IAN SCOTT: What I’d like is for folk to, I think, be open-minded to the fact that sometimes our thinking is based on old science, based on what we were taught, and that we come to believe and normalise certain practices. And we don't challenge. We don't challenge the status quo.
CAMILLE MERCEP: Welcome to Pomegranate, a CPD podcast from the Royal Australasian College of Physicians.
Medical practice is dynamic, and continually changes with evidence and experience. But costly or prolonged treatment doesn’t always translate into better outcomes for patients. In response, the College has launched EVOLVE—a partnership with specialty societies to identify and reduce low-value care.
EVOLVE is part of a growing international movement to examine clinical and consumer decision-making about overused, inappropriate, or potentially harmful medical treatments. As a founding partner of Choosing Wisely Australia, EVOLVE represents the College’s major contribution to the campaign: helping Fellows develop and promote lists of low-value care items for their specialty.
Professor Rachelle Buchbinder directs the Department of Clinical Epidemiology at Monash University, and has co-led the development of a low-value care list for the Australian Rheumatology Association. Associate Professor Warrick Inder, who teaches at the University of Queensland, is president-elect of the Endocrine Society of Australia, which has also recently released a list of their top 5 low-value items. They are joined on the podcast by Associate Professor Ian Scott, Director of Internal Medicine and Clinical Epidemiology at Princess Alexandra Hospital, and whose research interests include the cognitive biases of clinical decision-making.
On today’s episode, they discuss why they support EVOLVE, the challenges and opportunities of the campaign, and how they work with patients when the best course of action is…to wait.
IAN SCOTT: I’m Ian Scott, I’m Director of Internal Medicine and Clinical Epidemiology at Princess Alexandra Hospital, and Associate Professor of Medicine at the University of Queensland here in Brisbane.
I would define low-value care as care that is inappropriate and confers little or no benefit to patients and may indeed expose them to possible harm. I think it was about 12 years ago really when I became interested in it for two reasons. One, having become aware of the evidence based medicine movement that came out of McMaster University, reading their materials and realising that a lot of what we were taught in medical school—and perhaps even in our training program—wasn't really relevant to our actual practice of medicine in differentiating between what was good care and what was suboptimal care.
I think low-value care, in many respects, is a legacy of our medical training. We are taught things in medical school, and subsequently in our training programs, that at the time are believed to be correct. But we know that scientific evidence evolves and changes over time, and that the half-life of medical knowledge is about seven years. So about half of what we've learnt within seven years is redundant and no longer correct—or has been totally reversed.
I think also it's driven by desire on the part of clinicians to be seen to be doing as much as they can for their patients. We all have a strong urge to try to improve people's quality of life, to reduce their morbid events, but we need to realise that sometimes not doing anything, or doing very little, may be a better course rather than trying to do everything possible.
RACHELLE BUCHBINDER: My name is Rachelle Buchbinder; I’m Director of the Monash Department of Clinical Epidemiology at Cabrini Hospital, and a professor in the Department of Epidemiology and Preventative Medicine at Monash University.
So when I think about low-value care, I divide care up into three categories. High-value care, where the person has a problem and this treatment's been proven to be effective and the benefits far outweigh the risks. Low-value care, is where there is clearly no benefit and even sometimes a possibility for harm. And I guess my area of interest is that big area in the middle where we don't really know where the threshold is.
In my area of musculoskeletal conditions and rheumatology there are several examples of low-value care. The one that's been known for the longest, I think, is imaging for non-specific low back pain. So in the majority of cases of people who present with a new episode of acute back pain, we know that imaging is not required—and we now increasingly know that imaging may be associated with potential for harm.
And the harms might include the radiation that the patient receives, but it also might include labelling. Because as the imaging tests become more sensitive you can pick up a lot of abnormalities that are part of normal aging, if you like. But because the report includes all of the pathologies, patients then become labelled with different conditions and then there is a whole cascade that happens after that. So the patient might get referred to a specialist or someone with an interest in low back pain. They may get injections that we know are not efficacious, they may get surgery. And these downstream effects might cause more harm than good.
WARRICK INDER: My name is Warrick Inder; I'm an associate professor with the University of Queensland, and an endocrinologist at the Princess Alexandra Hospital in Brisbane.
I haven't been working on low-value care historically over my career, but I do actually have a personal interest in eliminating episodes of care or unnecessary investigations that really are more of a burden for patients than beneficial. I really don’t believe that there are subspecialties where it's more prone, I think it's really across the board—medical, surgical, radiological, pathology…pretty much every specialty would be able to pull out examples of low-value care.
IAN SCOTT: The trials that have gone and looked at traditional practices where people have said, “There hasn’t been any good proper randomised evidence here in relation to this practice, perhaps we should do a trial and look at what happens”—in those instances, up to half have been discredited.
The COURAGE trial was a randomised control trial of coronary revascularisation plus optimal medical therapy, versus optimal medical therapy alone—in a large number of patients with stable coronary artery disease, predominantly exertional angina. And that showed that coronary revascularisation did not confer any benefits in terms of mortality, recurrent myocardial infarction, or subsequent urgent revascularisation.
So that was a major challenge in the paradigm up until then, which was that if patients had more than mild angina they should have an angiography. And if the angiogram showed stenosis in their coronary arteries those stenosis should be re-vascularised either by stenting or by bypass grafting. And that was the tradition up until then for many decades.
That trial really challenged that convention, and it was heavily criticised. A large body of cardiologists really didn’t believe the trial, or felt that the patients in the trial were highly selected and therefore didn’t apply to their own practice. But the trial nevertheless was a large study, with I think a representative sample of patients, and has led to changes over time. But those changes have been slow and not of the magnitude that you would expect based on the results of the trial.
WARRICK INDER: My interest in EVOLVE was largely because of my position with the Endocrine Society of Australia. I chair the Medical Affairs Subcommittee and am the current President Elect.
I think EVOLVE has a couple of purposes. First of all, to identify aspects or areas of low-value care, and then secondly to disseminate the identification of these, and perhaps giving practitioners alternatives to undertaking these low-value cares.
And often the alternative is actually to do nothing. And that very fact was brought up at our EVOLVE meeting that was attended by a number of different specialty societies in Sydney. It was actually really good, we had a representative from the College of General Practitioners and we had a consumer representative. And interestingly, the consumer representative was very encouraging of doctors explaining to patients that, “Look, the alternatives here include not doing this test for the following reasons.” So I think providing you can justify your clinical rationale for not doing something, that would be perfectly acceptable to the consumer groups.
The lists of low-value care items have been developed in different ways by different specialty societies. So we were very fortunate to have the assistance of Jason Soon, who is a member of staff at the College of Physicians. He managed to find, using desktop research, a list of 44 possible low-value interventions that had been identified by other programs across the world. So ESA then evaluated this list of 44, we then thought of a few extra ones, and really by a process of elimination managed to whittle that down to approximately eight that we thought were particularly important and relevant to the Australian setting.
At that stage we undertook a literature review on those eight interventions and formulated a Survey Monkey that was sent out to Endocrine Society of Australia members. And they were asked to rank and order the top five most important practices or interventions to avoid. We had a response rate of almost 150 of our members, and from that our top five list was formulated based on the member's responses.
RACHELLE BUCHBINDER: Generally within my specialty people are very supportive of the EVOLVE concept and identifying low-value care practices. Many of the items that we've chosen we can look at the numbers of tests that have been ordered using MBS data. We can look at the number of treatments—particularly things like different surgical procedures— again using the MBS items. And I think it's something that we should be monitoring, testing different interventions, and if it's not obviously changing then we need to think about strategies for improving the implementation into practice.
IAN SCOTT: The rationalist perspective when it comes to quality improvement is based on good science, in the sense that we want to use tools such as education campaigns, revising clinical guidelines, doing clinical audits, and academic detailing. There are a number of techniques, and they're all based on the premise that if we present good science to what we regard as rationally-thinking human beings they will change their minds and do things differently.
Those techniques are effective up to a point, but they still are limited. What limits them is that we really haven't exposed, teased out, and discussed the cognitive biases that are still inherent to a lot of decision-making.
As human beings, everyone is biased to some degree. Our experience and what we've come to believe strongly does affect our thinking and how we respond to new information and new evidence. The one that I guess comes strongly amongst physicians is loss aversion. That means that we will feel pain, psychic pain, if we feel that a patient has come to a bad outcome because we didn’t do something. Whereas if we do something with good intentions but the patient may then have a bad outcome because that care was inappropriate, we don't feel quite as much psychic pain.
WARRICK INDER: One of my colleagues, Professor David Thorpe from Adelaide, made a comment at a recent meeting that endocrinologists spend a large amount of their time worrying about mild conditions. When something is very severe the clinical features are very obvious, and you can make a diagnosis usually without too much trouble and you initiate treatment. When someone presents with a very mild abnormality, the question is: “Is this a variation of normal, or is this the early beginnings of a disease condition?”
And one of the key ways of making a diagnosis in some conditions is the use of time as a diagnostic tool, to see whether there is progression in the patient symptoms and signs. Once that then becomes obvious, we can then be in a position to intervene with a lot more confidence—and knowing that our intervention is much more likely to result in some benefit to the patient. Rather than over-treating something that's very mild, that may in fact be a variation of normal, and expose the patient to potential side effects—both from their investigation and their drugs, or procedures that they're undergoing as forms of treatment.
IAN SCOTT: If you’re a general physician working in a regional town, you don’t have access necessarily to all the MRI and SPECT scans that may exist in the big tertiary hospitals. You have to confront the fact that, “I can't send this patient hundreds of miles away for these particular investigations, so I'm going to make the best judgement I can based on the evidence that I have in front of me, and the results of the investigations that I do have access to.”
And that can show that in many cases the high technology doesn’t actually add value to the patient's care. It doesn't necessarily improve the diagnostic ability, and doesn’t necessarily change what we plan to do in terms of clinical management.
So the patient still has a good outcome, and indeed in some cases has a better outcome because they're not subjected to a lot of high-technology medicine that may induce some harm—but also at a much lower cost and inconvenience to them and to the healthcare system. So I think if we could train or expose our physician trainees to see good, seasoned, experienced clinicians make decisions under conditions of uncertainty, without having access to a lot of high technology, they can see that they can provide just as good care and have just as good outcomes. And also to some extent have a better degree of professional satisfaction, because they’ve relied more on their own clinical acumen and reasoning, rather than default to a lot of investigations in a high-tech environment.
RACHELLE BUCHBINDER: There has to really be a systems, a whole of systems, approach to this. So we need to have support from the profession as a whole, we need the patients very much to understand what we're doing and why we're doing it, and we need to change their expectations for care. And I think that's a really important barrier to changing clinician behaviour and that's the patient expectations. In my research we found it's much easier to change clinician behaviour if you can change the patient knowledge and patient expectations.
WARRICK INDER: You know, we get referrals for an opinion. And if my opinion is that the suggested procedure or investigation is not indicated, I will absolutely tell that to a patient. And the representative at the meeting from the College of General Practitioners was very supportive of receiving written letters, and written information from physicians, to explain to GPs what best practice would be so that the GPs can avoid some of these things that are being identified through EVOLVE. And rather than it being considered, you know, "Big Brother telling us what to do,” he actually felt that the GPs would be very receptive to this.
We were using the example of the thyroid ultrasound, and he said, you know, if he received a letter from the endocrinologist saying, “Thank you for referring Mrs Jones about her thyroid; by the way, we don’t need to have an ultrasound prior to assessing these sort of conditions,”—he would say, "Oh OK that's good, next time I see someone like that I won't order an ultrasound.”
So that sort of just one-on-one communication from the specialist would actually change behaviour.
IAN SCOTT: Certainly patients, and I think the evidence also suggests that physicians, overestimate the benefits of treatments and underestimate the risks and harms. That may be to some extent because the evidence for benefit is given high publicity, and the risk of harm is not given the same emphasis. Or in the clinical trials, the risks and harms have not been measured as diligently as we measure the benefits. So I think that we need to ensure that people are given the correct information as much as possible, that they can weigh up the pros and cons and decide then whether they think it's in their best interest to adopt this form of care.
The question of shared decision making is a challenging one, and I've certainly had the experience of patients coming and expecting me to do things. My experience has been that if you sit down and ensure of course that you’ve taken a history and a proper examination, and you’ve given patients time to tell their story, then you give them a reasoned argument as to what you think is wrong and what we should do and what we shouldn't do. And often I will preface my terms by saying, "Well if I was in your shoes, this is how I would be thinking.
Now, that's not to say that the patient has to adopt my thinking, but it then puts it in a personalised form and I'm giving them, “Yes, I'm identifying with you; I can see where you’re coming from; I empathise with you.” In my experience that diffuses a patient's unrealistic expectations. And you’ve done it in a way that recognises the patient's rights to have their story heard and for them to voice their preferences. If you adopt that sort of approach then in many cases the patient goes away at the end of the interview satisfied, and what they thought they wanted in the beginning has now changed considerably.
RACHELLE BUCHBINDER: Patients are very surprised to hear when you tell them that actually there's good evidence that “this doesn’t work.” They're very surprised about that when they’ve been receiving that treatment and been told that it works from others, even when they can see that it hasn’t worked when they’ve had that treatment. I mean, I spend my life taking people off opioids, telling people not to have surgery, stopping MRIs…that's really my practice lately. And it's hard, and not everybody will agree with me. And not everybody will adhere to what I've suggested.
IAN SCOTT: A cognitive autopsy really is a process of reflection on a case that you may be currently caring for—or a case of a patient whose episode of care has been completed. But what you’re doing is that your dissecting out the care that was given in a reflective environment with some time, some quarantine time. And with colleagues who are open-minded and prepared to give a balanced view of their perspective on the care of this particular patient. And to learn from that, “OK, what could we have done better, what did we do right, and OK, what should we do in the future.” So it's a form of cognitive dissection versus anatomical dissection, in terms of the traditional use of the word autopsy.
A resource that I particularly like is the “Less Is More” series in JAMA Internal Medicine. These are pithy, concise case reports of people who have not had appropriate care, and sometimes they’ve had poor outcomes. They're just one page, easy-to-read—they start with a short clinical summary of the case and are then followed by a commentary which also cites the evidence as to why this care was low-value and perhaps what we should do instead. They come out on a fortnightly basis and I distribute them to my staff every time they come out, because the patient stories can be powerful.
RACHELLE BUCHBINDER: Having thought about this area a lot, I think that we have to be careful and bring everyone alongside with us. We don’t want people to do something that they're not comfortable doing, we don't want to be critical of our colleagues; we don't want to be punitive. This has got to be something that we all agree is important for the benefits of our patients. And I think by thinking about the health outcomes as the central value we should all be able to work together to reduce low-value care.
CAMILLE MERCEP: The EVOLVE website publishes lists of low-value items from specialty societies that have contributed to the campaign so far. Visit evolve.edu.au to see your specialty’s list and to access resources to help you use it. To find out how to contribute to lists in development or to get involved in future reviews, contact your specialty society. You can also email the College—please write to firstname.lastname@example.org.
Many thanks to Rachelle Buchbinder, Warrick Inder, and Ian Scott for speaking with us on today’s episode. The views expressed are their own, and may not represent those of the Royal Australasian College of Physicians. To read more about any of the studies or other resources mentioned on the program, visit the Pomegranate website at racp.edu au/pomcast.
Pomegranate comes to you from the College’s Learning Support Unit. The program is presented by Camille Mercep, and this episode was produced by Anne Fredrickson. Next month we’ll be focusing on the growing field of adolescent and young adult medicine, which will be launched as a separate traineeship by the College in 2017.
MICHELLE TELFER: Although medicine for infants and young children—as well as medicine in the older age groups—has improved dramatically, health outcomes for adolescents and young adults haven't improved to the same extent. And that's where I find adolescent medicine is a really interesting and progressive field to be in.
: Please join us.