Official Report (Hansard)

PDF version of this report (193.25 kb)

Committee for Health, Social Services and Public Safety

Review of Waiting Times: Gooroo Ltd

The Chairperson: I welcome Dr Rob Findlay.  The normal procedure here is that you give a 10-minute presentation, and then we open it up to questions and answers.  You are very welcome, and we appreciate you taking the time to come before us today.

Dr Rob Findlay (Gooroo): Thank you, Chairperson.  It is a privilege to be here.  My introduction probably will not take as long as 10 minutes, but I would like to talk a little bit about one of the issues that I have drawn attention to in my briefing note.  I would like you to imagine for a moment that you are a manager in a hospital where you have a lot of long-waiting patients on your waiting list.  You would like to do the right thing for those long-waiting patients by treating them.  Now imagine that you have the capacity laid on and you are treating your long-waiting patients.  What happens to the proportion of the patients that you treat who are long waiters?  If you are treating your long waiters, the proportion of long waiters being treated will go up for a while, because you are treating lots of long waiters, and the number of long waiters left on your waiting list will go down.

So we have done the right thing by treating our long waiters, and the number of long waiters left on the waiting list has gone down.  It is very simple stuff.  What I want to explain now is how the English NHS got on the wrong side of that very simple concept and ended up in some difficulty with its waiting-time targets.  I think that you have a copy of a piece of paper that looks something like this?  Do you?

The Chairperson: Yes.  It is in the tabled papers.

Dr Findlay: This is an extract from the 'Guardian' website.  It says:

"Waiting lists grew 61% in the past year".

At the bottom of the page, I have put the statistics that this story came from.  You will see that the number of patients admitted during the month who had waited 26-plus weeks rose from 7,360 in June 2010 to 11,857 in June 2011.  Those of you who are good at your 61 times table will know that that is a 61·1% increase.

What does that 61% mean?  Does it mean that the waiting list got bigger by 61%?  No, it does not.  Does it even mean that the number of long waiters got bigger by 61%?  No, it does not.  It means that the number of long waiters being treated went up by 61%.  If you treat more long waiters, you would expect there to be fewer long waiters left on the waiting list, and that is, indeed, what you find.  If you look at the second row of numbers, you see that the number of patients still waiting who had waited over 26 weeks fell from 95,814 to 93,123, which is a decrease of 2·8%. 

So, waiting lists grew 61% in the past year?  Sounds terrible, but that is not what happened.  They treated their long waiters, so the number of long of waiters remaining on the waiting list went down, and long waiters got their operations.  So, it is a good thing that happened.  Interestingly, if you challenge, as I did, newspapers that print reports like that, you find that their response is almost invariably the same:  we are reporting the figures that the Government put out.  So, my word of caution to you is that if you in Northern Ireland were to implement targets in the same way that the English did and then follow those figures, you might end up with headlines like that even when actually things have got better, not worse. 

To go a little bit deeper into the story, you have a chart from University College London Hospitals (UCLH).  There are two charts.  At the outset, I should say that I do not in any way blame the hospital for what it did — for what I am going to explain to you that happened here — because this was during the era of targets and terror.  If you were an English hospital that was failing against the targets, the consequences were very severe, both financially and in terms of your personal career. 

What the top chart shows is the performance of this hospital trust against the headline target in England, which was that 90% of the patients being admitted  — being treated as an inpatient and a day case — should have waited less than 18 weeks.  On the left, in 2008, you can see the target being achieved for the first time as the line drops below 18 weeks.  The hospital successfully achieved the target every single month thereafter.  It kept its nose so clean that if you look at the first marker in 2011 — I have put a couple of markers on the timeline — you will see that UCLH did not appear in a list of 28 hospitals that were failing on the long-waits target.  A few months later, the Prime Minister, David Cameron, honoured the hospital with his presence, so that he could give a speech in which he pledged not to lose control of waiting times. 

If you look at the bottom chart, you will see what was happening underneath those figures on the waiting list itself.  The dotted line shows how long 92% of the patients on the list were waiting, and you can see that, in 2011, it was not far off a year.  The solid line that is plotted on the right-hand scale shows how many one-year waiters there were on the waiting list — who had waited more than a year since referral.  You can see that that peaked at around 1,700 one-year waiters in just one hospital trust just before the Prime Minister came down to give his speech. 

So, on the one hand, according to the headline target, we have a very rosy picture of short waits, but, on the other hand, if you look at the waiting list, you can see that there were patients waiting for a very, very long time indeed.  If you want the true picture of what is happening on waiting times, always look at the waiting list.  Do not simply rely on looking at the waiting times of those patients lucky enough to get treatment.

The final chart that I will draw your attention to shows England's overall figures for a similar period.  Just after the latest figures were released in June 2011, the deputy chief executive of the NHS in England wrote a very fierce letter to all the chief executives of every NHS organisation in which he said:

"it is unacceptable for performance to fall below the expected standards as it did in February and March 2011."

Look at the chart:  in which months did the NHS fall below an acceptable standard in England?  If you look at the dotted line, you can see what the position was like on the waiting list.  You can see that 90% of the waiting list was waiting up to 19 and a bit weeks at the peak around January 2011.  That was when there were a lot of long waiters on the waiting list.  In February and March, which are the months that the deputy chief executive drew attention to, the NHS treated its long-waiting patients and sorted the problem out.  What he was criticising was actually the NHS fixing the problem, not the months when the NHS had the problem in England.

I have devoted my few minutes to this point because it is something that has got the NHS in England in some difficulty in the past.  It is an apparently technical and arcane distinction, perhaps, between the waiting times of the patients being treated and the waiting times of the patients still on the list.  You might think that the one would be a reflection of the other.  What I have hoped to explain to you is that they are not necessarily a reflection of each other — and, worse than that, if you apply a target to the waiting times of those patients being treated, it distorts the behaviour of managers.  All targets distort behaviour — that is their purpose — but this distorts their behaviour in particular ways that are unhelpful. 

With that, I will conclude.  I look forward to your questions.

The Chairperson: Thank you, Dr Findlay, for that very useful overview.  First, this is something that the Committee takes very seriously, for very obvious reasons, in relation to the delivery of health.  We are researching this issue, taking evidence and talking to the Department and other experts in the field, and one of the lessons is that it is the complete journey time that needs to be examined.  Also, there is a sense, particularly coming from the professor from Trinity College, that to throw money or finances at one part of this particular issue may not, in his interpretation, resolve it, and can often be seen as rewarding bad practice in some ways.  I would welcome your views on that.  Specifically in relation to the referral-to-treatment (RTT) system that you talked about, we have a response in front of us from the Minister, and he states that it will be challenging both financially and logistically, and he is referring to the disparate reporting systems that exist across the trust.  I am very interested in how credible you think that is for preventing us from looking at the whole patient journey.  I am interested in your view on that.

Dr Findlay: Measuring the journey time from referral all the way through to treatment has obvious benefits over measuring each stage of treatment separately.  The principal benefit is that a typical surgical patient may take the following route through the system. Imagine that you go to your general practitioner with a sore knee.  Your GP may look at your knee and say, "I don't like the look of that.  I'd like you to go and see a consultant, please".  So, you are referred to the hospital.  At that point, the clock starts.  You might then have a wait, and then you will see the consultant for your first outpatient appointment.  On a stage-of-treatment basis, your clock then stops.  On an RTT basis, your clock carries on ticking, because what happens next is the crucial thing that RTT targets capture, but stage-of-treatment targets do not capture, which is that the consultant may say, "I don't like the look of your knee either, but I would like you to have a scan to check".  So, you go off and have a diagnostic.  You then need to see the consultant again after the scan has been done, so that the consultant can make a definitive decision about whether you need surgery, and you may need a further test and then to come back again after that.  That cycle of diagnostic and follow-up outpatient appointment can take a lot of time, and there is a possible perverse incentive, which I will come back to in a moment.  After the decision has been made to offer you surgery, you then have the wait for inpatient treatment, which, under a stage-of-treatment target regime, is captured as a waiting time in its own right, and then you get your operation.

The potential perverse incentive is this:  imagine that you are a manager in a hospital again, and you have a very large waiting list for surgery.  You cannot keep up with the demand for surgery.  You want to try to slow the number of patients who are arriving on your inpatient waiting list.  I am not saying that this is happening in Northern Ireland — I do not know, because I have not seen the data — but there is the potential to delay patients at this diagnostic and follow-up outpatient stage where they are not tracked by a stage-of-treatment target.  That is a potential difficulty with the stage-of-treatment regime.  With RTT, the clock carries on ticking all the way up to the point at which they get their definitive treatment.  So, there is a clear advantage for RTT.

However, it is correct to say that RTT is more difficult to measure.  There are two main kinds of RTT target.  One measures how long the patient waited, but it only captures that measurement at the point where the patient gets treated.  As I have just explained, that can get you in hot water, if that is what you monitor.  The better thing to do is measure referral-to-treatment waiting times while the patient is on the waiting list.  That is more difficult.  If you are measuring RTT waiting times at the point where the patient is treated, all you have to do is say, "You have been treated, so let's look back.  When was your referral date?  It was back then, so you have waited this long".  That is the figure that you report as the waiting time..  If you are trying to capture all of the patients who are still waiting, you have to measure their waiting time since referral, at every stage in their journey:  when they are waiting for outpatients, when they are waiting for diagnostics, when they are waiting for follow-up outpatients and when they are waiting for surgery.  That might be two, three, four or five different IT systems, and it might include some paper-based waiting list management systems in the hospital, and you need to capture the data for every one of those.  That is a big challenge.  So, if you propose to implement referral-to-treatment waiting time targets — which, I think, would be a good idea — you should be advised that it is likely to take some years before you can implement them with good coverage in practice, and that during the transition period, it would be a good idea to retain the stage-of-treatment waiting time monitoring and targeting while the new system phases in.  Both systems should be based on the patient still waiting, and not focused primarily on the waiting time of patients as they come in, as previously explained.

You raised one further point, Chairperson, about rewarding failure, throwing money at the places where there are problems, and not resolving the problem.  There are ways around that.  That question falls into several parts.  The key to the issue about rewarding failure is to understand that waiting times are a function of two things.  First, how long is the queue?  Secondly, in what order are the patients treated?  Think of your everyday experience.  If you go to a supermarket, it matters that the queuing system is fair so that you have a free choice of which queue to join.  You can estimate which queue you will get to the front of quicker.  If you go into the post office where there is a single queue, everybody gets seen on a first-come-first-served basis, which is fair.  If you have a lot of queue-jumpers pushing in ahead of you, you will wait longer.  Queue-jumping pushes waiting times up and can be unfair.

In hospitals, you have patients with urgent clinical conditions such as cancer or aneurysms that need repair.  You have a number of patients whose clinical need is greater than the clinical need of others.  Those patients absolutely must be treated ahead of the others.  They need to get treatment for their clinical condition quickly enough.  Incidentally, while we are talking about waiting tines, the single most important thing that any hospital can do is to make sure that its urgent patients are always treated quickly.  They are jumping the queue.  That pushes up the waiting times for others, but that is OK because that is a good reason to jump the queue.

To the extent that the long waiting times are caused by inappropriate waiting-list management, which is mostly pulling patients out of order when there is no good reason to, you do not want to reward that kind of behaviour by throwing money at it.  However, that is easy enough to detect.  If, on the other hand, the problem is caused by the waiting list being too big, there is a fair argument that money is the appropriate solution.  The hospital has a queue that is very big, and the way to get rid of a big queue is to treat the patients.

There would then be the argument, "If we throw money at a long queue, it will draw in demand and the queue will never go away."  If you eliminated the queues at one hospital in isolation, I agree, it could draw in demand from  the surrounding areas and you would apparently have achieved nothing at that hospital.  However, if you did it across the whole of Northern Ireland, you would not necessarily see that.  When the referral-to-treatment waiting-time targets were achieved in England, GP referrals stepped up significantly; I estimate it at about 22%.  However, that is not a tsunami of demand.  It is not losing control of demand.  It is not opening the floodgates.  It is 22%.  Bringing down waiting times may increase demand — some of that demand may have arisen anyway — but it is a counsel of despair to say that there is no point in trying to shorten waiting times because we will only end up back where we started.  I do not think that the evidence supports that.

The Chairperson: That advice about looking at the waiting list as opposed to the patient waiting time is extremely useful and important.  You have probably answered my question around what you view as the reluctance in the health service to look at RTT.  You have explained that it is a complex and difficult process.  Are you aware of other regions that have moved to RTT?  Can you give an indication of the cost?  We are being told that one of the challenges is financial constraints.

Dr Findlay: I cannot answer the question about cost, I am afraid.

England is quite a long way down the road with this.  From memory, I think that it was 2004 that they announced it, and the target was first met in 2008.  Initially, it was measured only on the basis of the retrospective wait when a patient got treated.  It now also captures all the patients who are still waiting.  Even in England, where they have been doing it for some years, they are still not capturing the whole waiting list.  If you look at the data by specialty and by hospital, you still see hospitals discovering waiting lists that they had not previously been reporting, and you get a step up in the number waiting; that still happens. 

Scotland has been doing it for a few years less.  If you want to seek evidence on this, Scotland would be a good place for you to seek a witness. I understand that you have somebody from Scotland coming here next week.  They might be able to help you to find an information analyst who is very close to the issues and can assist you.

The Chairperson: Yes, we have looked at the Scottish model through the Assembly Research processes.  That is very useful.

Mr Beggs: You said that it would be easy to detect hospitals that tend to produce better figures by moving onto other waiting lists, massaging the figures etc?  How would that be easy to detect?

Dr Findlay: Our company has done quite a lot of research in this area to understand the dynamics of waiting-time management.  We have looked at the arrival rate of patients on the waiting list and how you can book patients in so that urgent patients are always seen first.  It is about following the three principles outlined in the first paragraph of my submission:  "safe, fair and short."  I would add one more to that:  "efficient".  The principles are that urgent patients must always be treated within their clinically safe limit.  "Fair" means that routine patients should, broadly speaking, be treated on a first-come-first-served basis.  I say "broadly speaking" because there will always be exceptions.  "Short" means that no patient should wait an unreasonably long time.  "Efficient" means that the hospital should be able to utilise its capacity fully in doing all of this.

We have established some very simple rules that hospitals can follow patient by patient to keep the waiting list continuously optimised.  By studying that in simulation, we have established the link between the size of the waiting list and the times that patients wait.  If you were to say to me, "Here are the figures for the number of patients who arrive on the waiting list at the given rate, this is the number of patients on the waiting list, this is the proportion who are urgent, this is the proportion who are removed without being treated", and the hospital is managing the list on either a fully or partially booked basis, I could tell you, for example, that, "In that case, you would expect 90% of the patients on the waiting list to be waiting less than x weeks if the list were managed according to those rules."  In practice, hospitals tend not to manage it in that optimal way, so you would allow a little bit of a tolerance.  However, beyond that there is scope for improving waiting times simply by managing patients in a better order.

We can do that visually as well.  This will look slightly odd in the Hansard record, but a normal waiting list looks a bit like an elephant.  You have the head dropping at the start where the urgent patients are coming in.  The elephant has a flat back where no patients are coming in because they are waiting their turn,  Then the elephant's rear drops off very steeply at the back where the patients are coming in, broadly on a first-come-first-served basis.

A real NHS waiting list tends to look like a dinosaur's tail.  You have quite a lot of patients at the front; then the waiting list dribbles on and on and on and on up to 20, 30 or 40 weeks.  It is not at all uncommon to see on a waiting list urgent patients who have waited much longer than they should have.  It is not uncommon to find long-waiting patients who do not even have an appointment for surgery, while, at the same time, very short-waiting patients who arrived on the list only two or three weeks ago and who have already got dates for their operations in the next week or two.  That is very common.

It is easy to detect when a waiting list is not being managed according to those well-accepted principles and to work out what it would look like if it were and, therefore, whether a backlog clearance is needed to bring waiting times down to a target level.

Mr Beggs: You said that some people can be on the list for a very long time.  I have come across constituents who have waited perhaps six months for exploratory surgery with one consultant and then been advised, "Sorry, it is not this problem.  It is actually a different problem."  They are told that the six-month clock starts again in the wait to see the second consultant.  Having to wait a year, or perhaps even longer, before you see the appropriate consultant seems excessive.  What is the case in England and Scotland?

Dr Findlay: Without knowing the specifics of the case, that sounds a little bit like a stage-of-treatment issue as opposed to a referral-to-treatment issue. Your constituent has waited a certain amount of time to see one consultant.  The moment they walked in the door to see the consultant, the clock stopped.  If they are then referred to another consultant, a new clock starts for that second wait.  In England — I think also in Scotland, although I am not 100% sure — if they were waiting to see the second consultant for the same condition, the original clock would still be ticking.  Therefore, with regard to the measurement against the 18-week target, the clock would still be ticking all the way through.  That is a cause of anxiety for tertiary hospitals in England.  If you have great big teaching hospitals with super specialists, they often have patients referred to them who already have 16 weeks on the clock, and it is very difficult for them to treat a patient within 18 weeks.

Mr Wells: This is a bit uncanny.  As we speak, my daughter was wheeled into an operating theatre five minutes ago for knee surgery, for which she has waited more than a year.  It was the result of a skiing accident, and although it was inconvenient and slightly painful, had the operation occurred in another year's time, there would probably be no difference in the outcome.  Has any empirical work been done to establish whether delays led to worse outcomes or is it just inconvenience to the patient?  Has any work been done to show that the longer you wait for routine procedures, the worse the outcome may be five years down the line?

Dr Findlay: That is a little bit outside my field of expertise, so I am not sure whether I can give you a definitive answer.  However, you would probably find that it would matter in some cases but not in others.  If some conditions deteriorate, the operation might save your life, particularly if it were an aggressive cancer, an aortic aneurysm or heart surgery.  As I said, that is a bit outside my field of expertise.

Mr Wells: Am I right in thinking that the vast majority of these are things like hips and knees.  I can think of dozens of constituents who have contacted me complaining about the 17-week wait.  Indeed, they have been put on the list for a knee replacement, but, at the end of the day, although it is very painful it is not fatal.  Is that the norm when we are talking about waiting lists, or are we dealing with a lot more complicated issues than that, if it is non-urgent?

Dr Findlay: Your question is more about rationing than waiting.  If the patient and their doctor have agreed that the patient should have an operation and if the NHS has promised the operation, the NHS should deliver on its promise.  If the NHS does not intend to operate on a patient because it considers the treatment to be of low clinical value, that should be clear to the patient at the outset.  That has been done in England, particularly for things such as tonsillectomies and varicose veins — I think, from memory — where there are procedures of low clinical effectiveness and fewer patients are being offered surgery for it.  You may be right in that some patients may not come to harm or very limited harm by waiting a very long time.  However, my view is that the NHS has made a promise and it should deliver.

Mr Wells: Have any trusts in GB come close to eliminating or dramatically reducing, and can that be done without throwing vast amounts of money at the problem?

Dr Findlay: What — to eliminating the waiting lists?

Mr Wells: To get it down to a month or something ridiculously short.  Is anybody coming close to that target?

Dr Findlay: Yes, there are some areas.  The curious thing about waiting lists is that some areas have very large pressures while others have very little.  This brings us into the area of planning.  The NHS is relatively good at big set-piece plans:  planning for the next financial year, doing a business case for a consultant or capital development — the big adjustments to capacity.  If you were the captain of a ship, that would be charting your course.  My impression is that what happens a bit less well is the constant adjustments on the tiller that keep you on that course and the continually adjusting capacity to keep up with demand.  That is something that the NHS is not so good at — planning on a time horizon of six weeks to a few months into the future.  What happens is that the capacity that the NHS is laying on drifts out of line with the demand for healthcare, so you end up with some areas with very big waiting lists and long waiting times and others where people can go home in the middle of the afternoon because there are not that many patients to treat.

This is an issue of planning, evening out the pressures and of the internal allocation of existing resource to the areas where it is most needed to relieve the pressure on the hardest-pressed areas by recycling resources from areas under less pressure.  There are areas that are under less pressure, and there is scope to do that.  It is not done as well as it could be, certainly in my experience in England and Scotland.

Mr Wells: You are saying that with proper planning you can greatly reduce the waiting list without throwing vast amounts of extra funding at them.

Dr Findlay: You can even out the pressures between different services.  Ideally, planning should be done at the level of inflexibility in the system.  Take orthopaedics, for instance.  You have consultant orthopaedic surgeons, but they are mostly sub-specialised now:  some do hips and knees or possibly only hips.  In a very big hospital, you may have surgeons who do only hip revisions.  Then you will have shoulder surgeons, spinal surgeons, hand surgeons, foot and ankle surgeons, to the extent that they do only one kind of procedure.  That is the level of inflexibility.

In orthopaedics in England, the pressures often tend to be in spinal surgery.  If you look at the orthopaedic waiting list in a big teaching hospital, the patients waiting out to a year are very often spinal patients.  It is difficult for the hospital to find the capacity in the system to treat them all — some could be 14 hours on the operating table.  So there is a particular sub-specialty where, over a long period, the planning has not made sure that capacity is aligned with demand.  At the same time, hand surgery may have quite short waiting times in some hospitals.

Mr Wells: Are there good examples of trusts on the mainland that have cracked this?  This is a big political issue in Northern Ireland; it is probably the big headline.  It is the waiting time at A&Es and the waiting time for routine operations.  To some extent, the efficiency of a health trust is measured by those two figures.  In my opinion, to some extent at the expense of, perhaps, quality of care, we have to get those figures down.  Are there examples in GB where proper planning rather than huge amounts of additional money have produced such results?

Dr Findlay: Not that I am aware.  It is a problem almost everywhere.  Waiting list initiatives are often done, so you do not have an early warning of pressure but actual pressure.  You have actual long-waiters.  Hospitals have difficulty booking in patients within a reasonable time, so they know that they are going to have trouble with their waiting times.  Therefore they lay on a waiting list initiative, negotiate extra capacity, put non-recurring money into it and clear the problem for the month.

Guess what, a couple of months later they have the problem again because it is not an issue of a backlog that needs clearing.  The issue is that supply is not keeping up with demand.  It is their baseline that is wrong.  They do not need extra now just to clear a bit of list.  The problem is that their list is growing.  They are not keeping up.  Even now, you will find examples in English hospitals of the plans for next year not being based on demand but on hospitals' ability to supply.

Let me give you a more sophisticated example.  A hospital is doing its plans for next year.  It is an inclusive process, involving all the stakeholders, all the general managers and clinicians in the hospital.  It is a devolved process and you say to the managers, "What is your plan for next year?"  They may go through a sum that goes something like this, "Well, I have this many consultants, they have this many sessions a week, and they do this many patients per session.  Multiple those three numbers, and you get the number of patients that we can treat every week.  Multiple that by 42 weeks of the year or however many weeks the consultants work and that is the number of patients that we will treat in a year.  That is our plan."  Where is the demand for healthcare in that calculation?  Nowhere.  It is a plan based purely on the hospital's ability to deliver.  I know that, in Northern Ireland, you have an exercise that is about planning capacity starting from demand, and that is the right thing to do.  That happens in most places in England.  However, it is by no means universal, and there are a significant number of hospitals in England that base their plans on their capacity to deliver and not on starting from the demand for healthcare.

Mr Wells: I was hoping that you were going to point us in the direction of south Essex or north Yorkshire so that we could go over and get the information and come back and solve our problem.

Dr Findlay: Most hospitals do long-range planning based on demand.  Many hospitals tweak their capacity at short range — the next two, three, four or five weeks — to respond to specific problems, but the crucial range is six weeks to a few months.  It is outside the annual leave notice period when you do not know the names of most of the patients and capacity is still reasonably flexible.  There needs to be good week-by-week planning in that area.  I do not know of any examples of hospitals that are really good at that.

Mr Wells: We had a brief chat about this over lunch, and it is a point that I made when Professor Normand was here recently.  It never ceases to amaze me when they build a new primary school in south Down, children just appear from nowhere.  It is extraordinary.  When you add the sum total of all the old schools that it replaced, suddenly, there are another 50 children, and we do not know where they came from. 

Similarly, if you build a new road, the traffic will increase to populate it very quickly, and, suddenly, you have as much of a problem with tail backs as ever.  Is there evidence that there is a latent demand that is not in the statistics, which will become evident if a trust is performing particularly well?  In other words, people who elude their GPs, or even themselves, by thinking that there is no sense in putting their name down for a routine operation because they will have to wait for ages.  Does it become a self-perpetuating problem in that more people come forward because the rumour has got out that you get quick treatment?

Dr Findlay: You would need to ask a health economist, but I will make two observations.  First, as I understand it, it is well known that you stimulate demand if you appoint a new consultant, particularly if it is a consultant who has a special interest that is not currently served in the local population.  In that case, you get people popping up who have that condition wanting treatment. 

My other observation is the new motorway that quickly fills up with cars; however, at least people are getting to their destinations more quickly.  Therefore it is not necessarily a bad thing.

Mr Wells: The demand on the service is still as intense if new people are coming forward to replace the more efficient model that you have.  You never get to the bottom of the pile because people are coming into the area for treatment because the rumour has got out that there is a particularly efficient model.

Dr Findlay: Yes.  I will go back to my earlier remarks about the experience in England.  In 2007, hundreds of thousands of patients had been on a waiting list for more than a year.  The 92nd percentile waiting times were out to nearly a year.  Now, there are only a few hundred patients on the waiting list for more than a year in England, and 92% of the waiting list is well within 18 weeks.  There has been an enormous reduction in waiting times in England across the whole country.  There has been a step up in GB referrals in that period — 22%, as I said earlier — but it is not out of control.  It rose and then it broadly stopped rising.  It was a step.  I think that a health economist would be a better person to ask, but I am not convinced that there is good evidence of the rationing impact of waiting times.

Mr Beggs: You talked about hospitals needing to react to demand in their plan.  Are there good examples of hospitals adapting quickly to that demand?  My perception is that generally what happens is that, in the end-of-year monitoring round, money comes and it is just fired off to the private sector for it to deal with rather than the hospitals and consultants putting in additional capacity and overtime for staff.

Dr Findlay: I recognise that scenario from England and Scotland, but particularly England, where the use of the private sector is more widespread.  Most hospitals, certainly in England, could do a better job of planning their baseline capacity in line with demand over all timescales, particularly the six key weeks to a few months into the future timescale, which is when they need to be responding to seasonality and changes.

The use of the private sector is widespread in England, although it is not necessarily an easy option for hospitals.  They are losing the income by paying for treatment in the private sector.  It may cost more or less, depending on the deal that they have struck.  I do not have a view on the merits of public versus private, but I will make one observation about the potential conflict of interests that it puts before the consultant.  If the same consultant is operating on a patient in the private sector as would have operated on a patient in the public sector, and the consultant is being paid for that, there is a potential conflict of interest for that consultant that could push them in the direction of maintaining long waiting times.

How do you solve that, if you are using the private sector and it is the same consultants?  One option is to try to make sure that it is not the same consultant.  It can be difficult to persuade patients to change consultants if they have already met them in clinic.  Most, but not all, patients typically say, "I would like to see the nice doctor I saw in clinic, please".  Therefore the ideal time to move them is before the first outpatient appointment.  Many more patients will agree to move at that stage than after an outpatient's appointment.

Mr McKinney: I had one point, but I now have two, because of the point that Roy made.  Does the use of the private sector to get rid of the problem quickly encourage bad practice?

Dr Findlay: What sort of bad practice are you thinking about?

Mr McKinney: Does it encourage bad practice in meaningfully dealing with the problem.  You are dealing with something that has erupted, or you are factoring in something to try to eradicate the problem, and you are not dealing with the problem as a whole.

Dr Findlay: Yes, I see what you mean.  I suppose that it could, but I think that it is another manifestation of not dealing with the underlying problem.  As I said earlier, a waiting time problem falls into two parts:  the waiting list is too big, or patients are being managed in the wrong order — or both.  In many hospitals, they do not know which of the two is the problem.  The analysis exists now to tell the difference.  Relying on short-term solutions, whether by bringing in locum and bank staff, exporting patients to the private sector, or by using some other short-term waiting list initiative, is part of the same approach of trying to fix a problem that we have this week or this month rather than realigning baseline capacity to fix the longer-term shortfall of capacity against demand, which, incidentally, is usually cheaper.  It is usually cheaper for the NHS to adjust its baseline capacity than it is to pay, usually, premium rates for locum or other short-term work.

Mr McKinney: Thank you for that; that is very helpful.  I was struck by your original slide, and I was thinking of lies, damned lies and statistics.  In essence, that was an interpretation by the paper of what you point to as otherwise good work, because there was a 2·8% drop in the overall figure.  Jim was not able to drag you to a particular place in respect of expertise or an example of good practice, but what are the drivers for political and management change, and what are the ingredients for change?  Can you sum that up for me?

Dr Findlay: Yes.  My favourite Aesop's fable is the one about the sun and the wind trying to get the fellow's overcoat off.  The wind blows and blows, and the harder it blows, the tighter the guy holds his coat around himself.  Then the sun comes out, and he says, "Ah, that's lovely", and he takes his coat off.  If you want something done, people have to want to do it.  How do you make them want to do it?  In England, they did targets and terror.  It achieved the result; there is no denying that.  When the coalition Government came in, they experimented with not having centrally enforced targets, and they were promptly rewarded with the failure that I showed in my third chart.  If they did the same, I am sure that they would be rewarded in the same way with another failure.  The targets-and-terror approach worked.

I am not a great fan of targets per se.  There is a risk that the target can be literally something that you try to hit as accurately as possible, but it should be a backstop, a minimum standard, something that the NHS usually comfortably exceeds so that it is rare and unnecessary to enforce it.

How do you reach that happy state of affairs?  If you were to have a target as a minimum standard, you may wish to set it not as a stretch goal for the service to get to but as something that the service should comfortably be able to achieve.  Then you should focus on getting well below that target by concentrating on the two underlying factors of waiting times — the size of the waiting list and the order in which patients are treated.

Put very bluntly, as I said in paragraph 12 of my submission, when focusing on waiting times it is easy to forget the waiting list.  That happened widely in England; people forgot to watch the waiting list.  If the waiting list is getting bigger, you have a problem; if the waiting list is going away, your problem is going away.  If you want to reach the happy position where there is no involuntary waiting in the NHS, getting rid of the waiting list is the way to do it.

Mr McKinney: You talked about people having to want to do it and you said that targets and terror worked even though it was not necessarily nice.  In our context, what is the driver?  Is it ministerial or departmental will?  At what point in the chain of command is the driver located?

Dr Findlay: I am afraid that I cannot answer that.  That is not my field of expertise; I am not a manager or a politician.  You had better ask someone else about that.

The Chairperson: Thank you for that.  It has been very informative, Dr Findlay.  Thank you for taking the time to talk to us.  There is a lesson in this, because I am reflecting on the fact that we have called this review a review of waiting times.  One of the things that we have learned today is that we should not lose sight of the waiting list.  There is a recognition from your evidence that this is complex but that processes such as referral to treatment are productive and can deliver if they are done alongside the stage-to-patient journey.  That has been very useful, and the discussion about targets is specifically relevant to us.  In your paper you say that even if there is a target it should not be a 100% target and that we should be realistic about people, obstacles and changes, the patient journey, cancellations and holidays.  That has been useful.

We really appreciate your evidence.  We are actively working our way through this process, and it would be very useful to share our recommendations with you and to hear your feedback.

Dr Findlay: I would be delighted to do that.

The Chairperson: Thank you very much for sharing that with us.