Official Report (Hansard)

PDF version of this report (191.74 kb)

Committee for Health, Social Services and Public Safety

Review of Waiting Times: Professor Luigi Siciliani, University of York

The Chairperson: I welcome Professor Luigi Siciliani, who is professor of economics at the department of economics and related studies at the University of York.  The procedure is that we will ask you to make a 10-minute presentation, and I will then invite questions from members.

Professor Luigi Siciliani (University of York): Thank you very much, Chairperson.  It is a pleasure to be here.  In my 10 minutes, I will try to give you an overview of maximum waiting time policies across the Organisation for Economic Co-operation and Development (OECD) countries.  Then, time permitting, I will discuss some waiting time measurement issues.  Most of this draws from a review of the policies that I co-edited for the OECD.  I will give you the headlines and a brief summary of the 300-page report. 

Across the OECD countries, the most common policy is to have a maximum waiting time.  You see convergence across several countries in the way that their policies develop by their specifying a maximum.  The way that the maximum is specified and used in different countries can be quite different, and that is where I want to identify three different groups.  First, the maximum waiting time is used as a target.  This was used in England and Finland, for example.  Under that, if providers do not perform within the maximum time and have patients waiting beyond that target time, they receive penalties.  That is one way to use the maximum waiting time. 

Other countries do not like that approach and prefer to mix it with incentives that are more related to choice and competition so that patients who wait more than the maximum time are entitled to seek treatment in other public hospitals or, maybe, in a private hospital at the expense of the public system.  That is a quite different way to use the maximum waiting times, and this approach has been experimented in Denmark, the Netherlands and Portugal. 

The last way to use the maximum waiting time is quite different and is a tool of prioritisation.  It is not so much about penalties and targets; maximum waiting time is used simply as an indication of the severity of a patient's condition.  Countries such as New Zealand, Canada and, perhaps, Norway have gone down this route, so their key focus is prioritisation. 

Generally, there is a trend to differentiate maximum waiting times, so you may have general maximum waiting times for all elective care but some shorter ones for more urgent care such as cancer and heart conditions.  I have observed that across a range of countries. 

My key policy message from the review is that we know that, generally, supply policies are not necessarily successful in reducing waiting times because you increase the supply and, possibly, the demand goes up by the same amount, so nothing happens with waiting times.  I have seen that a key ingredient of success is to combine increasing supply with a maximum waiting time, which results in an increase in the supply but keeps the demand under control.  Having a maximum waiting time means that referrals cannot go up so quickly.  In a few countries, this has led to quite substantial reductions in waiting times. 

For the rest of my few minutes, I will go through some examples.  I will go through the English examples quite quickly because you are probably quite familiar with what has happened there.  England started with quite high waiting times in 2000:  more than 200 days for a few elective procedures.  As targets were introduced and more resources were put into the system, waiting times went down quite rapidly from more than 200 days to 50 days for some elective procedures such as hip replacements.  One key issue with maximum waiting times is the potential for misprioritisation.  Quite often, doctors make the point that one potential problem with a maximum waiting time is that it really applies to low-severity patients.  In a system using prioritisation, the patients with lower severity tend to be treated later.  So, potentially, you could have the maximum waiting time being applied to low-severity patients at the cost of higher severity patients waiting relatively longer.  Maybe there is some evidence that this behaviour is happening.

Finland had a similar experience, so it introduced maximum waiting time targets.  In 2000, it had very long waiting times, similar to the English ones, and it also managed to reduce them.  Again, this was achieved through a combination of extra resources and a healthcare guarantee that specified a maximum waiting time for patients. 

I will move on now to the second type of policies, which involve more choice.  That has been experimented for a few years in Demark, where it was said that, if a patient had waited for longer than a certain time, they could go to, for example, a private hospital.  That seems to have contributed to some reductions in waiting times.

One policy that I really want to mention is that of the Netherlands.  Between 2000 and 2010, there have been quite some dramatic reductions in waiting times.  In 2000, it had, on average, 15-week waiting times for a few elective procedures.  In 2010, this was only five weeks, so waiting times went down quite a lot.  How did the Netherlands do that?  On one hand, it introduced some maximum waiting times, although its preferred terminology was "socially acceptable waiting time", which specified, for example, a maximum of seven weeks for inpatient treatment, and that was combined with another rule that 80% should be treated within five weeks.  This is another alternative way to specify the maximum.  You set a maximum waiting time for everyone, but you say that a proportion of the patients have to be treated within a shorter timescale.  In doing so, you tackle, to some extent, the issue of misprioritisation, and the focus on low-severity patients is lessened. 

In the Netherlands, that was not the only reason for the reduction in waiting times.  Introducing a maximum waiting time would not have been as successful if the Government had not at the same time relaxed the capacity constraints.  In 2000, hospitals changed from a fixed budget to activity-based payments.  Specialist remuneration also changed from lump-sum payments to a fee-for-service scheme.  All of that contributed to increasing the volume of activity.

These days, policymakers are no longer concerned about waiting times.  Their main concern is the sharp rise in health expenditure.  Sometimes, there are big trade-offs to be made.  You can solve some problems — for example, reducing waiting times — but that can come at the cost of significantly higher expenditure.

The third type of policy is the use of waiting times as a prioritisation tool.  The New Zealand experience is quite interesting.   There, patients are divided into three groups.  Patients can be booked in straightaway; be certain that they will be treated within six months; or put into active care and review, which means that they are sent back to their family doctor to be checked.  The idea is that some patients on the low-severity spectrum should not be put on the waiting list but sent back to their family doctor for review.  That is more like a demand management policy. 

The New Zealand Government tried to facilitate this policy through the development of prioritisation tools.  To be practical, I will give you an example of how these tools look in practice.  For hip replacements, a form asks what the patient's level of pain is when in motion and when at rest. Each level has a score attached to it, which gives a severity index.  There are some formalised guidelines and tools that could facilitate prioritisation, and they are linked to the formulation of the maximum waiting time.

The last policy that I want to mention is the one in Norway, which uses waiting times as a prioritisation tool in the sense that every patient is entitled to an individualised maximum waiting time.  That is quite different from the maximum waiting time policy that we normally observe in England or Finland.  In Norway, a specialist sees a patient and, based on severity, decides on their maximum waiting time.  One patient could be told six months, the next patient three months, so it is an individualised maximum waiting time.

Those policies solve the potential problem of prioritisation that comes from unconditional maximum waiting times, but, on the other hand, perhaps they introduce too much discretion into the system.  The early evidence from the evaluation of the Norwegian policy is that it did not change dramatically waiting times in any way.

I have given you an overview of the policies on maximum waiting times and tried to distinguish between the three main types:  use maximum waiting times as a target; combine them with choice and, perhaps, some involvement of the private sector; or use them as a tool for prioritisation.  I have some measurement issues to discuss, but I will pause in case there are questions.

The Chairperson: Thank you very much.  While conducting this review, we have been continuously told that we need to look at the complete patient journey and not just a part of it.  Is that your sense of the issue?  Should we be looking at the entire patient journey from the time of referral for treatment?

Professor Siciliani: Based on the current evidence, that is a good idea in general.  Several countries are moving towards the idea that you should follow the patient journey.  One reason is so that you have a complete measure of waiting times, but also so that, generally, we can understand the patient-care path better if we do that.  So it seems a good idea to me.  Some countries have moved in the direction of measuring the referral-to-treatment time (RTT), but it is still early days.  England and Denmark use referral-to-treatment time, but we do not have too many countries from which to draw experience and judge whether it has worked well. 

From the English experience, I observe that some complications arise from using the referral-to-treatment measure.  You need to report at least three different types of measurement.  Normally, in England, you have the admitted pathway and the non-admitted pathway, and the number of patients on the waiting list and the number who have been treated.  That leads us on to the measurement issue.  There are two main ways to measure waiting times:  take a snapshot of the number of patients on the list and check how long they have been waiting; or check the total number of patients treated in, say, a given month, and find out how long they waited.  Those are really two different measures, and there are benefits and disadvantages to using each.  Possibly, it would be a good idea to have both.  When you put that idea in the context of measuring referral-to-treatment times, the snapshot measure does not tell you whether the patient will be admitted because patients on the list will be a mix of the two distributions.  However, with the number of patients treated, you know whether they received hospital treatment, or maybe some just saw a specialist and did not need surgery.  So you can see that the idea of measuring referral-to-treatment time is interesting because it covers the patient journey.  However, it brings in some extra complications, which mean that, in practice, you may need at least three different measures of waiting times.

The Chairperson: Is it your view that there needs to be a separation of waiting lists and waiting times?

Professor Siciliani: There are different orders of measurement.  The most basic distinction is between waiting time and waiting list.  The waiting list is simply the number waiting.  I think that, possibly, policymakers have moved away from just counting the number of patients on the list.  Perhaps, what you are suggesting — maybe I have misunderstood — is that we measure the waiting times of the patients on the list.  There are benefits from and disadvantages to measuring that.  If you measure the waiting times of the patients on the list, you have a reasonably accurate and responsive measure of what is going on in the system at that time:  they are the patients on the list at that time, so you can check the distribution and the average.  It could be useful for monitoring purposes.  On the other hand, it measures an incomplete waiting time because, by definition, the patients on the list are still waiting, so you do not have the full measurement of how long the journey has taken.  That is where the other measure is really useful:  if you have the average waiting time of patients treated within a given month, you know that they have completed the journey.  That is a more accurate reflection of the patient journey, but it can be made only with a delay.  As the patients have now been treated, your measurement is actually of the numbers of patients treated in the past, so you do not have a very accurate measure of what is going on with patients at that point in time.  In that sense, there is scope for measuring both because each captures a slightly different aspect.

The Chairperson: You have given us some detail of a decrease, or positive impact, on waiting lists and waiting times.  Has there been any analysis of the impact of that decrease on health outcomes or even economic outcomes?

Professor Siciliani: Some research has tried to test whether patients who wait longer have worse health outcomes.  In my assessment, that literature is not huge and is still a little preliminary.  I am aware of some studies, mainly from Canada, that tried to test whether longer waiting times affected the health outcomes of patients waiting for a coronary bypass.  The study seems to suggest that, while they were waiting, patients were more likely to die while waiting or maybe had a higher chance of having an emergency readmission.  So, for coronary bypass, there seems to be evidence that goes in that direction.

There is other evidence that relates more closely to orthopaedics departments and, for example, hip replacements.  That evidence seems slightly more optimistic, in the sense that longer waiting times do not affect how much pain the patient experiences.  Whether the patient waits six months or three, the condition does not get worse with the passing of time.  That is my reading of the evidence, and those are the two primary examples that I have been able to identify.

Mr McCarthy: Thank you very much for your presentation.  Are we omitting accident and emergency?

Professor Siciliani: So far, I have omitted accident and emergency.  The primary focus of my review was elective surgery, so it is really about the long waiting times for patients.  From my reading of the literature, I think that the issues that you have with waiting times in accident and emergency are quite distinct from those for elective surgery.  So all that I have said applies to patients waiting for elective surgery.  You are right.

Mr McCarthy: We could keep you here all day talking about accident and emergency, but you have not gone down that road. What are the most important aspects of treatment times that we should seek to measure in order to get an accurate view of performance?

Professor Siciliani: The referral-to-treatment waiting time seems to me to be a good measure, in the sense that it captures the full journey.  I suggest capturing the three distributions.  Once you have those, I do not see why you should go somewhere else or try to seek collection of other data.  One thing to keep in mind is how you want to collect the data:  ideally, it would be nice to collect it from existing sources so that you do not have to start a new data collection.  My understanding is that England had a separate new data collection for RTT.

Mr McCarthy: Is there a critical mass of countries measuring referral-to-treatment times?

Professor Siciliani: At the moment, the only two countries that use RTT are Denmark and England.  Other countries are not measuring referral-to-treatment times.  That is, I think, because it is still early days, and, although people like the idea of measuring the patient journey, the other two waiting time measures are traditional.  Many countries have chosen either one or the other, so they all focus either on inpatient waiting times or the patients treated.  So there is lots of heterogeneity and, hopefully, over time, we will converge on the measure of time from referral to treatment.

Mr Wells: The stats are very interesting, and the UK does not come out very well in your analysis.  I suspect that, within the UK, Northern Ireland is probably even less impressive.  You are citing data from countries such as Norway, the Netherlands and Denmark, which are all very wealthy countries that allocate a huge proportion of their income to social welfare and health treatment.  Your English is extraordinarily good, but have you ever heard this question:  are we comparing apples with oranges?  In Denmark, for example, where tax rates start at 40%, if you throw enough money at the problem, presumably you have enough money for a good programme to address waiting times.

Professor Siciliani: I see your point.  However, you might think that in the Nordic countries, where public budgets are quite high, there should be no waiting times at all.  So it is perhaps surprising that even countries such as Norway or Denmark have significant waiting times.  Some of those countries have been successful, but even countries such as the Netherlands had a significant problem.  It is true that they put in quite a lot of extra resources, but now even the policymakers are saying that health expenditure is rising too quickly, and it could be that they will scale it back.  I think that many countries are struggling with the waiting time problem.  It is not all about income and how rich the country is.

Mr Wells: I was intrigued by your comment that, in some countries, if you cannot get a hip replacement or whatever done within the appropriate target time, you have a right to go to a private provider, get the work done and bill the health service.  That nugget of information was in the middle of your presentation.  Will you give us a little bit more detail about it?  That is quite revolutionary.

Professor Siciliani: That is the Danish experience, and perhaps I should qualify that policy a little bit better.  A patient, after the maximum waiting time, is entitled either to treatment in another public hospital, maybe in a different region, or through the private sector.  In practice, the number of patients who choose to do that is not massive, so, in a way, although the policy states that that is the case, in practice, there were some increases in private sector involvement, but they were not massive. 

If you think also about the English experience, where the independent sector was brought in, that sector can do 5% of extra elective procedures like hip replacements.  Private sector involvement is one ingredient in this policy, but I would not say that it has the critical mass to be the contributing factor that would reduce waiting times, even in Denmark or the Netherlands.

Mr Wells: The model here is different.  The patient does not have the choice.  If the health trust finds that the wait is too long, it can commission the private sector to carry out a number of procedures.  Indeed, in Northern Ireland, we allocate procedures to hospitals in the Irish Republic, so people will go down to Dublin or Athlone and get treatment.  What is so radically different about what you describe is that, once a certain date passes, the patient has a right to knock the door and say, "I demand that this be done."  Does that then concentrate the minds of the health trusts or equivalent in the sense that, presumably, they do not want to pick up that bill and, as it is more expensive, they try to stay within their limits?

Professor Siciliani: Yes, that was the intention of the policy.  The Government hoped that this would have a disciplining effect so that the initiating hospitals or local authorities would try to stay within the maximum waiting times in order that this would not happen so much.  I think that Portugal followed a similar idea and route, although it is not listed in my presentation.  You will have another presentation next month.  Again, in practice, that works maybe as a threat, but there are still not a huge number of patients who exercise that choice. 

There is a combination of factors.  There is always a choice of how to structure the involvement of the private sector:  do you just try to involve the private sector and ask whether it can do this extra elective care, or could you use the same resources for the public hospitals and ask the doctors working there to do that extra volume of work?  There are some different policy options here that, maybe, in the bigger picture, could be equivalent in the sense that, if you have some extra resources, you can give them to the private sector; the private sector might recruit doctors working in the public hospitals; or you could have public hospitals with public doctors and ask them to do extra sessions and tell them that you will put some extra resources on the table.

Mr Wells: Is there evidence that having a high-quality service for procedures and short waiting times generates more work?  In other words, if people perceive that there will be a short time to wait, as in, for instance, in the Netherlands, which is an extremely good performer, does that just cause more work because GPs and patients are more likely to demand that type of surgery rather than making do?

Professor Siciliani: That tension is always there with waiting times.  When you expand the supply, your primary concern is that demand will go up and then you start to do marginal care.  That is where the maximum waiting times policy comes in:  you say that the maximum has to stay within the five weeks, three weeks or whatever period has been set by the regulator, and that keeps the focus on keeping the referrals more limited.  Otherwise, demand would go up, and you would not be able to have a short waiting time.  In that sense, it is, I think, a combination of supply and maximum waiting times that help to keep the demand under control.

Mr Wells: An obvious question is why Italy does not feature in any of this research.

Professor Siciliani: Italy was covered in the review.  However, it has 20 regions, and there is quite a lot of heterogeneity in it.  That made Italy quite difficult to cover in my 10 minutes before the Committee.  It is there, but we would need a long digression.

Mr D McIlveen: Thank you for your presentation.  From the point of view of the work that we do on a day-to-day basis in our constituencies, to the person who is waiting for a hip replacement, knee replacement or whatever it should happen to be, this is hugely important.  I think that this piece of work is really important.  A lot of the media focus is on accident and emergency, but it would be wrong to ignore the suffering that people quietly endure on a day-to-day basis, as their name creeps slowly up the waiting list.  You have given a very helpful overview of what different regions are doing.  Can I encourage you to come off the fence a little?  I say that with the greatest of respect.  Can you give us an indication as to what, in your experience, works best?

Professor Siciliani: I risk repeating myself a little.  The combination of extra supply with maximum waiting times works best to make sure that waiting times for hip and knee replacements are brought down to acceptable levels.  Many governments try to create those extra resources, but how do we make sure that that does not just translate into extra referrals and waiting times do not change?  We need that combination.  Having read reviews and evidence from all these countries, I made that my main conclusion:  we must combine the two.  In the past, most countries would act either on one side or the other.  They would increase supply and hope that waiting times would go down, but that would not happen; or they would decide that patients should not have to wait longer than 30 days without receiving an explanation, but that would not work either.

So, to me, the key is understanding that waiting times are a demand-and-supply phenomenon.  You can create extra supply in different ways, either by increasing capacity in the public sector or by involving the private sector.  Those are the alternatives.  However, the key is demand.  We must make sure that demand for referrals does not respond so quickly to reductions in waiting times.  We hear stories that, when waiting times reduce, doctors tend to make more referrals.  That is the balance that has to be struck.  To me, that is the key.

Mr D McIlveen: Perhaps I will reword the question, and put it in a different way.  If the Chairperson were to ask you where she should book a flight to in order to explore best practice, where would you suggest?

Professor Siciliani: I would suggest the Netherlands, because that is where I saw this happening.  There, they did not like the phrase "maximum waiting times".  They brought doctors round the same table and asked what was the reasonable maximum time that patients should wait on medical grounds.  That seemed like a good starting point.  Once that had been settled, they asked what resources were needed to bring supply to that level and to make sure that it stayed there.  Therefore, I would say that the Netherlands was the best example.

Mr D McIlveen: In your research, did you find that there were any obvious societal differences between what is happening in the Netherlands and what happens in the UK?  As my colleague said, we can pour money into things, but, if there is no change in society, culture or how the public thinks, it can result in throwing good money after bad.  Is there a cultural difference between the attitude to healthcare in the Netherlands and that in the United Kingdom?

Professor Siciliani: In the Nordic countries, and perhaps in the Netherlands as well, you do things more by consensus and almost by unanimity.  Everyone is brought round the table to make sure that everyone agrees on the plan.  That is not necessarily always the case in other countries.  It may be more top-down, and they say "These are the targets; get on with it".  That is the major cultural difference.

Mr Beggs: Thank you for your presentation.  You talked about the considerable improvement in New Zealand, and you can see quite a dramatic drop in its waiting times over a one-year period.  Your explanation for that is that a mixture of issues has brought about best results in certain areas, including an increase in service delivery and a part review of the waiting lists.  Can you tell me a bit more about the review of waiting lists?  It is easy to remove someone who has already been treated; that is non-controversial.  However, you also indicate that many patients were informed that they would not be assessed.  How do they get on the waiting lists if they do not need to be assessed?  I am trying to understand how people would be removed without some sort of clinical assessment.

Professor Siciliani: It is a reasonably common phenomenon that I have observed in various countries that, when the waiting list problem suddenly becomes more of a real policy concern, the first thing that policymakers try to do is make sure that the data are of good quality, and that is when the reviewing of the data happens.  I have an old paper that contains some very similar pictures for England and its early initiatives on waiting times, and, basically, the waiting list data were not very good in the sense that, on that list, there were patients who were on other lists in other hospitals, or perhaps some patients had already received treatment.  Therefore, in a way, patients who were not supposed to be on those waiting lists were part of the review of waiting lists.  That was inflating waiting lists for no reason because those patients were not supposed to be there.  I have seen in a few countries that, when you try to make a jump in the quality of the data, in the first review of the data, the waiting lists will suddenly drop and will then stabilise, since, once you have taken out all the patients who were not supposed to be on the waiting lists, they are gone.  Once the quality of the data becomes acceptable, those patients are not added to the waiting list.  You have that one-off effect.  The drop is only partly due to that.  However, other things were going on at the same time, including a bit of an increase in supply and perhaps some patients having sent back a negative carer review.  That was just one of the ingredients.

Mr Beggs: I am trying to see for what other reasons people would be brought off their waiting list and returned to the GP.  I recently came across a constituent who had been referred from his GP to a consultant for a hip replacement.  Part of the process in Northern Ireland is to go through an Integrated Clinical Assessment and Treatment Service (ICATS) centre.  He waited for a few months to go there.  Having passed that and it having been identified that he did need a hip, he was referred to a consultant for whom there was six-month waiting list.  Once he sees the consultant, there will be another six-month waiting list for the operation.  You said that, in New Zealand, people wait six months for treatment.  Is the definition of "six months waiting for treatment" the same everywhere?  Six months does not really mean anything to my constituent:  he has been waiting more than a year.

Professor Siciliani: The New Zealand policy was six months to see a specialist and six months to get surgery.  I do not know whether that answers your question.  I mentioned initially the active carer and review group, which is the most interesting thing for the New Zealand experience.  When the policy was initiated and introduced, the government decided on a clinical threshold over which patients can benefit from surgery and an economic threshold, and those were the terms used in the policy documents.  Ideally, all those patients could benefit.  However, there were perhaps 200 patients but money for only 100 to be treated; that is where the active carer and review group came in.  It was the gap between what could be funded and afforded by the public system and the patients who could be treated.  In a way, it is the rationalisation of demand.  Rather than add patients to the waiting list and hope that, at some point, they will be treated, the idea was to make demand management more explicit so that some of the patients who would never be treated — even if they were on the waiting list — could be sent back to the family doctor for active care and review.  That was the idea.  I am fascinated by the New Zealand policy, because it looks quite different from any other policy that I have seen in any other country.

Mr Beggs: Have they significantly moved the point on the threshold at which you would get treated?

Professor Siciliani: I think that the idea has evolved over time.  I do not think that the threshold has been operationalised in a very precise way.  The answer to your question is that the threshold is still informal.  It is there, but it is not precisely measured.

Mr Beggs: You say that it is informal.  Does it just qualify whether the money is there to allow you to be operated on and then people are prioritised based on that and, therefore, if you are not a priority, you are not on the list?  Is it as simple as that?

Professor Siciliani: The family doctor and the specialist have to work together.  The idea is to prioritise patients early on so that you can rank the severity.  You use that information to decide the patients who are entitled to public treatment and those who are not.

Mr Beggs: So it might not be a good area to look at.

Mr McKinney: I have two consequence questions.  Are there financial implications in maximising supply to deal with waiting times?  Does your proposal for the best model cost more or less or is there evidence to show that it is within budget?

Professor Siciliani: The policies that have been successful, in the sense that waiting times have gone down, were all accompanied by a significant increase in resources.  They are not policies that come free; they probably implied significant extra expenditure.  Most of the policies were in the early 2000s and onwards, before the recession.  In those times, it was a little easier to provide extra resources.  Now that we are in recession, it will perhaps be more difficult to implement similar policies, because the budget is more constrained for everyone.  Replicating that idea, starting at a higher level of waiting times, will be more difficult.  You can still try to experiment and introduce maximum waiting times with not so much of an expectation for an increase in supply.  However, it will be harder to obtain a reduction in waiting times.

Mr McKinney: Over time, has that cost stayed up?

Professor Siciliani: The way to think about waiting times is as a dynamic phenomenon.  In past policies, one way of reducing the waiting list was to introduce a one-off increase in supply, which would make it disappear.  However, that does not work.  You need a steady increase; otherwise you will reduce the backlog for a period, but it will tend to come back.  The policies are long term with constant increases in expenditure.  If you have an increase in supply, it should be a permanent and not a temporary increase.

Mr McKinney: Do the proposals have any positive or negative impact on other waiting lists associated with the health service?

Professor Siciliani: Normally, the more recent policies try to affect all elective care.  In that sense, they should cover quite a lot of the healthcare setting.  That seems to be a good way of doing it; otherwise you would be concerned that, if you focus on one subset, other areas may suffer.  Here, the concern may be about elective versus accident and emergency in the sense that you focus on elective procedures at the possible cost to accident and emergency.  I am not aware of evidence to support that, so I cannot comment specifically on it.

Mr McKinney: I wonder whether savings are to be had or whether additional costs would be imposed elsewhere as a result of doing this.

Professor Siciliani: I am not aware of any evidence as result of trying to look at how extra resources being provided to, or a focus being put on, maximum waiting times could have diverted incentives from other areas.

Mr Dunne: I think that most of the points have been covered.  Is there any evidence that the use of the private sector goes a long way to improving the waiting times?

Professor Siciliani: No, that is just one complementary policy among others.  Normally, the private sector is used in conjunction with other policies that reinforce it.  So far, from what I have seen, there are no policies that introduce the private sector only, so it is difficult to disentangle what is due to the private sector compared with the other policies.  My overall impression is that the private sector was an ingredient and was one of the determinants that helped the success of these policies, but it was not, perhaps, the number one.  The focus on the maximum waiting time was more important.  The increase in supply could be obtained through the private sector but could also be obtained in other ways.  Perhaps increasing the sessions and changing the contract for specialists and doctors who work in the public sector could be another way, as could building new public hospitals.  Those are alternative policies, and I do not necessarily see that the policy of using the private sector in itself is the one that dominates.

Mr Dunne: What about the internal processes in the existing services?  Is that an issue that is worthwhile addressing?

Professor Siciliani: You are referring to the dual practice issue.

Mr Dunne: Yes.

Professor Siciliani: That is an important issue that needs to be addressed, because people in health policy know that the dual role of doctors generates a conflict of interests.  Especially on waiting lists, you know that, if the waiting list is longer, the patient will be more willing to go private if they can afford it.  In that sense, there is a conflict of interests.

Mr Dunne: That is a real risk.

Professor Siciliani: It is a risk.  First, you should think about why you want dual practice to exist.  Is it because doctors are more willing to work in the public sector?  Perhaps it brings other benefits.  Do you really need it?  Perhaps you could change that element and say that doctors work for the public sector only.  Those are all policy options.  There is a conflict of interest there, and that has to be taken into account when deciding policy.  I do not think that there are any easy fixes.

Mr Dunne: There are no easy answers.

Professor Siciliani: No.

Mr Brady: Thank you for the presentation.  In your paper, you give an example from western Canada of prioritisation guidelines on hip and knee replacement.  That seems to deal with the person as opposed to the system, because it asks whether the pain on motion is none/mild, moderate or severe, whether pain at rest is none, mild, moderate or severe and asks the patient to assess their ability to walk.  Therefore it seems to be looking at how the person is affected as opposed to putting them into a system where they become just a statistic.  Is that how they look at it in western Canada, because, if a person is in severe pain, they will be prioritised, presumably because their mobility and quality of life is affected?  That seems to be patient-centred rather than looking at numbers and statistics.

Professor Siciliani: Let me say a bit more about the Canadian policies, which resemble those of New Zealand.  We know that prioritisation goes on informally and that doctors are pretty good at it.  The idea behind the policies was to develop a tool that would help doctors to do something that they already knew how to do a little.  It is correct that that system puts the patient at the centre.  It was quite costly to develop those guidelines, because, basically, you had to bring together doctors, patients and members of the public. The guidelines are interesting tools.  They help the prioritisation but, at the same time, if you are thinking large scale, they are potentially quite costly to develop, in the sense that we have one for hip replacement and three or four others, but they still cover a small subset of care.  So, these tools can help but there is a danger in the sense that they are costly, so I do not necessarily see them as the solution to the whole problem.  Ideally, if this were to be costless, then —

Mr Brady: The point is that it is the solution for the patient.  If you do not have a system that is patient-centred, it seems to me that you do not have a proper system in place.  Ultimately, if you are the person suffering severe pain because you need a hip or knee replacement, you are not interested in somebody else's waiting list; you are worried about yourself and how you may be prioritised and dealt with.  It may be costly, but at least it is dealing with that particular person's problem.  That could probably be multiplied a thousand times over, so it seems sensible.  I accept that it is a tool, but it is a sensible way of dealing with an increasing problem.

Professor Siciliani: I will complement what you said by referring to the Dutch experience and the idea about determining the maximum waiting times by simply looking at it from a regulatory perspective rather than having maximum waiting times that are acceptable to the patient and society, which seems a good way to go.

Mr Brady: When you talk to someone who is in severe and chronic pain and is waiting for a hip replacement, and then talk to them again afterwards, you know that it has really enhanced their quality of life.  That is particularly true of older people who may not have many years left.  To increase and enhance their quality of life seems to be a good thing.

The Chairperson: OK, professor, thank you very much for that.  What I take from this is that you suggest that there is not one quick fix but it is a combination of a supply issue generally, which I expect also takes in issues such as GP contracts, workforce management and the maximum waiting times target.  I suppose, for us, the issue is around enforcement of that as well.  As we conclude, it might be good to get your thoughts on the best models of enforcement.  We can have all the extra supply or have maximum waiting times without proper models of enforcement.

Professor Siciliani: There have been key experiences around enforcement.  To have targets for maximum waiting times and then to attach heavy penalties, such as people losing jobs, seems extreme.  That is at one end of the spectrum and is excessive.  However, you do need enforcement.  The current policy in England in which part of the revenues are retained if the target is missed seems to be a compromise.  You need enforcement otherwise changes are not going to happen, but what I outlined is not as radical as the extreme fines that had been put in place.  So, attaching some serious financial incentives to the maximum waiting times seems to be a reasonable compromise and balance.

The Chairperson: Thank you for that, and thank you for taking the time today.  Your evidence has been informative.  Obviously, the Committee and wider community feel strongly about this ongoing work.  I take this opportunity to suggest to you that, as we develop our draft report on waiting times, we feed it back to you for your thoughts.

Professor Siciliani: It would be a pleasure.

The Chairperson: That is great.  We will take it that we can forward that to you and you can feed in your thoughts.  So, thank you for your time and have a safe journey home.

Professor Siciliani: Thank you.  It was a pleasure to be here.