Examination of Witnesses (Questions 20-39)
MR NIGEL CRISP AND MR PAUL JENKINS
WEDNESDAY 6 FEBRUARY 2002
20. Your costings presumably are on the basis of what you have to pay to the nurses in NHS Direct itself?
(Mr Jenkins) Yes.
21. Of the ones that have come from the NHS, about 80 per cent could have been working otherwise in the NHS?
(Mr Jenkins) That is right.
22. It seems to me that you are under costing the real cost to the NHS of NHS Direct in that if those nurses were not working for NHS Direct you might be able to use them as NHS nurses rather than agency nurses who I imagine always cost you more.
(Mr Crisp) You are getting into the process of planning and priority setting which we go through in the department and in terms of how you spend your money you have to make decisions about whether you believe that this service has done a lot of things to speed up and improve care.
23. I am not saying that it is the wrong decision to use the nurses in this way, but if you are going to judge whether it is right to do it you have to get the costs right. It seems to me the overall cost to the NHS is not the 80 million a year but rather more than that because, on top of that, you ought to be adding the extra cost of having agency nurses in A&E who you otherwise would not have had to use.
(Mr Crisp) We should not assume that these nurses would go and work in A&E. You can make comparisons with nurses working in coronary and heart disease or in A&E.
24. We are offering a completely new service and as a result of that we have fewer nurses available to work in A&E departments or wherever. Therefore, we are using more agency nurses within the NHS so the overall cost to the NHS is not just the direct cost of NHS Direct; it is also the opportunity cost that you have lost because you have to have more agency nurses working in other areas of the NHS.
(Mr Crisp) The argument, with respect, applies to our extension of coronary heart disease services, does it not?
25. I have no doubt it applies elsewhere but if we are to judge the value of NHS Direct we ought to be judging it against the real cost of NHS Direct which is not just the direct cost but also the additional cost that the rest of the NHS has suffered.
(Mr Jenkins) One of the things that NHS Direct is doing, particularly with out of hours GP services, is reducing some of the demands on medical manpower, so we are using nurses to help manage another problem in terms of manpower within the NHS, in terms of the number of available doctors. One of the interesting byproducts of the NHS Direct call centre infrastructure is that we have been able to use that to deliver the NHS's own agency staff service, NHS Professionals. The telephone side of that service is now delivered from our call centres and using a lot of the same infrastructure and expertise. That is a way of reducing to the NHS the costs of employing short term, temporary staff. It is a byproduct of the fact that we have invested in infrastructure for the delivery of the main NHS Direct service.
26. What you appear to be saying is that there are other ways in which NHS costs have been reduced by the introduction of NHS Direct. Fine. If so, let us include those too. What I am suggesting is we really ought to get the costs right and not just base ourselves on what is the initial, direct cost of NHS Direct because there may be other costs and perhaps other savings involved. If we are really to get a handle on whether this is worthwhile or not, we ought to get the costs right. It seems very odd that we have been talking for years about the government needing to introduce these major schemes as pilot schemes, first of all, and then to evaluate your pilot scheme. When you have evaluated your pilot scheme, you take a decision as to whether to roll this out across the country. In this case, you seem almost immediately to have rolled it out without anybody having a chance to evaluate whether the pilot scheme was showing good effect or not. Why was that done?
(Mr Crisp) There was some evidence coming out of those pilot schemes. There were three originally. There was also a view taken that the early indications were good and that this was meeting an obvious need.
27. There was a view taken that the early indications were good? What evaluation was done?
(Mr Crisp) The first formal evaluation beyond those done internally was the one referred to in these papers.
(Mr Jenkins) We have had an ongoing programme of evaluation through the rolling out of NHS Direct and we got feedback from that at three stages during the lifetime.
28. That was an evaluation of the total effect or the pilot schemes?
(Mr Jenkins) It was looking at all aspects of the pilot schemes, issues around take-up, popularity, consumer satisfaction, impact on other services and clinical safety. We have a growing body of evidence through the course of rolling out the service that proves the concept and is giving valuable input into how we roll it out. The original White Paper gave a commitment to roll out NHS Direct. We have always been clear that evaluation of the service and piloting was about the way that it was rolled out and what its eventual scope could be rather than the concept. You can argue that that might have been different, but those were the rules of the game.
29. In effect, you are saying that this was yet another government scheme that was not properly piloted because the decision was taken before any evaluation was done that the whole thing would go ahead. That may be the right decision but it seems to be another case in which the government, having said it would do a pilot, in this case has not.
(Mr Jenkins) It depends what you want out of piloting. Piloting here has very strongly influenced the way in which the service has developed. The model of NHS Direct as it is currently organised has very much been influenced by a continual process of trying piloting, learning from that experience and modifying as we go along. The concept of having NHS Direct or not was not piloted because that was a decision ministers took at the outset.
30. Can I go on to paragraph 2.28 on page 16 where we learn that a number of people claim that they followed the advice they were given, 97 per cent, but quite a few did not follow the advice. What has made you say that they did not? What sort of advice are we talking about here? Were they advised to take some pills and they did not or were they advised that there was nothing much wrong with them and they continued to think there was? What difference are we talking about here?
(Mr Jenkins) That has been followed through in Sheffield's evaluation. There may be a difference in the time frame in which people take the advice. Maybe they are told to contact their doctor in four hours and they contact the doctor in 12 hours. Also, people's conditions will change. They are worried at a particular point. The advice was appropriate at that time. They wait a couple of hours and the problem goes away. My experience with the health pattern of my children shows that to be very common. In general, we find that at the point of patients presenting to the service there is a high level of appropriateness about the advice we gave.
31. Our witness was too polite to say so, Chairman, but, when you made your phone call last night and you faced that long delay, had you considered the possibility when you said it was for a parliamentary report that they were looking for a psychiatric nurse? The Report reads very favourably and I am very impressed with it. If I ask about some figures that may not necessarily sound too good, I hope I have got them wrong. If we look at table 5 on page 14, permitted average number of calls per full time nurse per month, if we look at the first column on the left hand side, there are nearly 900 calls per month per nurse. Which site is that?
(Mr Jenkins) East Midlands.
32. It looks a lot until you work out seven days a week, 24 hours a day and that works out at one call every 50 minutes; yet your target is about six and a half minutes per call. That figure suggests a degree of under used capacity but if we go to the other end of the table on the right hand side there you have what I work out as being 220 calls per month, which works out at seven calls per day. That does seem in both cases to imply that, despite the good figures, there is a tremendous under use of manpower capacity and, in the second case, very gross under use. Would you agree with that?
(Mr Crisp) We have to have a balance between the availability of nurses to answer the calls and the number of calls coming in. There will always be some down time within a system. Secondly, this is a very wide variation, much wider than we want to see. Since we have had the same systems operating across the whole country, we are able to make better comparisons and the extremes are narrowing, but there is still a lot of work to do to get the productivity up to the levels that we want.
33. I was interested in the left hand side and the right hand side, to see if we could explore any particular reasons. The right hand is one quarter of the workload of the left hand figure. One call every three hours 20 minutes would suggest that there is something wrong at that site. You have 22 sites and I think you gave an answer that there are 1,150 whole time equivalent nurses. That works out at five full time equivalents per site.
(Mr Jenkins) About 50 per site.
34. What sort of variation do you have in the total number of nurse full time equivalents available at each site?
(Mr Jenkins) It varies in full time equivalent terms between 30 at the smallest sites to over 100 at the busiest.
35. Has anyone been inspired to provide you with a piece of paper to show which the two sites were?
(Mr Jenkins) It is West London on the left hand side and Essex on the right hand side.
36. What do you think would account for such an incredibly wide variation?
(Mr Jenkins) In terms of working out the overall number of calls per nurse, the amount of available time on the phone is reduced by allowances for leave and sickness. Also, time for professional development and training. One of the reasons why there is so much variation in this data is that this data effectively predates our move to a single, national system. One of the key factors that we explored in the procurement for the system we have chosen was the average call length those systems generated. Because nurses are following some set protocols, the system does affect the total transaction times. The sites on the right hand side of the graph historically use one of the latest to move over to the new decision support system and conversely on the left hand side.
37. With the prediction being one every 55 minutes, why is it that you are not able to meet your target of replies within five minutes?
(Mr Jenkins) Those calls do not come in every 55 minutes. Probably there is one every 55 minutes in the dead of night but you still have to have two or three staff on to cover that for safety and clinical support reasons. At 11 o'clock on Sunday morning, we are probably talking about 70 or 80 calls.
38. Are there predictable peaks consistently across the country?
(Mr Jenkins) There are predictable peaks in general. There are within those very dramatic surges of demand within a period.
39. In the briefing we have from the NAO, we are told that NHS Direct is operating safely. They quote a figure of only 29 reported cases of an event during a call that caused physical or psychological injury to a patient, fewer than one for every 220,000 calls. At first sight, that sounds very impressive but when you look at the wording again it only applies to events that took place during a call. What evidence do we have of the safety of the advice that is given in what happens to the patient after the call? If someone drops dead two minutes later and had been told to take an Aspirin, that would not be included in these statistics, would it?
(Mr Jenkins) It would.