'Beyond Bricks and Mortar - Building Quality Clinical Cancer Services' Symposium 2011

Strengthening Collaboration to Improve outcomes for Patients - Professor David Currow

Up to Radiation Oncology

prev pageprev page| TOC |next page

Professor of Palliative & Supportive Services, Flinders University; Chief Cancer Officer and Chief Executive Officer, Cancer Institute of NSW

Download presentation by Prof David Currow (PDF 1216 KB)

Norman Swan:

.. and during the course of the conference please feel free to come up and talk about things that you think should be on the table for discussion that might not be, and I’ll be sure to introduce those elements as we go along.

Our first speaker – our second speaker of this session after the opening is David Currow, Professor Dave Currow, who chairs at Flinders University, is a palliative care physician and palliative care researcher of international repute, has been responsible for doing research in palliative care that other people said was impossible, through his randomised clinical trials for example. He is Chief Cancer Officer of New South Wales, and Chief Executive of the Cancer Institute of New South Wales, having just been Chief Executive of Cancer Australia. Please welcome David Currow.

Prof. David Currow:

Thank you very much for your introduction Norman – very generous words and thank you Brian and the team for inviting me to be here today. There’s a wonderful perversity that brings together 170 people to talk about cancer. That perversity is that the rest of the health of our community is sufficiently good that we live long enough to get cancer, because age is still the greatest single risk factor for us getting cancer. And as we reflect on what we need to do to improve great outcomes, already the challenge for us is how do we do that in the broad context of health.

So I’d like to consider over the next few minutes what’s working well in cancer control and how’s that reflected in the Australian context; what are the areas where we can improve cancer into the future; and how can we collaborate more effectively in order to drive systems improvement right across our community.

So what is working well? Australia enjoys some of the best health outcomes in the world. For disability adjusted life expectancy we are second only to Japan, and I don’t need to remind anyone in this audience that for 50% more expenditure per capita, the United States comes in at number 25. So we’re doing something right in order to live long enough to get cancer, and when we do get cancer our mortality rates – and these are a little old, and the new set is slightly better than it was then – but our outcomes are incredibly good. The countries that sit above us are either from Scandinavia or Switzerland, the unifying factors I think have to do with very well distilled alcohol drunk to keep warm. And I’m delighted that NH&MRC have increased the recommendations from their original draft to two standard drinks a day with no alcohol-free day- drinks a week. It’s fantastic.

What’s that sorry? They are only standard drinks, thank you Norman.

I think the important issue is not only as Brian has pointed out is incidence increasing, but importantly mortality is going down, so as you can see the match in colour here between the solid lines and the dotted lines are diverging, and are predicted to continue to diverge into the future. That says that for every person who’s getting cancer your chances of being alive and well in 2011 are much much better than they were 30 years ago and are likely to be much better in another 20 years time.

And as we look out to 2036 we can see that that divergence between these two lines continues into the foreseeable future. I’m not sure I trust projections to 2036 but this is work that has been done using Australian data.

If we look at that change over the last two decades for all cancer, our five year survival has increased from well under 50% to in Australia over 64%. And if we look – and even if we exclude prostate cancer from that, given the large blip that we saw in the early 1990s of the early diagnosis of a number of people with prostate cancer, the numbers still hold up very strongly.

And if we look across this slide we can see the extent to which that improvement reflects cancer control. Cancer control particularly in terms of treatment once people are diagnosed with cancer. And the news there is excellent that for most cancers we have seen in both genders significant and sustained improvements even in the last decade.

New South Wales was part of a benchmarking process which was finally published at the end of last year. It was really to ask how are we doing in comparison to other jurisdictions nationally and indeed internationally. It was driven out of the United Kingdom, which as many of you will be aware has invested more than a billion pounds Sterling a year for the last decade with a view to trying to improve cancer outcomes. So this was benchmarking against Scandinavia with the exception of Finland, the United Kingdom with the exception of Scotland – not sure why they didn’t want to come and play – the four largest provinces in Canada, and finally Victoria and New South Wales.

One single analysis was performed by Michel Coleman and his team in London, and so for the first time we probably are comparing like with like in whole of population registries that have excellent coverage. Notwithstanding the editorial by Peto and Burrell in the British Medical Journal last year. Where did we sit within New South Wales? We sat first with colorectal cancer and we sat in the top group for the other three cancers that were benchmarked. But importantly the work doesn’t finish there, there are four more modules in which we’re participating, particularly looking at the beliefs and behaviours of the population as a whole; the beliefs, behaviours and ability to behave – ability to act for general practice; importantly, looking also at the actual treatment that is offered to people who have been diagnosed with cancer, and then finally drilling down into some specific cancers.

At the end of the day it’s going to be important to understand each of those steps because the variations are really quite marked. If we look at this slide, it looks as though we’re running well with colorectal cancer; the same figure in England today is less than 55% five year survival. So roughly comparable populations, roughly comparable health services and yet an absolute difference of more than 10 percentage points in five year survival from the same cancer.

I think one of the challenges in Australia is how do we stay ahead of the game? How do we stay at the forefront of improving survival, and certainly as we’ll come to in a moment, we need to decrease incidence in the first place, but we also need to be working assiduously to understand what are the key success factors that lead to us doing demonstrably better than some of our peers.

Where are the areas where cancer control can be improved? Cancer still is the biggest single burden across our community in terms of life years lost or by a number of other measures that we may choose to use. It still is having an enormous impact right across our community, and if we compare that to the news in cardiovascular disease dealt with in the four graphs on the left of this slide, we haven’t seen the same dramatic responses that we have in cardiovascular disease. We’ve got great news and there’s no doubt that we have improved five year survival and that that is reflected across a broad spectrum of cancer, but we still have a long way to go.

In terms of defining improvements in incidence, decreases in incidence, smoking rates continue to fall, and I know AIHW have just released their national data, each of the jurisdictions measure slightly differently and we’ll be releasing New South Wales data comparable to this I hope in a couple of weeks with adult smoking rates down to 15.8%, so a 1.4% absolute drop in one year. And if we can continue that momentum, we will make the 2020 aim of less than 10% of the adult population smoking. I think the really great news here relates to school children, with only 8.6% of them smoking regularly. That is good news for the future, and we need to build on that strongly.

Our screening rates: lots of variation across the country and I think if we are to be fair for breast, we need to take into account as the national evaluation did the number of women who are having bilateral mammography with no other procedures through Medicare. We also probably need to be a little more generous with ourselves and ask ourselves has a woman had a pap smear in the last three years, has she had a bilateral mammogram in the last three years? Because if we take that into account, then our figures for both of the first two cancers on this slide come up to about 70% participation. And much more realistically reflect what we’re seeing in both incidence and mortality.

Colorectal cancer: I think the biggest challenge that this slide throws to us all in cancer control is how do we encourage men to actually participate in an evidence-based program that has shown a decrease in mortality. The program is not entirely evidence-based, but colorectal screening has certainly shown a decrease in mortality and we’ve had that information for a long time. How do we get men engaged in that process?

We still have major gaps in outcomes across our community, however. And the sobering thought from John Condon and his group is that stage for stage, cancer outcomes for Aboriginal and Torres Strait Islander communities are demonstrably worse. This is not simply late presentation. We can no longer simply say look, it’s about remote services, we can’t get to the people. This is having presented with cancer, stage for stage, worse outcomes. And if we can’t do something about that in the next five years, in the next ten years, then we will have failed the community in which we live.

What about rural and remote? Again, it’s very glib to say this is simply about access; if we had better access we wouldn’t have this problem. We’ve started to unpack that, and I draw your attention to the last row of this slide. This is asking what are the odds of presenting with localised cancer. And if you look at metropolitan versus rural/remote in NSW and look at lung cancer, you are 27% less likely to present with localised disease. Now part of that is access. But part of that is also getting a message out there that we are able to make a difference to cancer if we find it early. That we have to dispel the nihilism that has been so much part of our thinking in cancer control across the community for so long. So although access may be part of this, we need to think far more broadly and engage the community as a whole as well as primary care in the discussions around how cancer treatments are changing in order to improve outcomes.

So finally, how can the system meet needs more effectively for people diagnosed with cancer? What are the collaborations that you and I need to foster if we’re going to build on a 64.4% five year survival for all cancers? There are a number of areas. Firstly we need to listen to consumers. And at a systems level I don’t believe that we’re doing that effectively today. We hear time and again of difficulties at transition points, of people falling through the holes in the floorboards at those times, and at a systems level we have not made the changes that are necessary to ensure optimal outcomes. How are we ensuring the physical and psychological wellbeing of people diagnosed with cancer – not just during treatment, and I don’t think we necessarily cover ourselves in glory then, but certainly in the transition back to life having been successfully treated for cancer across our community? It’s a major challenge for us and we invest very little resource, and yet there’s a burgeoning evidence base that says we should be investing heavily in this space, and that there are health economic benefits at a community level that would justify such investment.

And for god’s sake, in 2011, why can’t we address parking? It’s not rocket science. It’s at the top of every single survey and we just need to do something about it. We need to get over it; we need to get over shrugging our shoulders and saying, “How can we possibly address the issue of where to park a car while someone’s getting treatment?”

At the same time we have a workforce of caring, committed clinicians, and so the rest of my comments today are entirely at a systems level. This is not about individual clinicians – every clinician that I have met who works in cancer is committed to the very best possible outcomes for the people whose care is entrusted to them. But at a systems level, how do we improve things? How do we harness that goodwill, that enthusiasm, that commitment, to improve outcomes?

So as we’ve already said we need to collaborate much more with primary care, and that will be a theme that you’ll see in several subsequent slides. We need to engage primary care in the conversations about where there are significant changes to cancer treatment, so that people are not told, “look, lung cancer, nasty disease” – and, as one of my colleagues used to say to his patients – “any treatment is a cruel hoax”. How do we move beyond that? And for much of our workforce when they trained, that was indeed the truth of it. So we need to think about how we can change that.
How do we ensure that having a reasonable suspicion that this person has the most efficient path to diagnosis, to a definitive diagnosis and a plan for care? And to suggest that we have anecdotal evidence is dismissive. The problem is that every time I talk to people who have experienced the system, this is where the system most is likely to fail. And yet cancer services are saying, “we’re not diagnostic services”. General practice is saying, “we’ve got the challenge of how to do this most efficiently”. Again, if we’re going to collaborate, we need to collaborate strongly in this space to ensure that that process can happen quickly and efficiently.

On the 19th of this month we’ll be launching the new online cancer services directory for NSW. The project officers covered more than 40,000 km in meeting with multidisciplinary teams and actually finding out from them what they do, where the parking is, and what services are available. I’d like to thank Cancer Australia for its generous funding of this; it was a great initiative, and it will be able to identify closest multidisciplinary teams for general practitioners who have not had a lot of contact with health services, or have not had a lot of contact with health services regarding that particular cancer. It’s geo-coded, it’s got Google Maps, it’s got fantastic patient resources, it’s got great GP resources, and if we’re going to make a difference we have to make a difference for people who are occasional users of cancer services, because that’s where the greatest problems are. It’s not for the 50% of GPs who have a well-developed referral network – although they can use it – it has to be across the system. And as we look at that in more detail, we get down to the descriptions that the multidisciplinary teams themselves want as their description of what they do, what they offer, and how they can best interact both with patients and with general practitioners.
Top of page
What about during treatment? Again, general practice primary care has been largely absent from these conversations, and at times not because they want to be. So in redesigning the architecture of eviQ, all of the protocols have an area for primary care or for non-specialist services who would be supporting those people. The emergency room, primary care, other physicians, other surgeons. How do we ensure that in a couple of very quick, easily digested, standard format screens, they can understand what the aim of this therapy is, what this therapy actually is, and what are the likely side-effects for which they need to be responsible in providing care? And so ensuring that we have point-of-care data, rather than trying to make experts of all clinicians be they an allied health nursing or medicine is a major platform if we are going to get better collaboration across the system. It is ultimately with the design to create a community of practice, an ability for practitioners to work seamlessly jointly in genuine shared care models, and the pilot programs that have been done nationally show great opportunity. The work of the National Breast and Ovarian Cancer Centre in the space of shared care shows that it is not only feasible, it is valued by patients, their general practitioners and their specialists and can be done effectively and indeed cost-effectively.

Multidisciplinary care: interesting to look at the literature as it slowly evolves, and I do acknowledge that it’s a very slow evolution but these are large and complex studies that need to be done. In terms of patients: improved survival, improved satisfaction, improved perceived quality of health care; for professionals: improved patterns of treatment, better use of existing evidence and better communication between clinicians at all levels of the system. They have to be good things; they’ve already been measured. Can we ensure that they are actually there at a system-wide level today, so that those sort of outcomes can be guaranteed to all people who come in contact with cancer services.

What about other collaborations? You know, we need to learn a lesson from HIV AIDS. They did some really, really sensible things very early on. I’m sure there was a lot of blood on the floor, I’m sure they had to bring in the sawdust and the hoses at morning tea and lunchtime, but they got together and said, “give us money for AIDS and we’ll do the fighting after that.” In cancer we want to fight between disciplines and between tumour streams. And we need to get over it. We actually need to say, “Give us money for cancer" because we will take responsibility for ensuring that we have the right spread of disciplines, not just in medicine, not just in nursing but including allied health, and that we need to go to government and to other funders with a united voice in that space. If you need another radiation oncologist, chances are you need another medical oncologist, you need some more allied health input, you need some more nursing input, and you need some palliative care. Sad as it is, one in two people who are diagnosed with cancer will have their life shortened directly as a result of that. And yet our planning at a service level does not acknowledge that; does not adequately support people and again does not help people in that transition.

So how do we get collaboration between service providers? We’re working to provide data for benchmarking in a way that I don’t think has been made available directly to the clinicians, and ultimately to a broader community before. By bringing together the NSW Cancer Registry with existing data sets – and I stress existing data sets – let’s look at what we can find. Firstly we have to take data and transform it into information, and that process is really working not only with biostatisticians, hopefully biomathematicians in the not too distant future, epidemiologists – it’s also working with clinicians, end users, coders, everyone who has input into that process. And although there have been enormous steps forward in particularly the eastern seaboard states in this area, the question is how do we identify and act on variations in outcomes that we see across the system? So we’ve embarked on a program – this is not using the clinical cancer registry data. Hopefully that will be available in the next couple of years. This is the population registry data with the admitted patient data set. And we’ve looked at surgery to start with, but we will be looking at medical oncology, radiation oncology, and those cancers that demand multidisciplinary care in exactly the same way. What we’re getting together now is the analysis plan to make sure that we’ve got it right, and that people have faith in the data that we’re presenting.

We’ve just looked today at length of stay, unplanned readmission for that entire episode – so if you’re in the private sector and go to a step-down unit at another site that would still be included in the episode – and then 30 and indeed 90-day mortality. Now 30-day mortality, 90-day mortality is incredibly important. You know, if you’re not going to see the AFL Grand Final, you kind of want to know that. And more importantly, if you’re not going to see the AFL Grand Final and you’ve been cured of your cancer through that surgery, you might be really pissed off. So if there is variation in the system at that level, we need to know about it and we need to act on it.

So these are standard indicators from the NSW health data dictionary – there’s nothing surprising there. As we look across, we’ve got two groups of slides: one looking at peer-group hospitals, because the example that you’re getting today is a very low-volume cancer surgery and we don’t need to identify that, but importantly you’re looking at peer-group hospitals. So principle referral hospitals A are large metropolitan teaching hospitals, B are slightly smaller but major metropolitan hospitals. Major metro are the step down from there and then major non-metro are mostly regional hospitals and as you can see, there’s the private sector.

Importantly on this slide, no great difference in age, but a very big difference down here in the co-morbidity rates of people seen in the private sector. Less than 20% versus more than 30% in almost all of the others, at least numerically.

The degree of spread at presentation: really not a lot to hang our hats on there and we can talk about the anomaly around distance at some stage and we’ll do that over lunch. But I do want you to look here – the number of emergency admissions sit at 3.3% in the private sector versus double digits for all of the public sector up to and including 22%, 23% for major metropolitan centres.

So what does that look like as we start to look at eight years of data? So it’s not a small data set that we have. Firstly I want to draw your attention to the range of, the number of procedures. Here in the private sector we’ve got from 1 to 95 over those years, 1 to 24 in major non-metro, 3 to 25 and if we look at the median numbers, the question is what is an adequate volume? They are not per year, that’s over eight years. Do you want to go to a centre that does three major surgical procedures of this order of magnitude? That’s one every two and two-thirds year; or do you want to go somewhere that at least does it once every five weeks. That’s really the challenge that we need to think about.

So what does that look like graphically? For length of stay greater than three weeks using principle referral hospital A as the reference point, you can see that certainly principle referral hospital B’s are statistically different to that. When we start to look at readmission rates, we see a big shift for major non-metropolitan services. When we look at mortality – and this is having controlled for age, gender, co-morbidity, emergency status and extent of disease, so it is comparing like with like. Even controlling for all of those things, private hospitals still do better. And to look at that in a table we see exactly the same data.

What I’d like to take you to now is to move away from peer group institutions to volume per institution. And this is where the rubber really hits the road. So this isn’t looking at peer groups, this is simply saying, “what is the volume of throughput for a particular institution per year?” And if we look, we’ve got 0–2, 2–6 and greater than 6. Again I haven’t identified what the procedure is but we can talk about that at lunchtime. What I want you to look at is 30-day mortality rate. If you’re doing one or two procedures, you’re mortality rate is almost 7% at 30 days. Down to 2.8% if you’re doing more than six of those procedures in the institution that year.

Now, the question is always the patient who comes in and says, “Look doc, if you don’t do it at the local hospital I’m not going to have it done. I’m going to go home and die.” Well, the bottom line is if that informed consent includes “You have three times the risk of being dead at 30 days”, then I’m happy for that person to go home. But if the informed consent does not include that, then we’re leaving out a really crucial point. And if we look at 90-day mortality the same trend is there: 10.6% down to 4.9% depending on volume. We’re cutting that at 15 procedures a year so that we have that picture also, but at the end of the day these are sobering data and at a systems level we need to respond to it. We need to say there are some things for which you need to travel if you’re going to get best possible outcomes; and some things for which you don’t need to travel, and I’m delighted to say what I haven’t shown you this morning are cancers where we do not see a volume outcome difference. And that means that there will be a quid pro quo, that means that there will be some horse trading around saying let’s stop people from – suggest that people don’t travel for some of these procedures because they can get world’s best outcomes at their local hospital, and for others let’s at least get their initial treatment somewhere else.

So the last collaboration that I’d like to draw your attention to is that if we’ve got data like that from a single jurisdiction without the input of Medical Benefits Schedule or Pharmaceutical Benefits Schedule data, how much richer it would be to follow people through the system between private and public sector, between in-patient and out-patient, and actually understand what’s really happening.

So the last collaboration that I’d like to draw your attention to is that unless we get strong collaboration between governments to share that data in a meaningful and timely way in order to understand the performance of the system, we have lost the greatest single opportunity to improve five-year survivals in the next five years. There are lots of things that we can do to improve but we actually need to understand how the system is operating and ensure that people are getting best-evidence practice informed by dynamic data that are reflecting how the system is delivering care today. If we can do that then it will be exciting in 2015 to look back and say, “we have made a huge sustained and affordable difference to cancer survival in our community because we were prepared to take on the system and improve the way the system delivers care.”

Finally, we need to invest in the science of how to get existing practitioners to take up new knowledge in an evidence-based way. The Cochrane reviews in that area are incredibly lean, and certainly the Cancer Institute NSW is directly investing in the science of improving the uptake of evidence into practice now. We’ve got a long way to go but at a systems level there is no doubt that in the foreseeable future if we can bring together consumers, primary care, other clinicians and government, we can make a sustained and very exciting difference to cancer outcomes into the future. Thank you.
Top of page
Norman Swan:

Thank you David. Time’s going on, but let’s take a moment to intersect with this shy and retiring presentation. How does health performance such as it is in Australia at the moment intersect with this? So you’ve got Medicare locals, you’ve got local hospital networks, in theory you’ve got more local autonomy – that could mean increased fragmentation. So what’s you’re thinking about how all this happens within this threat and opportunity here with current reform?

Prof. David Currow:

Look, I think there is a very real threat that if you disaggregate larger networks there is a chance that you may indeed start to get worse outcomes because every local health network says we can do everything here. But equally I don’t believe that clinicians when presented with quality, credible data do anything but respond to it, and I think what we’ve lacked in the system is timely credible data in which clinicians have been a part of both the collection, the analysis and the interpretation in order to make these decisions.

Norman Swan:

But one of the key points you make is the sharing of data across jurisdictions. It’s bloody hard at the moment with large states. Now you’re going to have Medicare locals and local hospital networks. Do you feel that there are enough – I know there’s a lot of discussion in Queensland about this, that certain things have to remain at state level. When you look across the cancer data etcetera across the jurisdictions with becoming increasingly local, do you think enough’s been held back by the state to allow that sharing to take place?

Prof. David Currow:

Well I’m delighted that the most recent draft of the agreement explicitly includes data sharing in a way that hasn’t been there before, and what can happen within states can continue to happen. I think the issue is that the Commonwealth and the jurisdictions have to come to an agreement to share key data, particularly around Medical and Pharmaceutical Benefits Schedules.

Norman Swan:

If you want to make a comment, by the way, what we’ve done is put standing microphones rather than somebody running around spare trying to find you with a microphone, so just come up and make your point and I’ll bring you in as it goes along. So what is a step-wise strategy for dealing with the appalling outcomes in Indigenous communities?

Prof. David Currow:

We need to address that at absolutely every step of cancer control. The Commonwealth has continued to invest in Aboriginal Quitline initiatives; we need to look at screening rates which are at least 20% lower than the rest of the community in NSW; we need to work with primary care and opportunistically ensure that we are identifying people at higher risk for particular cancers; and certainly once someone from an Aboriginal or Torres Strait Islander community is identified as having cancer, I think at a systems level we need to change the way we approach that. What is their nutritional status, what are the supports that are in place, what are the travel restrictions that may have them making choices that are not going to necessarily optimise cancer outcomes but are absolutely congruent with that person’s belief system, and how do we understand that and start to make a difference?
I hope that radiotherapy in Darwin is going to make a big difference in terms of some of those issues, but the largest communities are still urban communities and we have a long way to go.

Norman Swan:

Just a final question in the lead-in to Jeremy’s presentation: do you think that the country’s been sold a pup with this $560 million regional cancer centre program to some extent, because you’ve just said multidisciplinary team-based care is state of the art. That’s what increases survival, gets better outcomes, people don’t fall between the cracks – you know the story, but the workforce issues in country towns, well-meaning, good people working there – I mean it’s very hard to get a proper multidisciplinary team together, particularly one that involves psychologists.

Prof. David Currow:

Of course we haven’t been sold a pup, Norman, absolutely not. We’re not talking about any country towns, we’re talking about country towns where people want to go and live. They’re lovely places, I’ve visited them!

Norman Swan:

We’ll come and visit you there. Thank you, could you please thank David Currow.

Top of page

prev pageprev page| TOC |next page