards have their own governance structure and should be held accountable for such matters. The reports were transparent and available to both ministers and boards.

When the reports were coming back, were you as a team considering why things were happening as a consequence of the target? Was that part of your work?

It is an additional step to say that everything was happening as a consequence of the target. However, by that stage, we were certainly looking at the situation in the round, not least because we were conscious of the fact that we had on the horizon a 12-week treatment time guarantee and the need to make some good guidance available to boards so that they could work towards the TTG. We were therefore learning all the time.

What we have tried to do with the new ways approach since 2008 is ensure that the guidance is clarified and refreshed as and when required, rather than issue repeatedly different versions of the guidance. We try to keep it live, if you like, as far as we can. I think that that is something that we should learn from, too, as we move into the treatment time guarantee.

I take that point, but it would be fair to say that the target is very driven and one to which ministers of any Government would give a huge amount of attention, and it is probably the most central target that the director general must deal with. Is it therefore given credence over everything else? Does that determine a set of analysis that leads you in one direction?

No, it does not. It is one of a number of targets in the suite of health improvement, efficiency, access to services and treatment—HEAT—targets. I feel that I would be failing in my duty to the Public Audit Committee if I was not to put having the NHS in financial balance at the very top of my responsibilities. However, there is no kind of pecking order in the HEAT targets.

We think that it is important to try to reduce waiting times for patients. One of the means that we have chosen to do that is to set a target. All the evidence shows that that is a way of getting people engaged. If we set a target for something, people pay attention, allocate resources and their time accordingly, and see the target as a priority. We have been trying to get people to see waiting times as a priority. As the committee heard from our colleagues from the health boards, they view the target as challenging but achievable. They also recognise that we have done our part to help them achieve it by doing the work that we did on the “how”, which John Connaghan talked about earlier, and by putting in some resources.

It is probably worth mentioning at this stage that one of the things that we are very fortunate to have in the NHS in Scotland is the Scottish partnership forum, in which we bring together the Government, managers of health boards and the representative organisations. We had conversations with them back in 2009 I think—John Connaghan will correct me if I am wrong—about a set of principles that would underpin the way in which we undertake performance management. One of those principles is that clinical outcome for patients always takes priority over achieving the target. The target is a means to an end, which is whether we can get patients better access to care.

My final question is: why do you think that it was ultimately a whistleblower who explained what was going on?

I do not know exactly how that got into the media. I first heard about the issue when it was reported in the newspapers. Once we had heard about it, from whatever source, we then did the right kind of things. That is why the sequence of events that John Connaghan rehearsed with you—ensuring that it was properly investigated, getting an independent source to look at what was going on in Lothian and, on the back of that, trying to satisfy ourselves through internal audit that that was not being repeated elsewhere—was an entirely rational set of things to do on the back of intelligence. We should be open to getting intelligence from everywhere. If people in the NHS have a problem with something, I would much rather know about it than not.

Sure.

13:00

I asked the previous panel specific questions about systems on the back of the Audit Scotland report, which highlighted the inadequate controls and audit trails and the limited information that was recorded in patient records.

As we are aware, all the boards are moving towards a common approach to producing statistics, not necessarily by using the same system, but by producing the same reports using common criteria. I asked the panel who has overarching responsibility for and control over that, but there did not seem to be an answer, other than that the boards were co-operating with each other.

Richard Copland knows quite a lot about what is occurring, so I will ask him to start off.

I told the committee a few minutes ago about the considerable chunk of data that is taken out of live operational systems in a board and sent to ISD, which puts it through a fairly complicated validation process. That system has been refined since 2008, and there are different layers of accountability for signing off on the figures. One layer is the chief executive of a board, who is the responsible accountable officer. He or she will sign off to say that they believe that the figures are correct and truly representative of their board’s performance.

ISD will take that information and produce a large set of publications on it, which will contain a lot of data. It will satisfy itself that the data are statistically sound and represent what has happened. Mr Feeley is the Scottish Government official who is responsible for overseeing the NHS, and our day-to-day role is to examine the information that comes in, and look at a wider range of information such as complaints and other statistics to see whether the data make sense.

One challenge that we face is that social unavailability and medical unavailability have no doubt helped many patients. As NHS Greater Glasgow and Clyde said, and as Audit Scotland recorded, boards made use of unavailability. Rather than resetting the clock for the target at the time, which was 12 weeks, they gave patients a period of social unavailability for a shorter time. In some ways, we are making a pariah of social unavailability.

One caveat concerns what was happening in NHS Lothian, where there was no engagement with patients. It was not as if—as the witnesses from the boards said—the board had no ability to record exactly what was said in some of the telephone conversations, which have now been superseded by writing to the patient. NHS Lothian was not recording the information, which came to light as a result of what is known as retrospective adjustment between publications.

With regard to the statistics, there will always be an element of retrospective adjustment. Errors in data will happen right at the end of the period, and ISD will scrutinise a change and decide whether it is acceptable. However, the NHS Lothian level was three times the Scottish average, which would then drive up the average. That is one of the clues.

As Mr Feeley said at the outset, the Scottish Government accepts the recommendations in the Audit Scotland report, which will mean that there will be even more scrutiny. I do not suggest that there was no scrutiny before, but there will be more scrutiny of some of that information at a much more granular level—for example, down at the specialty and sub-specialty levels. That is the way that the NHS and health services around the world are being driven.

It is not that there is a single person who is accountable; a number of people are accountable for their bits along the way, which I think makes the system much stronger. Below that, at board level, the board itself, individual managers, clinicians and governance and audit committees are all accountable for things.

You are saying that there is a sort of collective responsibility to make the system work.

There needs to be collective responsibility to make it work. It relies on people at the point of care doing the right thing, all the way through to NHS senior managers doing the right thing. There needs to be some level of collective responsibility.

As you know, we have replaced social unavailability with something called patient-advised unavailability, and that might give you some of the reassurance that you need about how things will happen reliably across the country.

I am concerned that the data be gathered in such a way that we can rely on the statistics that come out, rather than having interpretations that might vary between boards, such that we are not comparing like with like.

Patient-advised unavailability means, first, that the patient is much more in charge, which is always a good thing. Secondly, the board must write to the patient to confirm that a period of unavailability has come along. Thirdly, there are eight codes that need to be applied. The reasons for social unavailability will be much more transparent.

There will be better records and a better audit trail. Crucially, that is all enshrined in the regulations that sit alongside the Patient Rights (Scotland) Act 2011. It is not an optional thing—it is regulatory. That ought to start to bring us the kind of intelligence that we need to avoid our having to sit around this table to discuss the matter again in two or three years’ time.

Your answers lead on to my questions. How many patients were coded as socially unavailable because they wanted to be treated in their local hospital or by their own consultant?

I do not have a number for that, which is partly because the systems do not record it, as you heard earlier from the witness from NHS Greater Glasgow and Clyde.

Would your job analysing the lists—and our job as the Public Audit Committee—be easier if we could work out how many patients had been coded in that way for that purpose?

I am always in favour of better information, which can help us to make better decisions, so I think that it would be helpful to have that information. That is what we are trying to get from the patient-advised unavailability material.

That was one of the recommendations in the Audit Scotland report of March 2010. Why did it take until August 2012 for that code to appear in guidance?

In 2010, Audit Scotland recommended that a specific patient choice code be applied. We had issued guidance to health boards in December 2009 that they could and should use social unavailability as a means to achieve that end—to give people the opportunity to elect to be seen by a specific clinician or at a specific location.

Even if we had introduced a patient choice code at that time, we did not have systems that would have been able to support the data collection that would have been required to underpin it. We would not have derived significant benefit from that. We were already starting to turn our attention to the new target—the treatment time guarantee—and to consider what legal arrangements to put in place for that. We saw it as an opportunity to be much more definitive about all those things.

That is an important point. One of the recommendations in the 2010 report requested that we have a patient choice code for location and for consultants. Audit Scotland and, I am sure, ISD were aware of that in the latter half of 2010, as they debated the make-up of the report. It is clear that we gave effect to that requirement to have a patient choice code in the guidance that was released in December 2009. However, we could not split down the fine detail between location and consultants at that point. We are now able to do so as we move towards the implementation of the TrakCare system, but the recommendation was clearly given effect in 2009.

It was given effect, however, by bundling all the patients and coding them as socially unavailable. That has led in part to the current situation in which the Public Audit Committee is unable to analyse fully why people are being coded in a particular way. The recommendation in question was not fully implemented, partly because the systems were not in place back then. Now they are in place. Why were the changes not made back then to allow these things to happen?

There are probably a couple of practical reasons for that. First, you should remember that at that point in time we were considering the implementation of the treatment time guarantee, which would produce the new patient-advised unavailability code. Secondly, systems development cannot happen overnight; it requires a lengthy timescale and I would not care to venture what might happen if you rushed such a complex matter and got it wrong.

You have heard from Mr Calderwood and other colleagues that they are in the process of merging systems. By the end of this year, 70 per cent of the population in Scotland will be covered by TrakCare and for the remainder of the population the systems will be rewritten to allow these things to be captured. All of that takes time; systems need to be tested and so on. Given all that, I think that we have done pretty well in giving effect to the requirements that Audit Scotland placed on us at the time to allow that to happen and to give patients access to that facility. However, with regard to the ability to record the fine detail of all that, I hope that you accept my recommendation on the risks of rushing such work.

I am trying to manage our time. Bob Doris, Willie Coffey and Ms Baillie still have questions. I ask colleagues to be as concise as possible.

I shall do my best, convener.

I think that, this morning, we have heard reasonable explanations, to which no one has particularly objected, about the rise and fall in the use of social unavailability codes. There seems to be consensus that some of the reasons are valid, but the concern that the committee is expressing is that most of this was not and is not auditable. We have been round the houses on the question of why it could not be audited; I think that in his last answer Mr Connaghan explained some of the issues in that respect.

I want to go over some of the reasons for the increase that we have heard not just from this panel but from the health boards, in order to ensure that I understand them clearly. First, the system that was used before the new ways approach had unavailability status codes. As a result of those, people would be sent back to their GPs or put on a separate, or what has been referred to as “hidden”, waiting list. If, in the move to a new system, a third of patients who otherwise would have been designated as being unavailable on another list are brought back into the system, that will by definition increase the number of people who are deemed to be socially unavailable. The difference is that, in the new system, they are all on one waiting list and are quantifiable, instead of sitting hidden elsewhere. Is that a reasonable explanation for some of the increases?

The increases can be taken in chunks. With a new system, one starts with zero; no one is on it, so you have to populate it, and one can see that populating happening through the first year or so of the implementation of new ways. According to the first quarterly report—from memory—about 5,000 or so had gone on to the system. I guess that when we look back at the introduction of patient-advised unavailability as described in the new TTG system, we will probably see the same set of figures. The populating of the new system from new ways was clearly a factor in the increase.

13:15

If you examine the jumps in social unavailability, you will find that they almost always occur at two times of the year, with a gentle rise over the summer. In the winters of 2009-10 and 2010-11, however, there were significant spikes of about 4,000 to 5,000. It is clear that the NHS takes a long time to recover from such spikes. The 2010 spike was during the worst winter that I can recall for cancellations, for patients saying that they were not going to travel and for boards having to say that they were sorry because staff were not in so people could not be given routine operations. Because it was such an issue, we took specific action and said to boards that we needed to put additional capacity in the system in order to start to treat the backlog. As we go through the timescales, there are two or three reasons in the history of social unavailability that explain the rise.

You have helped me to skip one of my questions, because I was going to ask about peaks during the winter.

For clarity, on the graphs to which the convener referred, some of the dubiety might be because there are two graphs on page 20. One of them refers to out-patients and shows a steady and consistent decline in use of social unavailability codes from June 2011. The other is for in-patients and day cases, on which it is undeniable that the decline is from December 2010. That is not a matter for debate; it is a matter of public record.

I assure Mr Doris that it is a matter for debate. That is a completely different reading of the graphs but—fair enough—we both have our readings on the record.

I can assure you, convener, that I have the ability to raise issues that I want to raise when it is my turn to ask questions, if you do not mind.

Absolutely.

I have put that on the record.

I can give the statistics to which the two graphs refer. In December 2010, social unavailability for in-patients and day cases was at 32.5 per cent and, for out-patients, the figure was 10.8 per cent. By September 2011, the figure for out-patients had fallen to 9.7 per cent and the figure for in-patients and day cases had fallen to 28.3 per cent. So, in percentage terms, there was already an evident fall for both types from that peak of December 2010.

I certainly should not have a debate about the numbers with a maths teacher.

That is a reasonable point, Mr Feeley.

The only thing that I will say is that NHS Lothian is a factor. The picture looks different, depending on whether NHS Lothian is in or out.

That is helpful. The figures are particularly helpful because the clear trend started before the NHS Lothian situation emerged. We can say clearly—irrespective of how individuals wish to nuance things—that explanations have been given for the increase, and that there is consensus that the explanations appear to be valid, even if that cannot be audited. There are clear explanations for the fall: when increased demand was recognised, increased resources were put in. Another reason for the fall was that it was no longer winter, which would tend to result in a fall. Is that reasonable?

As I explained earlier, it is a material factor that, as a result of behaviour among the board chief executives—I think a colleague from NHS Tayside explained about training—patients were no longer held with open-ended dates on a social unavailability list. It is evident that more patients were being returned to GPs with no more prospect of treatment than they had previously and that, in that respect, boards cleaned their lists, so to speak. That is another factor that we need to consider as part of the reason for the fall.

I have a final question on the factors. As boards are gearing up for patient-advised unavailability, which is about having a discussion with patients on their choices and preferences, is it reasonable to suggest that the more patients are aware of their rights, the more they will decide to use them and the more we can expect an increase in patient-advised unavailability, which in the past was called social unavailability? Is it reasonable to suggest that we should watch for that in future figures?

As John Connaghan said, our expectation is that the figures will start to rise. As I mentioned, social unavailability was introduced to be fairer for patients, and Audit Scotland’s commentary on it in 2010 confirmed that it was, indeed, fairer for patients.

What we are trying to do with patient-advised unavailability is make a step change, to put the patient absolutely at the centre of decision making about whether they are unavailable and to ensure that the patient has a record that shows why they were unavailable and the period of that unavailability. We are trying to ensure that we have better records than we had in the past about why decisions are taken, so that decisions are auditable and transparent.

I have no more questions. There are compelling arguments that explain the trend and I do not think that any committee member has argued that they are not valid reasons—although they might not be all the reasons—for the trend. It is just deeply disappointing that none of it was auditable; audit scrutiny is the purpose of this committee.

Thank you for your explanations. I hope that the next time you come to the committee we can consider the audit process. That would be helpful.

As I said to the previous panel, the Auditor General set out to see whether there was widespread manipulation of lists across the NHS in Scotland. She did not uncover evidence to back up an assertion that there was such manipulation. In their evidence, the Scottish Government witnesses have reiterated that there is no evidence to support such an assertion, and the three health boards categorically rejected the accusations as being false and baseless. However, accusations remain on the table. Members who have made such accusations are duty bound to withdraw them.

Mr Feeley, how can we ensure that we will be in a better position in a couple of years’ time, when we might have to consider a similar report? You said that you have accepted Audit Scotland’s recommendations. How can we ensure that false and baseless accusations cannot be levelled at the NHS?

TrakCare will be much more widely implemented—if not fully implemented—so there will be a direct connection between the clinical interaction between clinician and patient and the information that we get about decisions that affect patients’ care. We will have patient-advised unavailability numbers, with eight availability codes, so we will have much better information about the reasons why people are socially unavailable. We will also have better information on the periods for which people are socially unavailable.

We will therefore have much richer intelligence, which ought to be “auditable”, as Bob Doris put it. That will be the key learning from the Audit Scotland report, which I very much welcome. Those improvements need to be made.

Thank you. On the main issue, are accusations that widespread hidden waiting lists operate in the NHS in Scotland true or false?

There are no hidden waiting lists.

Thank you.

Mr Feeley, would you say that there was certainly misuse of the social unavailability code, which caused the numbers to rise substantially and then fall dramatically after the practices in NHS Lothian were uncovered?

There is no evidence of that.

Is the truth of the matter that there is no evidence to support a contention that all was well? [Interruption.] That is exactly the point that Audit Scotland made. Members would be wise to pause and reflect that Audit Scotland did not deliver a clean bill of health.

Mary Scanlon asked whether I disagreed with Audit Scotland about anything. Caroline Gardner could not have been clearer in the evidence that she gave to the committee. She said:

“As we have said in the reports and as I have said today, we have not found any evidence of manipulation at all.”—[Official Report, Public Audit Committee, 27 February 2013; c 1238.]

That is a strong statement.

I think that she went further, but I do not have the Official Report in front of me—I am happy to fish that out for the committee.

I do not intend to trade information about graphs or statistics. I realise that I cannot match Richard Copland’s enthusiasm in that regard, so I will not try. However, I want to try to understand processes. Based on what Mr Crichton said to the committee, I am assuming that reports on the statistics go to Mr Copland.

We get a number of reports. I was one of the ones to get part of that five day pre-release just before the last publication. Like other parts of the Scottish Government, we ask ISD to do additional bits of analysis for answering parliamentary questions—[Interruption.]

Colleagues, we have done well for a long time, but there are too many conversations starting now. We do not have to go on for much longer.

I am sorry to interrupt, Mr Copland.

So yes, we get statistics. Every month, via ISD, we also get some high-level performance figures. ISD takes them from its systems and, because publication is on a quarterly cycle, they do not go via the new ways approach. We get a raft of information to scrutinise.

When you say “we”, do you mean yourself, Mr Connaghan, or whoever is director general for health at the time?

The information is very much operational information that I and some of my staff would take up with a health board. If I was concerned about something—and okay, I am not long through the door—I would raise it with John Connaghan. The information is not hidden and anyone who wants to see it can see it; it is for operational, day-to-day use.

Would that information have been available to you as part of the chief executive meetings?

For the period 2000 to 2005, I was head of the access team. The sequence that Mr Copland has outlined is correct, but one important fact needs to be borne in mind. All the data that we get on pre-release access are made available publicly anyway. As civil servants, we get the information five days before it is released so that we can get our minds around it in order to brief ministers and so on.

We supplement the data with additional management information. As Mr Copland said, that management information comes in monthly, and it mirrors what would have been available to us quarterly. We use that to spot any early-warning signs of capacity issues, for example, such as whether waiting lists are growing. If we spot any such signs, we can either debate those issues with the chief executives at the regular monthly meetings or, if we consider it to be necessary, we can have a special meeting with a health board if we find something to be concerned about.

Things happen from time to time. Earlier today, Mr Calderwood outlined the fact that there was a spike in ophthalmology and orthopaedics, which was tackled shortly thereafter. That is the kind of issue that we would raise.

I simply wanted to establish that you saw the statistics and that they were available to you. Indeed, you have gone further and said that they were supplemented by additional information.

You mentioned that you brief the cabinet secretary, so the cabinet secretary would have available to her or him the range of statistics that you would have.

John Connaghan can answer that question, but it is important for the committee to recognise that the statistics are generally available. They are published.

I understand that.

They can be accessed on the ISD website. They are not in any way secret.

I agree that you and the minister get lots of information. I am simply asking whether such information would be drawn to their attention or whether you would expect the cabinet secretary to interrogate a website.

No; it is drawn to their attention.

Okay. It is helpful to know that.

Audit Scotland raised in its original 2010 report the problems with social unavailability; I am happy to provide you with the quotations. It talked about not accurately recording all relevant information about patient waits, about the difficulty in demonstrating that all patients are being managed correctly in line with the new guidance, and about wide variations in the levels of social unavailability across 12 health boards. That implies that boards were applying the code differently, which has implications for patients. That contradicts what I believe you said earlier—that the Audit Scotland report did not raise issues of social unavailability. Clearly, it did.

We can check the record. If that is what I said, it is not what I meant to say. I do not think that it is what I said. I said that Audit Scotland had no concerns about the level of social unavailability. I recognise that it was concerned about the processes that underpinned the use of the code and about variations between boards, but it did not say that social unavailability was rising at a level that was far too high and that that was causing it concern.

13:30

If I quoted you incorrectly, I apologise. We can check the Official Report.

Audit Scotland asked in its report for guidance to be issued. That was in March 2010. It reminded you in March 2011 and was told that that guidance was imminent. Why did it not come forward?

We have been over this ground. In December 2009, we made provision for boards to use the social unavailability code to recognise patient choice. That gave patients what they needed. That was our first—

I am sorry to interrupt, but why was Audit Scotland told that the guidance was imminent rather than that you had already published it?

There was an on-going set of guidance issues. John Connaghan has the list in front of him.

I can clarify that. In the 2011 follow-up report, three pieces of guidance were mentioned. One was on effective patient booking and a first draft was issued in February 2011. It was characterised as being a live document. With a live document, as we go through the months we take the queries and revise it. That was the first piece of guidance and it was issued at or around the same time that the audit report came out. The second piece of guidance was an additional NHS Scotland access policy, which was issued in April 2011, a month or so after the publication of the report. The third piece of guidance was the national NHS Scotland waiting time guidance. We have already explained how that transmitted into the TTG guidance. It was eventually published in 2012. Of the three pieces of guidance that were required, two were published swiftly and one was work in progress.

Thank you very much. That is a helpful explanation.

I have two very brief questions. The cabinet secretary said that both the rise and fall in social unavailability were somehow down to IT systems, yet we have heard today that there was a need for further education, that there were capacity issues—which you have highlighted—and that the fall occurred during preparations for the introduction of the treatment time guarantee in October. Are the health boards wrong?

I think that we have rehearsed the reasons for the rise and fall in social unavailability.

So, it was not down to IT.

The IT issues are around being clear about the reasons that lie behind social unavailability.

At no time did the cabinet secretary say that the whole problem was down to IT systems; he said that that was just one of the issues behind the problem.

We now have both interpretations on the record.

Thank you, convener. I am happy to provide links to the various statements made by the cabinet secretary.

I have a final question. We have heard a lot about accuracy today, but the key issue for me is not so much the accuracy of the statistics as who was monitoring them. With the benefit of hindsight, do you think that you should have picked up the problems with the rise in social unavailability before NHS Lothian?

No.

You do not think that there is a problem.

The rise in social unavailability in the period 2008 to 2010 occurred largely for the reasons that we have explained to the committee today. It was not an unexpected rise; it was a gradual and steady rise. Where there was a peak or spike in social unavailability, we looked at that spike and took appropriate action. There was no reason for us to act in any other way.

Thank you, convener.

That brings this session to a close. I thank Mr Connaghan, Mr Feeley and Mr Copland for giving us so much of their time. It is very much appreciated.

13:34 Meeting continued in private until 13:39.