Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
Community Affairs References Committee
Health services and medical professionals in rural areas

HUMPHREYS, Professor John, Centre of Research Excellence in Rural and Remote Primary Health Care


CHAIR: Welcome. I understand information on parliamentary privilege and the protection of witnesses and evidence has been provided to you.

Prof. Humphreys : Yes, indeed.

CHAIR: We have your submission, No. 32. I invite you to make an opening statement and then we will ask you some questions.

Prof. Humphreys : Thank you for the opportunity to speak with you today in relation to your inquiry. Today I speak on behalf of the Centre of Research Excellence in Rural and Remote Primary Health Care. It is a collaboration of staff from three universities—from Monash University's School of Rural Health in Bendigo, the Centre for Remote Health in Alice Springs and the Department of Rural Health at the University of Sydney. So it is a multi-site collaboration. My colleagues Professors Wakerman, Lyle and Perkins and Dr McGrail are not able to be here today, and nor should I because I should be in Adelaide, where we are running a rural research capacity workshop. So they are all involved in other activities, but they would wish to be here and I speak on behalf of the centre.

I am aware that by now the committee will have heard many people talk about this complex problem of how to ensure an adequate and equitable supply of health services and medical professionals in rural areas, and so I do not intend to go through a lot of that stuff. What is clear is that, despite well-intentioned government policies over the last two decades, there is still a shortage of doctors in rural and remote Australia. Any recent improvements largely reflect the increasing number of international medical graduates who, in effect, have limited choice in where to work. Access to health care remains problematic in many small rural and remote communities and to date there is little quantitative evidence of the effectiveness of workforce incentives in redressing the situation.

Over the past 25 years I have undertaken research in a number of areas: in the nature of general practice, the requirements of viable rural practice, international medical graduates, the role of incentives and the impact of classification on resource allocation. The document I have given you today demonstrates 25 years of research, and I have tried to summarise it for you in one page. The references on which it is based are on the back there. You may disagree with the assumptions on which I operate, but I doubt you will disagree on the evidence because the points I have made there emanate from a significant amount of research that has been undertaken over a long period of time. The recommendations and conclusions that I draw from this are consistent with those in the submission. So I do not intend to go through that for you. What I would like to do, though, is to indicate that I think it is very clear that the shortage of doctors in rural areas will only be resolved when they are appropriately remunerated and supported in the work they undertake.

Under the existing ASGC scheme for targeting workforce incentives there are clearly problems. The existing schema is not equitable and, I would argue, is not effective. This is particularly because of the inherent heterogeneity in the ASGC categories 2 and 3. ASGC is the Australian Standard Geographical Classification, and I will refer to it as such. Currently we have a situation where doctors who are practising in large, well-supported communities, in environmentally attractive areas, in resource rich areas—places such as Coffs Harbour, for instance—are eligible for the same types of incentives as those who work in small inland, remote communities. As you will be aware, this is clearly inequitable. It is also an ineffective use of resources.

Neither the tweaking of the boundaries that occurred with the recent review of the ASGC undertaken by GISCA nor the introduction of the new mesh block, which was outlined to you, I believe, by the Australian Bureau of Statistics in their presentation, will address the fundamental problem. The fundamental problem is that dependence on geographical criteria alone does not adequately reflect the issues that are responsible for the difficulties associated with recruitment and retention of doctors into rural areas. Our proposal deals with that.

What we have proposed to you in our submission and I think in a follow-up article which I sent to the secretariat and which I hope you have had access to is a fit-for-purpose classification as the basis for distributing resources. This classification is based on the nature of medical practice across rural and remote areas of Australia and the context in which they operate. In other words, our classification is based on what doctors do and the environment in which they do it. What we have tried to do is to make sure that doctors undertaking similar activities in similar settings receive appropriate incentives and support for their practice and that they are differentiated from those practising in contrasting environments.

Our classification, unlike the ASGC or the ABS type classification, is based on six sentinel factors that impact upon the recruitment and retention of doctors. Four of these are professional factors and two of them are non-professional factors. And the choice of these six sentinel factors is based on the evidence that has been derived over many studies in Australia over the last 20 years, and those references are documented on the page I have given you. Specifically, we take account of the total hours that doctors work, the after-hours on call they do, whether they do hospital activity or not, and the difficulty in taking time off. These are critical factors that impact upon doctors' desire to go and practise in a rural environment and the length of their stay there. The two non-professional factors are the employment opportunities for partners and the local education facilities.

What we have been able to do using these as the basis of our classification is to show that population size used in conjunction with the geographical location provides a better surrogate measure. What we did effectively was to correlate those factors with the existing ASGC classification and, yes, there is an association. When we also correlated it, though, against the town's size and population, we get a much more significant correlation—in other words, it is a much finer, more sensitive discriminator. The only area in which we needed to use location was in those small communities where, by the addition of location, geographical isolation, remoteness, if you like, we start to get greater sensitivity. What this does in effect is separate out those small communities of, say, 500 people which might exist in the shadow or the catchment area of a metropolitan area—they are in quite desirable areas but they are in the commuting belt—from those small communities of 500 or so which are really quite remote and isolated.

That is the basis for the classification we have put forward. Effectively, we tested how well ASGC differentiates and we found in fact that population size categories that we derived from this were much more able to differentiate the nature of practice. We chose in our paper six categories, I believe, and that is because governments do not like to review the world as it is. There are 1,500-odd small communities. We do not want 1,500 policies; we want a number of policies that group like with like. Our classification chooses the optimal number of groups and it minimises within-group variance and maximises the between-group variance. In other words, it groups like with like and separates unlike from unlike, which is really what the program wants to do.

Our classification has been widely publicised and I think it has been readily accepted by doctors, by and large. As you will be aware, with any classification scheme there will always be some anomalies. But, by and large, we think that our scheme works in between 95 and 99 per cent of the cases, and we have had a validation from the rural doctor community. Unfortunately, despite our best efforts, to date the Department of Health and Ageing have been somewhat reluctant to engage in this issue. I believe that this committee probed the Department of Health and Ageing, as I read in the submission, and also asked some pertinent questions of the Bureau of Statistics.

Our centre of research excellence, which is funded for four years and has another 2½ years to go, continues to undertake research, such as this, on rural workforce issues relating to the provision of primary health care. Importantly, we have developed a framework for knowledge exchange. So we are not just doing research in a blue sky arena. We want to translate our research into the policy environment and into the practice environment so that policies and practices are evidence based. We are optimistic that this opportunity to talk with you today will illustrate one important way in which this has happened. Thank you for reading our submission and for listening to me today. I am happy to answer questions and to elaborate on any points that you wish to raise.

Senator NASH: Thank you very much. I am very impressed that you managed to get 25 years into one page. That was a very good overview. Why do you think the department is less than inclined to engage with you on the proposed classification system? Reading through the submission, it seems very sensible and it seems that it would work. Why do you think there is a reluctance on the part of the department to engage in at least considering this?

Prof. Humphreys : I think there are a number of reasons. I have been raising this with the department and staff there for some years. I think governments are often risk averse. They have a number of existing arrangements—for example, they depend heavily on the Australian Bureau of Statistics, which does an excellent job. I think in their presentation to you the Bureau of Statistics were wise and outlined the limitations of what they deal with, which is, effectively, a geographical classification. Unfortunately, it is not a fit-for-purpose classification for this instance. There is no one generic classification for all. Classifications are always for a purpose. We were able to draw on the data from the 'Medicine in Australia: balancing employment and life (MABEL)' study that we are undertaking to get appropriate data and match it with the geographical data.

Sometimes one could be a conspiracy theorist and argue that they already have contractual arrangements with organisations that do work for them. I believe that the review of the ASGC undertaken by GISCA was enormously disappointing. This notion of tweaking around the edges with anomalies is really not the way to get a systemic solution to a problem that has been around for a long time. We have had a number of classifications over time, some of which have needed to be updated, and it is important to review new developments. I think we have good relations with the Department of Health and Ageing, as we do with many professional organisations. That is our business. But probably you are closer to the scene of action than I am in terms of how the world works. I will say that if I were given retention grants there are occasions when I would give them to the Department of Health and Ageing because every time we go there there is a new person who is dealing with the problem. The turnover in that area is often problematic.

Senator NASH: Yes, there is not a lot of corporate knowledge. When we were asking the department about this, it was put to us that this needed more time to bed in and that it had not been going very long, so they could not get a very good handle on it at this point. But the principles of it actually seem to be wrong. Can you see in any way, shape or form why giving it more time to identify whether or not it would work could be successful?

Prof. Humphreys : I am a lover of Yes, Minister. I always loved Sir Humphrey's five reasons why things will not work. I think the principles underpinning this are very clear and they are not going to change. What might change is access to better data. If we could incorporate Medicare data we may get better representation along the way.

Let me explain: the original data for this is a study where we have annual national surveys of doctors. We get about 10,500 to 12,000 responses each year, and that is the basis. It is not all doctors, but when we have done an analysis against the population of doctors it is a very good sample. It is very representative along the way. If we were able to have access to Medicare data, for example, you might see how robust our suggestion is, but the principles that underpin it do not change. That is, you really want to take account of what doctors do in their practices and the context in which they do it: how much they are supported, what the nature of the community is and what the demands of the community are. That should be the essence of how you reward and remunerate.

There has been a long history of this. It goes back to the relativity study that was done in the nineties. There have been a number of good proposals over my lifetime to the Department of Health and Ageing, which have, unfortunately, been very selective in the way they take these things up. I suppose what I am saying is if you are going to spend scarce resources you need to stop this nonsense of having pilots thrown out here and there and get it used systemically.

Some of our research, for example, on rural workforce retention has shown that you can actually save money by doing things differently. If I can exemplify the point, we did a study where we looked at retention patterns of allied health professionals, nurses and doctors and we showed that if you put certain incentives in place at certain times, and kept that person there for a period, you would save money. You might spend $10,000 at a certain point of time by increasing the length of stay of a professional by 12 months. Where that savings comes from is if that person leaves the recruitment cost is twice the amount you are giving in retention grants. Premised on the basis that that service is an essential service for that community and you will therefore want to replace it, you are expending money.

We have shown how things can be done without necessarily costing more money. That is not to say that more money is not welcome; it is to say that we have to appropriately target the way we use the money. We need good recruitment policies and good retention packages as a strategy—not necessarily just throwing money at a person, whether it is a nurse, an allied health professional or a doctor, but as a package that takes account of their continuing professional education requirements, the infrastructure requirements and the family requirements. It might even be in the form of flights to support a spouse to go and visit children in a boarding school in a metropolitan location.

That package, fit for purpose and adjusted to context, can make a difference. Our work shows, for example, that housing is much more important in some of the isolated remote areas than another bundle of dollars there. If you look at the evidence there is plenty there about which you can do things. There is a reluctance to have flexibility in programs.

Senator NASH: I was just about to use that exact word: flexibility. It seems from everything that you are saying that the need for flexibility and not just one-size-fits-all is really important.

Prof. Humphreys : Indeed.

Senator NASH: In terms of remuneration, there has been a lot of discussion around raising the profile and raising the prestige of GPs out in the regions, and part of that may well be, bluntly, to pay them more. Have you done any research on how much it would cost government to increase remuneration across the board to these GPs working out in the bush?

Prof. Humphreys : No. What we have done, though, in our study of viable rural practices, is an assessment of about 150 practices across the country, looking at the income structure of those practices in different environments. We used this as the basis for some settings and benchmarks around things. What we showed there, effectively, was the need to provide an adequate primary care base for those communities. Something like 20 per cent were going to go out backwards because they did not have sufficient income. They were bulk-billing their communities because the communities are poor communities that cannot afford co-payments. At the end of the day those practices will go out backwards or general practitioners will leave because they are not getting a sufficient amount in the practice environment. I think the second thing is it is not about an absolute amount of dollars for a doctor; it is about remuneration for what they do. That is the important thing. If you are on call, and are called out twice a night for the week you are on call, or if you are doing a one in two arrangement, you are severely restricted in your quality of life, the work-life balance and the impact on the family. Sometimes that financial incentive or remuneration can compensate by allowing them to go to the coast or to the city or whatever they want to do.

I think we have to be realistic about the rural and remote general practice environment. People do not go there forever. They go there for a period of time. What we really want to put in place is a mechanism where practices are remunerated in such a way that it remains an attractive environment. If the existing doctor goes because their kids reach high school—or it is time or they have had enough—it is still an attractive place for other people to come to. The community then benefit in the assurance of quality primary care along the way. I do not think we should think about doctors being there forever, particularly with the modern graduates—the gen X or Y people—who really do want this work-life balance.

Senator NASH: I apologise if this was not your submission but one of the submissions did have just what you are saying there—easy entry, graceful exit.

Prof. Humphreys : That is right. That is the work of Ian Cameron from the New South Wales Rural Doctors Network. We have worked closely with them.

Senator NASH: I thought that was a very good way of putting it.

Prof. Humphreys : It is.

Senator NASH: Just finally, on this issue of the non-financial incentives, you talk about housing and improved working conditions. One of the things I am concerned about that is happening at the moment is that local government is providing housing and that social requirement and it is not really the responsibility of local government. How would you see the provision of things like housing happen under what you are putting forward?

Prof. Humphreys : I think that is a really good point. There are two factors here. I am a little disappointed that one of the key issues that the health reform agenda let drop was this ability to bring funding arrangements between the Commonwealth and states together much more closely. I think that would have been a wonderful thing to happen, but states are what they are and want state of origin matches and one thing and another. It is a problem. There is that issue obviously.

The second issue is the silo mentality of government departments. I would like to see regional development, housing and employment talking a lot more about how they can deal with these things. As the author of the original National Rural Health Strategy 1993, one of our recommendations then was that the minister for health should talk to the minister for education to start this dialogue. I do not think that has changed. It is unfortunate sometimes that local governments have to resort to this activity. They want to and they do it with good intention but sometimes it separates out things in such a way that it does not create the systematic solution we are trying to get. Regardless of the community—whether it is Queensland, Victoria, New South Wales or wherever—every doctor requires an adequate housing structure, adequate schooling and adequate employment for spouse. It is really regional development and primary health care coming together.

We have to stop thinking about health as being an exercise in consumption of dollars and more as an investment. It is an investment in the community, an investment in quality of life, an investment in productivity. Whoever invests in that community provides a legacy that the community can still keep building on when the doctor goes. I am an idealist. I live in the ivory tower and the world looks good from up there.

Senator McKENZIE: I want to flesh out a point you made in your submission that the current classification system penalises small and more densely populated states such as Victoria.

Prof. Humphreys : I am not sure whether you are aware that under the change from the rural and remote metropolitan areas classification, RRAMA, which was the original one, to the Australian Standard Geographical Classification, the ASGC, with the Accessibility/Remoteness Index of Australia, ARIA, in between, states like Victoria have a greater geographical area subsumed under one category. Whereas the old RRAMA scheme would have representatives of all bar one of the seven classifications, under ASGC that does not happen. You get communities like Bendigo and Shepparton grouped with small communities like Rushworth and Elmore, which have very specific needs. In one case they do not have a GP; in another case they have to operate in a different sort of model. But they get the same incentives, if you like, and fall in the same category. Bendigo is a city of 100,000 and Shepparton has 50,000, and they have many more environments. That is effectively what happens. As you move to the big states like WA or Queensland, then it plays out a little bit more, but you still get that coastal strip. You get the Proserpines coming in with the Mackays, the Beaudeserts with the Moretons, and so it goes on. That is what we are talking about there. Obviously, it is a scale thing and our classification is designed to look at a national scale. You could refine it down with a state like Victoria or Tasmania if you wanted to.

Senator McKENZIE: When I was reading the mapping you had done with all the variables playing into why people make the decisions they do—

Prof. Humphreys : Was this the black-and-white graph or the coloured graph? The black-and-white is messy because it was originally in colour. I do have a colour copy I can leave.

Senator McKENZIE: That would be great. My question went to the smaller centres. As you get to under—

Prof. Humphreys : Five thousand?

Senator McKENZIE: As you decrease down that scale, are they typically located in a geographic area or are they dispersed randomly?

Prof. Humphreys : I suppose originally they were like corner stores: they serve a purpose at a local level, but over time many of those centres lost their raison d'etre. Their economic base changed; they lost functions like railway functions if they were a railway town; their commercial structure changed as companies vertically integrated and they no longer had a role. Many of these small towns are now the legacy of a historical antecedent. To that extent they are widely spread. You do not get lots of new little towns developing. Instead you get the revitalisation of those towns that have some environmental amenity or, in the case of mining, some new resource is discovered. You get the stagnation of those that still struggle with their economic base and are largely inhabited by an ageing population with young people going elsewhere.

Senator McKENZIE: I probably did not phrase my question appropriately. Are there certain geographic regions where there is a greater density of towns under 5,000 than other geographic regions?

Prof. Humphreys : Inevitably, yes. There are closer settled areas, if you like. In Gippsland—

Senator McKENZIE: I was thinking of Gippsland specifically.

Prof. Humphreys : What we do in our work is always to try to select densely populated areas, so Dr McGrail operates in the Gippsland region; we are in the Mallee region, which is much more widely dispersed with communities up to 100 kilometres apart, but they are still in that 700 to 1,000 population.

Senator McKENZIE: Does that impact?

Prof. Humphreys : It means that the model of service provision inevitably can vary. In densely populated areas you can often run a hub-and-spoke type arrangement. You can congregate your providers and have them rotating out, if you want to, much more easily than you can in widely dispersed areas, where that geographical impediment of travel means you want to maximise the in situ presence of providers.

Senator McKENZIE: Does your classification system allow for that?

Prof. Humphreys : In terms of a model for resource allocation, it does. In terms of the nature of the model, further work is required. You will see in some of our references there, we have used different sorts of models—hub-and-spoke models, visiting models, discrete models, integrated models. One of the critical things I would stress here is that our research is showing that where you have a good model of provision—whether an integrated community health centre or a general practice—where you have got that right, it becomes an attractive place for doctors to be recruited to and sustain their contact. Inevitably where the problem lies is where the practice environment does not provide professional satisfaction and the town has limitations for family. They look at this compared with other places and say, 'Well, I'm not sure. I'll go to the coast, rather than go inland.'

Senator McKENZIE: I have a couple of examples in Victoria where that is exactly the case. I have just one more question. You make several comments around 'bang for buck', essentially. I am just wondering if you know how much we as a nation spend on rurally targeted programs. You can take that on notice.

Prof. Humphreys : That is a really good question.

Senator McKENZIE: The second part of that is about how we go about evaluating it. Do we have a standard view of the world?

Prof. Humphreys : There are two points, then. The last time I looked at one of the government audits that were done of rural expenditure or programs that targeted rural things, it was something of the order, I think, of $700 million, but that did not necessarily include mainstream programs. There is a need to separate out the specific from the mainstream. I think Di Wyatt was talking about rural proofing, and New Zealand certainly has that sort of thing, where they run things by to look and see what happens. I do not have figures, and they are not always easily obtained, about the specific rural things. It also depends on whether you aggregate those things that come from different silos, if you like, whether it is health or some other thing.

Senator McKENZIE: Or education.

Prof. Humphreys : One of the things about your evaluation is that evaluation by the Department of Health and Ageing is notoriously bad. It is always an after-the-event situation done by a consultant. Good evaluation really starts with the program to establish the baseline figures—so what it is like before you implement a program and whether you can monitor it along the way. That has the advantage that you can tweak it if things are going wrong along the way, but also you know what the difference is in terms of the incremental shift from where you were at time T to where you are at time T+1 or T+2 or whatever you want to do. We have tried desperately in our work to make sure evaluation is built into things.

To give you an example, we were able to get some funding—the Department of Health and Ageing have given us the last three years of funding—to take an evaluation of the Elmore primary care centre, which is now in its fifth year. What we are doing there is a clinical audit, a community survey and a lot of other stakeholder interviews to see what is happening in that particular service such that it is not only maintaining its viability but producing the sort of health outcomes that you want for the community and the workforce outcomes in terms of sustainability and things like that. What we have been able to show is how this service has been able to sustain itself well despite the policy changes and despite doctors coming and going and things like that. We are using that information, if you like, to try to publicise it so that other small communities, in setting up their practices, can look at what they need to put into place to make sure that viability is sustained and that they have as much capacity as possible to attract suitable providers, whether it is a doctor, a nurse practitioner or whatever.

Senator McKENZIE: Thank you.

Senator MOORE: Where is Elmore?

Prof. Humphreys : It is about 50 kilometres north of Bendigo.

Senator MOORE: So it is a rural community, and you are evaluating that particular service at Elmore.

Prof. Humphreys : Yes, indeed.

Senator DI NATALE: Thanks for the excellent submission. I am surprised that it is something that has not come up before, but it is the first time I have seen a very clear correlation between population and a whole range of factors that are much more relevant for the decisions that doctors make when choosing to work in a particular area rather than simply relying on geography. I suppose what you are saying is that population is a proxy for all these other things that people are looking for. Would that be a fair summary?

Prof. Humphreys : Yes, indeed. Really what we were trying to do is to facilitate this for policymakers. If you can get a surrogate that accounts for a large proportion of the explanation, and that surrogate is readily available, as population data are, and can be readily updated, as the ABS can do, it makes the life of policymakers a lot easier to make sure that the basis for their resource allocation is always state of the art. So effectively that is what we were trying to do: to look for the best surrogate rather than have to go out every year, do an annual survey and wait six months to collate the results.

Senator DI NATALE: It seems to me there is a very neat correlation there, and that is very handy because, as you say, it makes life very straightforward. I am looking forward to having a discussion with the other senators to perhaps discuss this a little further. I suppose, going back a step, the issue that you raise which I think has been the surprise to me in this is the dearth of appropriate evaluation given the amount of money that is being spent. I get the sense we still do not know what the right balance is between relocation and retention incentives, or even the mix between that and the overall increase in remuneration through the other options that might be available—or if either of those are particularly useful, at all. In fact, it may be that all of these other factors that are outside of our control and are not easily modified are the biggest influences in people making a decision to practise in a regional area. Do you have any comment on that?

Prof. Humphreys : Thank you; that is a very useful point. One of the things our work has done is to separate out issues of recruitment and issues of retention. While there is always an overlap between those—retention reflects how well you recruit—a lot of the factors that are associated with the selection and recruitment of a person are different from those that either trigger a person to leave or encourage them to stay. So we have tried to research that sort of thing.

We have battled desperately with this issue of trying to get good evaluation data. We had the nonsensical situation where, in one of the projects that we were doing which was funded through the Department of Health and Ageing, we had to use part of the money to go through freedom of information to get a document that the department had—the results of an evaluation it had conducted—as part of the building blocks. That is the nonsensical kind of secrecy that goes on in terms of they way consultancies are done.

I think it is fundamentally important that once we have identified the nature of the lever that we are trying to pull and allocated some money to pull that harder, we ought then as a matter of obligation see how well it is working in achieving the desired outcome. I think that is a real issue. There are two components. The first is the ready access to data for academics such as those in our own centre, because good data exist but they are often very difficult to get access to. The second issue is getting the evaluation process built into both the research and the program roll-out—the implementation of the program. That would overcome, to a large extent, that problem. As I said, it enables governments to tweak and monitor along the way but it also enables them to look, after a sufficient time period has elapsed, at what the effect of the program was.

Senator MOORE: Professor, we have talked many times on this stuff. But in terms of the proposal you have put before us you have replaced one model, which was based on geography, with another one that is based on population. You said in your statement that the work you had done was about 95 per cent, it seemed, process. I would like to know about the other five per cent, because we seem to find a lot of our time in the system arguing about the ones that are not happy.

CHAIR: The five per cent.

Senator MOORE: The five per cent. I can remember hours of debate about two communities that did not fit the RRMA thing. You also talked about flexibility. From your perspective of research, how do you do that? We replaced one model with another model, and that was very attractive in terms of the kinds of data that was used. All the stuff that went before it about the MABEL data was fine and it shows people's interest. That then translated to a population model. How do you think we should look at the nature of the flexibility?

Prof. Humphreys : I think there is always a case for special circumstances in life. I do not think we ever get the 100 per cent that we want. We try and approximate that as nearly as possible. I recall that when I sat on the retention grant committee there were letters coming in from doctors who were in the Dandenong Ranges of Melbourne bleating that they were undertaking rural practice and deserved their $12,000.

Senator MOORE: It happens on the Sunshine Coast.

Prof. Humphreys : Well, there too. I think Maleny is a pretty attractive sort of place.

Senator MOORE: I do also.

Prof. Humphreys : The reality is that we try to get to the optimal point where the between-group differences—the unlikes—are separated out. With the retention grant some money was always held back for those exceptional cases that might exist. I cannot nominate what they would be in this, because I have not looked at the whole geographical spread.

I have had letters from doctors saying that the stuff I do is the worst thing they have ever heard of in their lives. That is fine. I am always pleased to know that I am not necessarily hitting their mark; it is an encouragement to hit it a bit harder next time. I think where the flexibility lies is that, to some extent, we have got to bridge this gap between the people who take responsibility for the program at the centre—that is, the Canberra bureaucrats—and what actually happens on the ground. I am a firm believer that, if you want to know how well a shoe fits, you ask the wearer, not the manufacturer. I think there are occasions where good CEOs, good practices, know what the best thing is to do for their situation. It does not mean they abuse the program guidelines or anything like that; it means they massage them in a way that fits different groups, whether it is an IMG, a female GP who has significant family responsibilities or an ageing GP who wants to move from nine sessions a week to five sessions a week. There are different ways in which that sort of support and incentive can be used.

We must have flexibility. Time changes the world, but I think there is always core business that remains the same; there are some things about practising in a rural environment that remain the same. People say to me, 'If you've seen one rural community, you've seen one rural community,' but I travel a lot and I can see the similarities between West Wyalong and Condobolin and St Arnaud as I go there. Okay, it might only be two-thirds of the variance, but, by gee, that is a good way to start if you have limited time and limited capacity. Then you can get to know the other things you need to layer on that. I am not sure I have answered your question.

Senator MOORE: You have; it is just it is struggling with imposing a model and expecting it to fit all purposes.

Prof. Humphreys : That is right.

Senator MOORE: I think governments always want to do that, and it never works.

Prof. Humphreys : It is a balance between the 'one coat fits all' and the uniqueness. All the world is unique if you want it to that extent, but there is a significant amount of order.

Senator MOORE: Can I just ask on notice whether your organisation or you yourself have done any work on the rural GP program that the Queensland government is using and that has been talked about several times? Also, we heard evidence this morning about the rural clinical practice processes and the 25 per cent rule at universities on rural doctors. Has your organisation done any work on those two things about how they are done and whether they work?

Prof. Humphreys : On the former, no; I have left that to organisations that are involved in that, whether it is GPET, ACRRM or RACGP—sorry about the acronyms, but you have heard these before, I am sure.

CHAIR: We're all going to be tested at the end of the day!

Prof. Humphreys : Apropos the 25 per cent, we have done a lot of work and we have published about what is happening in the support that is being provided through the rural clinical school training program and the RUSC program, and I have to say I think that program has done very well at the front end of the training spectrum. It has encouraged universities to train their entrance programs, their selection processes; it has created dean's lists and opportunities for disadvantaged rural people. It has done a lot there in terms of early immersion. It is in need of a little bit of scrutiny at the moment, because I think there is a bit of fudging of some of the figures around 25 per cent and what that really means, but effectively it has been a very significant and worthwhile impetus to get the front end of the training spectrum right. Within the graduate and undergraduate medical programs, I think things are going well. One of the big problems, the big black holes, remains the intern year, the PGY1 that I think Di was talking about, where things fall apart a little bit for all the wrong reasons, and then of course it depends on where students take up their specialty, whether it is in a college of general practice or not.

Senator MOORE: Is that work on your website?

Prof. Humphreys : No, it is not. You can access it through the Committee of Deans of Australian Medical Schools, through their MSOD—there is another one for you—the Medical Schools Outcome Database and Longitudinal Tracking Project. That is available through the Medical Deans Australia New Zealand website. There is a special thing there.

Senator MOORE: We will track that down.

Prof. Humphreys : If you cannot, I am happy to assist there.

Senator MOORE: That would be great. Thanks.

CHAIR: This may be a dumb question. Why can't we do a more complicated formula? With your sentinel indicators here, why can't you do a series of equations and plug in the data and come out with a system that is more flexible? I am looking at your table 3 here with your new classifications. Some of those, as you said earlier, seem to be anomalous. So why can't we have a more complicated system, if it just means that, for every town, we plug in a series of indicators with measurements against each of those indicators that actually does measure all of the indicators for why somebody would or would not locate in a town?

Prof. Humphreys : Such an exercise would academically be feasible and possible. Whether it is a worthwhile thing to do in the practical application is a moot point. In terms of understanding, I think we have actually done a lot of that work. If you look at our detailed investigations with the models program, we actually surveyed all doctors. We went intensively looking at the economic and practice activity of practices. We then visited, bought the practice for a day, and interviewed everyone there and looked at it, so you can see what happens. The problem with that is that you will always end up with some aspect of uniqueness. We were able to tease out those things that are common across the board like the demands of being on call and the way they increase as you become more and more remote.

CHAIR: Yes, which is why your indicators seem pretty reasonable. My point is, and getting back to the point that Senator Moore made, we still have that five per cent. So with the current system we have anomalies which people have concerns about. With your system we are going to get five per cent anomalies.

Prof. Humphreys : Five per cent? I do not know but, yes, there will always be anomalies. There would be some places that perhaps do not fit that for some exceptional reason. It might be the Falls Creek practice that is a seasonal practice, for example. Because of that one factor it does not necessarily equate with the group in which it fits.

CHAIR: What is the percentage of anomalies that we have with the current system, with the RA system?

Prof. Humphreys : I would think it is actually quite huge. Since the introduction of the ASGC, I have followed the complaints. I looked at the submissions that came in and I saw many of them saying that it was just inequitable. I am based here and it is inequitable for me. If I were a recent graduate looking out there I would say, 'Why the hell would I go there?'

CHAIR: There seems to be a lot. We have certainly had a lot of complaints. What I am looking at is how we get a system where we reduce the number of complaints. You cannot get a perfect system but there must be a way we can reduce them. It seems to me that the more we can deal with uniqueness the more we can actually recompense and get a system that meets people's needs.

Prof. Humphreys : I do not have an answer to that for you. I think there is a bit of a 'suck it and see' approach.

CHAIR: It is getting down to how we build in that flexibility.

Prof. Humphreys : I think it would need to be road tested. We have not road tested this. We have not had the funding to road test it out there, but it could be done. The department could examine it, for example, and do a comparison of this versus the other things. Then, with rural doctors organisations, have a look to see how it plays out.

CHAIR: Thank you. We have given you a bit of homework and we have taken on some ourselves.

Proceed ings suspended from 12:17 to 13:04