Assessing, communicating and managing cardiovascular disease risk

[Music] Welcome to the Australian Prescriber Podcast. An independent, no-nonsense podcast for busy health professionals.

Hi, and welcome to this Australian Prescriber Podcast. Today we're talking all things cardiovascular risk. So I have with me, Professor Garry Jennings. Professor Jennings is the Chief Medical Advisor of the National Heart Foundation of Australia, and he and Professor Mark Nelson, who is the Chair of General Practice at the University of Tasmania, were co-chairs of the expert steering group for the 2023 Australian Guideline for Assessing and Managing Cardiovascular Disease Risk. Welcome to the podcast, Professor Jennings.

Thank you.

Let's start with a bit of background as to the 5-year cardiovascular disease risk tool, which most of us will be familiar with and have been using for some time. And I gather it hasn't been touched for 10 years before this, and yet there's been a whole lot of new evidence emerge over that time. And one of the issues was that it tended to overestimate cardiovascular disease [risk] for the modern Australian population, and also, it had a limited number of variables, and adding in some other variables adds to the accuracy of the tool. Is that right?

That's right. We've been using a 2012 absolute risk guideline, which included a calculator based on the Framingham population, which is probably 60 years old. And of course, as everybody listening knows, the demographics of heart attack, stroke and cardiovascular disease has changed significantly during that time, and so has the prevalence. As you said, the present equation overestimates risk in low-risk populations, which is our general population, and underestimates risk in high-risk populations, like our First Nations people. So we certainly needed to come up with something more contemporary and better.

What did you base this tool on? Was it a New Zealand model?

Yes, we looked around the world; there are many, many different ways of calculating risk with all sorts of different variables on them. They're all derived in different populations, and many of those populations are quite different to Australia. We were fortunate to find that the PREDICT equation, derived from a very large New Zealand population, met all the criteria that we were looking for, for this equation. So that's been adapted. It still needed to be adapted to the Australian population, so we've gone through an exercise in modifying it slightly for our population, but of course New Zealanders are pretty close to us anyway.

I'm glad to see that our neighbours and colleagues across the ditch have added so much to our health system, so thank you for that.

Yes, and I only wish we collected data as systematically and effectively as they've done, with something like a quarter of a million patients followed up.

Kudos to them. So this podcast, there's a number of factors we could concentrate on, but in particular we're going to concentrate on the assessment of risk and where one fits into the 5-year percentage risk calculator. I do notice that the new guideline recommends that we assess a broader range of people, a slightly broader range, and we are talking, of course, about people who have no current past history of known cardiovascular disease. How have we expanded the group to which we apply this tool?

It's been expanded in the age range, as you said, 45 to 79, and we hope to actually increase that a little bit as more New Zealand data comes in. It's also been applied and validated in people with diabetes who are aged 35 to 79, and we're also recommending that it's done earlier in First Nations people too. So more people are getting it. But as you say, this is not for people that have already got cardiovascular disease, they're at high risk, we're not going to do anything different with them, but it's for the rest of the population.

I guess it's a central point with all these risk calculators is that the real use for them comes in that middle range of people, doesn't it? So [for] people who are highly unlikely to have a problem, it's not really particularly useful. People who we know have a problem, again, we're really looking at that middle range of people, as defined by the article, who gets the tool.

Talking of who doesn't get the tool, just briefly, before we move on to the tool, is people over the age of 80, because it's presumably not quite accurate enough or doesn't give us enough information about what we should or shouldn't be doing with those people. And there's also a couple of other exclusions, based on renal function and familial hypercholesterolaemia.

Yes, familial hypercholesterolaemia is an important one, because it's much more common in the population than perhaps many people realise. And of course they're all at high risk. And then there's the relationship between chronic kidney disease and cardiovascular disease, which we know is a very strong one. And we don't feel that a risk is going to add any value in people with an eGFR less than 45, or a reasonably high level of urine albumin–creatinine ratio. So they're at least the initial groups where we think you probably just go on and treat them as high risk anyway.

Thank you. I'm talking with Professor Garry Jennings who co-chaired the guideline production and has written the article in Australian Prescriber. And now, Garry, the familiar factors to me are age, sex, smoking, blood pressure and cholesterol, and certainly I can't count the number of times I've plugged them into the old risk calculator, whether on my computer software, or gone to the National Heart Foundation website. But let me ask you about some of the extra things you've thrown in, which give a bit more nuance to the individual's risk. So first one is cardiovascular medications used within the last 6 months. What do we do there?

Yeah, that might sound a bit paradoxical, because these are things we give people to make them have a lower risk, but of course they were put on them for a good reason in the first place. And we know that the residual risk with these medications is never brought down to zero. And through the relationships identified in that large New Zealand population, they do make an independent contribution.

Some of the others that are new are a history of atrial fibrillation, and by that we mean real atrial fibrillation that's been confirmed on a 12-lead ECG. We know that the social determinants of cardiovascular disease are really important and it's not easy to plug that into a calculator as a single factor, but we have used the socioeconomic indexes for areas, which is known as a SEIFA ranking in Australia, based on people's postcode.

Now, when you do that and put it into the calculator, you will know that some of the people in your local area might live in a low socioeconomic postcode, but actually they're not at all disadvantaged, so you do have the option of modifying that input, based on what you know about the patient. And this has been a very important feature of this new guideline. It's not something that we're relying on a robot to do; the clinical judgement comes into it all the time.

I'm curious about that, I'll just interrupt you there for a moment. I mean, I think it's actually a fine idea. It's one which, perhaps, might cause some controversy, I would imagine. The SEIFA ranking of postcodes, I assume that gets updated, itself, as postcodes change. So the calculator works, even if you put no postcode in. So it's really up to the judgement of the clinician as to whether the person in front of them is, perhaps, typical in some ways of the socioeconomic stages of the person within that suburb.

Yes, you'd only adjust it if you thought that they were not typical of the local area. And the final thing is, type 2 diabetes status. And that's a really new thing. We've tended to assume in the past that all diabetics are at high risk, but in fact, what this population follow-up showed was that there are actually some quite low-risk people within the diabetes population. So there's the option, and you don't have to do it, but there's the option if you know someone's got diabetes, that you go into a separate subequation, put in some more variables, and out of that, you'll get a more precise estimation of risk, taking into account how much added risk is provided by their diabetes. And sometimes you get a surprise, it's not that high, but of course, most diabetics are at higher risk than people without diabetes.

So once you pop in that they do have type 2 diabetes, I gather a pop-up box sort of asks you how long they've had it for, their HbA1c, the albumin-creatinine ratio, the eGFR, body mass index, and whether they're on insulin. You then add in those as an addition.

Yes. I mean, you may not have all those variables, in which case you don't need to do it and it'll just assume that they've got average diabetes risk. But you can refine the estimation by following that sub-equation.

Looking at risk categories now. A big-ticket item with this new guideline is that you've changed the risk categories. So you still get a percentage number, but the naming of low, intermediate, and high risk has changed from the previous guideline.

Yes, this reflects a number of things. In part, it's mathematics, because we're using a different equation with different variables. In part, it reflects the fact that the prevalence of cardiovascular disease in the general population has fallen since Framingham was developed. And whilst it is different, the actual numbers, we actually don't expect it to make much difference in terms of the proportion of the population people are seeing who will be allocated to each of those risk groups. We think that probably about 15% or so will be in the high-risk group, and the rest will be in these lower risk groups. And of course, it's the intermediate group that we're really interested in knowing what to do with them.

Thank you. And just to clarify for listeners, so low is less than 5%, intermediate 5% to [less than] 10%, and high is 10% or higher. Another thing I was interested in is the reclassification factors. I think a lot of family physicians, GPs on the ground, do think, "Well, the person falls into this category. I know them, and I know that they're really at this extra bit of risk that the guideline hasn't taken into account." And I think it's great that some of these things can now be taken into account, based on the individual GP or healthcare provider's knowledge of the person in front of them.

These are really important. And as you say, they do reinforce the fact that it really needs to be a health professional doing this health assessment because clinical judgement comes in. They're things that we know are important, but we couldn't really put a number to them to put them into the calculator, so we leave it to the practitioner to make an assessment about whether they're relevant or not.

And there's things like family history of premature cardiovascular disease, and I mean a real family history, not someone that's had people in their 80s with heart attack or stroke, chronic kidney disease, severe mental illness, particularly important, we know, in its relationship to cardiovascular disease. Ethnicity; some ethnicities have higher risk than would otherwise appear, and some, indeed, have lower risk, and you can reclassify that way too.

And another good example is, we know now, people are turning up to practices with a result of their coronary artery calcium scan. Now we're not recommending those for everybody, we don't think they're ready for prime time for mass population screening, but if you've got the result in front of you, you'd be crazy to ignore it. And so that's another thing that can help you reclassify somebody's risk.

I think that's a very important point that you are not actively recommending it. We do have to be aware that, compared to the other factors we are adding in, it does cost money, whereas virtually everything else you mentioned is sort of a known factor already. Given all the other factors we know, how many people, does it really alter their risk from one category to another?

That's exactly right. We're not saying it's without value and, in fact, the guideline does give some hints to particular people where it might be of considerable value. What we're saying is that if you would apply it at a mass population level, apart from the equity issues, it wouldn't make a huge difference at population level, even though some individuals may be very much reclassified by the result.

Let's move on now to once you've categorised the risk, it's about communicating that risk. And I guess most clinicians are fairly used to this, but I was very pleased to see an online tool that's been developed and is on the website there, which has a 100-person chart, which is adjustable in real time. Tell me about that.

Well, firstly, our evidence review said that there's no single way of communicating risk to people. You've got to have a number of different techniques and the 100-person chart is a very good way. We've also got a, sort of, speed dial thing, too, and this is the point at which we think, and hope, the practitioner will turn the screen towards the patient and they'll discuss it together. And they might play around and say, "Well, let's look at what happens if we lower your cholesterol, we put you on blood-pressure lowering tablets, you stop smoking…” that kind of thing, and the display will respond accordingly. So it's part of the discussion and, really, the negotiation with the patient about what should happen next.

I'll just read out one sentence from your article which I thought does sum up an excellent approach, and that is that your guideline recommends combining risk communication tools. So that's like the 100-person chart, for example, with behavioural strategies, for example, motivational interviewing, personalised goal setting or health coaching, repeated over time. And in some ways, that pretty much sums up the job a lot of us do.

Yeah, it's not a set-and-forget situation, is it? It's one in which you've got to go along a journey together.

Professor Garry Jennings, we'll finish with a subject dear to my own heart, which is the considerations for First Nations people. I've worked in that area for more than half of my career. We've talked about the adjusting according to ethnicity, once you do the cardiovascular risk tool, and that includes various ethnicities, which are listed in the article and in the guidelines, but in terms of First Nations people, it also broadens when you use the tool in the first place.

Yes, as you know, 45 [years] might be too late for many First Nations people, and this needs to be seen in the context of other things that are recommended and available. There is a consensus statement which says that, at the age of 18, we should start assessing risk in First Nations people. And that's through the usual kind of risk factors, although not necessarily plugged into this equation, and follow them up and manage them.

And then as we said at the beginning, the age range for First Nations people, where this whole thing can be applied, is also younger than in other people. And we're much more likely to see things like kidney disease, which already puts people into high-risk situations, diabetes, and the other comorbidities that occur with cardiovascular disease, more commonly in younger First Nations people.

Well Garry Jennings, thank you so much for giving us that insight, and I refer listeners to the Australian Prescriber article and certainly tap on your favourites online, the National Heart Foundation of Australia's risk tool, which is the updated 2023 version. It's very handy, there's lots of information, and hopefully it will make our job a little easier. Thanks for coming on the podcast today.

Thank you very much.

[Music]

Professor Garry Jennings declares a financial conflict of interest with Sanofi, Novo Nordisk and GSK.