Sam Chang: Hello, everyone. My name is Sam Chang. I'm a urologist at Vanderbilt University in Nashville, Tennessee, and I have the honor of having a session today with Dr. Manish Patel. Manish is a professor of neurological oncology at the University of Sydney. He actually does his clinical work at Westmead Hospital in Sydney as well, and we have had a relationship ever since we were fellows together at Memorial Sloan Kettering. Now, we can say decades ago, and we've asked Dr. Patel actually to give us a presentation on some of the initiatives that he has advocated to help improve quality of care for our patients with more advanced or significant cancers. And so I've asked him to talk about some of the things that he's found in terms of quality and how to help improve quality of care for all our patients.

So, Manish, thanks again for spending some time with us, and I look forward to your presentation.

Manish Patel: Oh, it's my pleasure, and thank you for the invite.

So today, I'm just briefly going to talk about developing centers of excellence for the treatment of major urological cancers in Australia. And I'm going to focus predominantly on radical cystectomy, but the principles of what I'm going to talk about can be applied to all of the major urological cancers.

And this has come about from what's happened in the UK. Back in 2002, there was the NHS Improvements in Care of Cancer report came out. And in that report, the NHS had significantly poor results compared to the rest of the world with regards to cystectomy outcomes. And then they implemented centralization of radical cystectomy and that resulted in quite substantial improvements in care. And so back in 2002, what they recommended was that teams should do radical cystectomy and each team should have greater than 50 radical cystectomies per year. And each team should serve greater than 1 million population, and surgeons who are doing less than five cystectomies a year were recommended to stop.

They've published on their outcomes, and what they've found is that the compliance to these requirements was actually improving over time, and you can see from this graph here, that the green, which is the providers' compliance, has substantially improved over time. The surgeons, which is the purple line, also compliance has improved over time as the total number of surgeries have actually increased. And this improvement in compliance has actually resulted in better survival. And you can see that the graphs to the right, those patients that were in compliant treating centers and surgeons had substantially better outcomes, survival-wise, than those that were non-compliant. So the UK has definitely shown us that centralization of radical cystectomy works.

So back in 2015, we undertook a program of research where we looked at the outcomes in New South Wales, and New South Wales is the largest populous state in Australia and has about 7 million people. And when we started this research, we found that there was actually 50 centers doing radical cystectomy and 39 of those were doing less than four a year. And we published these outcomes. What we found is that the perioperative mortality, if you look at the green line, was quite low for those that were doing high volume. This is per center. But those that were doing low volume, which is the blue line, over the years, they still maintained a fairly high mortality rate. And so this is, of course, quite concerning.

And then we went on to have a look at the disease-specific survival of these patients. And you can see, compared to the low volume centers, the high volume centers, which we stated was seven radical cystectomies a year or higher and I know that's substantially lower than what the UK advised, but there's actually a disease-specific survival improvement in the high volume centers. So we know in Australia that there's the same application to centralization as we saw in the UK, but Australia's got bigger problems than the UK with regards to geographical distances. We have large areas, and patients don't like to travel, so we have to contend with that somehow.

Based on our research, the New South Wales Cancer Institute, which is a government body, obviously saw a need to improve cancer outcomes, particularly radical cystectomy outcomes. So they then went ahead and asked all the low-volume centers to stop doing cystectomy. And as a result, the low-volume centers arguably were quite upset and argued that although there were low volumes, some centers still had good outcomes and wanted to continue doing it. So they then started collecting statewide data on all patients having radical cystectomy. This statewide data, unfortunately, is only administrative data like 90-day perioperative mortality, greater than a 21-day hospital stay, and readmission rates. But you can see from this representative graph that, certainly, the higher volume centers had lower perioperative mortality, had less prolonged lengths of stays, and less readmissions.

So, the question now arises, well, where do we go from here? In my opinion, government administrative data is useful, but it isn't really a substitute for quality of care because you need to take into account case mix, which is the cancer mix, the general health of the population, and geographical considerations, rather than just volume of a center mortality rates and length of stay and readmission. So we'll talk about this, but we're in the process of trying to collect more detailed quality data to evaluate each center and each surgeon to allow better quality of care while still maintaining radical cystectomy, particularly in rural or remote geographical locations.

Thank you.

Sam Chang: So, Manish, you've spent some time in the States. You understand that our healthcare system is totally different in terms of the government control or lack of government control. Tell me what you think would be some key takeaways for US hospitals.

Manish Patel: Yeah. So I think the first thing is that you don't want the government telling you what to do. I think you need to take the bull by the horns and be proactive on this. And as a society, like the AUA or the SUO, need to take the initiative and start evaluating centers and creating a framework on what they think is a quality center and people need to meet those achievements. And if they meet those achievements, they do radical cystectomy. I think once you leave it to governments, they will start using criteria, which you don't necessarily agree with, but can't argue with, and once governments get involved and in the US, the funding model is all through the insurance companies, you lose your control on how to maintain real quality.

Sam Chang: What is your take on transparency of results? In the US, everybody's quite sensitive in terms of identifying individuals, per se. Many times it's anonymized in terms of individual surgeon and outcomes, but we're moving towards at least attempting transparency for price within the US system.

I guess, from my standpoint, is how do you determine what are the best quality metrics? I mean, just as you say, we also are really attempting to try to determine what is quality. And the question is we have trouble ourselves as surgeons in defining what is really important. Is it complications? Is it et cetera? So how are you guys, as you look at the quality metrics, how are you helping to determine that so that you all determine it as opposed to the government?

Manish Patel: Yep. So look, I'll answer your first question first, which is how do you do a program like this and maintain anonymity, et cetera. So across Australia, we have a prostate cancer program called the Prostate Cancer Outcomes Survey, I think it was, and all the data is collected by data collectors, which are independent of the surgeons in the institution, and the surgeons are individually fed back their individual results. And otherwise, it's kept anonymous. Only the surgeons have their own results. And the aim is that if you see your own results and if they're bad, then you, yourself will be motivated to improve. A lot of surgeons don't know that their results are bad, right? So that's the first thing.

Their data that is completely kept anonymous and hence, it's a very non-threatening way of doing things. There's collected data that's presented, but only individual surgeons will have their own individual data. So that's the first thing.

The second thing is how do we identify what quality outcomes are? That's complex, and I think it has to be done with regards to the case mix. For example, in our public hospitals in Australia, the patients are a lot more sick than what we see in our private hospitals and hence, their complication rates, their lengths of stay, et cetera are going to be longer. They have more socioeconomic issues, so we can't get them out of hospital.

So, I think you need to benchmark like for like. And initially, we need to benchmark, say, percentiles. Say, if you are in the lowest 20 percentile for this particular type of cancer in this particular socioeconomic setting, then you need to get above that level. Rather than set a benchmark, I think the benchmark needs to move as we start improving quality with time.

Sam Chang: Yeah, I think that's very insightful in terms of two different things of, one, obviously assessing the situation for individual patients, setting up systems to help evaluate prior to any intervention, kind of the... just like you say, we have case mix index in the US looking at multiple parameters. We could look at comorbidity indexes. We could look at frailty indexes. We could look at a lot of different things to help evaluate those patients and level the playing field.

And then the idea that you brought up that I think is intriguing, which I think makes a lot of sense, is as opposed to a threshold level everyone needs to meet, as you say, is have an evolution of percentiles. And as people improve, you want lift all ships, raise everybody. And I think that would be something that would be... I think we could start off as an institution in terms of individuals. Then you can talk about the region, then perhaps even large areas.

I could see this actually working at our VA system, which is a government system. Tracking of results is actually quite good. You have different regions throughout the country in terms of geographic, and that's something that could be done, and that's something that now, you're making me think this could be a great quality initiative for our VA programs that could then be, hopefully, transferable out.

As you looked at this data, what do you think... is it... because as you know, it may not be the individual surgeon. It may be the hospital. It may be the postoperative care, maybe the lack of social work in terms of being able to support each of these individuals in terms of results. Do you think that there is actually a threshold number because that's how you kind looked at it or would you feel comfortable with certain institutions doing just a few as long as they maintain their quality? I mean, to me that makes... if they're still doing quality procedures, they should be able to do that. And I guess that's the advantage of kind of what you're setting up if measure how you do and you'll be able to continue to do it.

What's going on now currently in Australia? These are being measured? Are those low-volume centers, are they still discouraged or are they still doing them? Tell me what's going on now.

Manish Patel: Yeah. So the threshold for low-volume centers was set by the government. We didn't advocate for that. And they were asked to stop. There are some low-volume centers that are still continuing on arguing that they still have good results. I know a few of these small volume centers and I know the surgeons very well and they're very well trained and I expect they have excellent results. And I don't think volume is the key. In general, yes. Statistically, volume does make sense, but you have individual centers that have low volume and have great outcomes and they should be allowed to continue doing it.

So that's why I think we need to measure better outcomes, particularly disease-specific survival, complication rates, [inaudible 00:14:43] complications, et cetera, and put that all into the mix as to who should be allowed to do it and who shouldn't.

Sam Chang: Yeah, that's where I agree with you 110%. I think I don't think we can fall, especially in the US too. I don't think we can fall in the trap of, "Well, you got to do at least 10 or you can't do them at all." I don't think that... well, that can't be currently mandated, but I think it assumes too much. It's too broad, just as you say because there are individual centers and surgeons that don't do high volume but still deliver quality care. So I really appreciate-

Manish Patel: [inaudible].

Sam Chang: Yeah. Go ahead.

Manish Patel: On the same note, there are also some high-volume centers that deliver poor-quality care.

Sam Chang: Exactly. And so to your point, volume is not a surrogate for quality. I think that is really an important message, and I think within the US, I think, if anything, if we make the results more available to everyone regarding these are how patients do, I think patients will self-select those centers that tend to have better outcomes. Will that happen? I don't know, but in the US, you already have people looking at Yahoo reviews and different types of ways that people are being evaluated. So I very much applaud everything that's been done in Australia in an attempt to try to determine what really is quality and what then leads to better results.

So, Manish, thank you so much for spending some time with us and really appreciate your efforts in an attempt to improve quality for all the patients that are undergoing major procedures for urologic cancers. And I look forward to future results coming from down under to see what's going on, and I look forward to seeing you soon in the US as well.

Manish Patel: That's great. Thank you so much for the invite, Sam.

Sam Chang: Absolutely.