Transcript

PLEASE NOTE: While an effort has been made to correct errors in this AI-generated transcript, some mistakes may have been missed. Users should take responsibility for checking anything drawn from here. Further, the views expressed are those of the guests and the podcast may not represent the RACP’s position on the subject. This is intended as a conversation starter about emerging evidence not as a replacement for guidance from clinical regulators or an expert-reviewed consensus.

MIC CAVAZZINI: Welcome to [IMJ On-Air] from the Pomegranate studios of the Royal Australasian College of Physicians. I’m Mic Cavazzini, and I’m publishing this one day before I go on long-service leave, after ten years at the RACP.

But don’t worry, I’ve cued up three more episodes to keep you going while I’m away. The first, on managing at TB outbreak in a remote Aboriginal community. The second, on cultural safety in palliative medicine and the last one on the RACP’s first youth advisory group that was involved in developing a positions statement on youth appropriate healthcare.

If you’re itching for still more more high quality audio, I suggest you look up PodMD in your podcast browsing app. Their episodes are usually feature specialists presenting 15 to 20 minute ‘need to know’ summaries for GPs and other generalists. For example, one recent episode featured an RACP gastroenterologist explaining the interpretation of colonoscopy and histology reports. Another, episode from an endocrine surgeon was all about the detection and management of thyroid nodules. You can find them also at podmd.com.au.

But today, you’re getting a taster of the College’s Internal Medicine Journal. The paper for today’s discussion was published in the June edition of the IMJ and is titled “Implications of race-neutral equations on interpretation of lung function in Australia.”

Spirometry reference equations are central to differentiating obstructive ventilatory defects, like COPD, asthma and bronchiectasis from restrictive ventilatory defects, like sarcoidosis, asbestosis and idiopathic pulmonary fibrosis. Lung function testing starts, of course with metrics like the forced expiratory volume in 1 second and the forced vital capacity. Across the population, those metrics will fall out on a bell-curve and you can pull up a predicted health score by inputting a patient’s sex, age and height into a reference equation. When you integrate the patient measurements, there are particular cutoff values that support a diagnosis of a particular type of ventilatory defects. A restrictive ventilatory defect is defined as a FVC below the lower limit of normal, equivalent to the 5th percentile. An obstructive ventilatory defect is defined as a reduced FEV1/FVC below the lower limit of normal.

Now, the shape of this bell curve I’ve described does change from one population group to the next, and it was classically thought that the ethnicity was an important driver in this. As such, race-specific equations were developed to provide an adjusted reference value appropriate to the patient in front of you. As noted in the paper, prominent examples of such equations have come from the National Health and Nutrition Examination Survey from the USA and the Global Lung Function Initiative from the European Respiratory Society.

But times they are-a changing. In recent years, the merit of this adjustment for ethnicity has been debated in major reviews in the European Respiratory Journal, Chest, and the American Journal of Respiratory and Critical Care Medicine. That perhaps it’s not race that drives population differences but other modifiable factors. Or if there is some pre-existing difference, this needs to be explored also in overlooked groups such as First Nations people.

The question for today’s researchers was, what effect would it have on an Australian patient cohort if diagnostic classification of chronic respiratory diseases was done using race-neutral equations of lung function? Now, I am not across all of this literature, because I wasn’t expecting to have to conduct this interview myself. Unfortunately, the IMJ’s editor for respiratory medicine was called away just minutes before recording, so here is me attempting to wing it with the able guidance of authors Dr Helen Luo, Professor Paul King and Dr Daniel Tan.

DANIEL TAN:      So I'm Daniel Tan. I'm a final year respiratory and sleep advanced trainee based at the Royal Melbourne Hospital and Monash Health. and non-clinically I'm a postdoctoral research fellow with the Allergy and Lung Health Unit at the University of Melbourne. I have an interest in the life course effects of asthma and CRPEDI lung health, as well as their pharmacological management.

MIC CAVAZZINI:                Thank you, and the paper's first author, Helen Luo.

HELEN LUO:        Hi, I'm Helen. I'm currently a basic physician trainee working at Monash Health. hopefully will become a respiratory advanced trainee next year. We'll see.

MIC CAVAZZINI:                And Paul, if you want to give us your two line CV.

PAUL KING:        I'm a respiratory physician at Monash Medical Centre. I've had long-term interest in lung disease, both obstructive and restrictive, and particularly the role of reference equations and the interpretation of these conditions.

MIC CAVAZZINI:                Okay, so who wants to give us the background? You've set it all out in the paper, the rationale for race-based adjustments that came from historical studies demonstrating systematic differences in lung volume between ethnicities.

DANIEL TAN: Yes, I can start off with the introduction. So, conventionally lung function is interpreted in reference to specific reference equations. So, where a person or a patient should be, referenced to their age, their height, their ethnicity. And we use these specific reference values to derive a percent predicted to determine where somebody sits relative to someone of those parameters  in the normal population.

Conventionally, we've used reference equations which have come from several big cohort studies, the major one being the NHANES study from the United States, and then subsequently the Global Lung Initiative collaboration, which included data from nearly 130,000 adults and children worldwide. There's been a growing recognition that some of the data that contributed to these studies may have been impacted by a number of social and environmental factors that result in the reference equations, maybe not being representative of a truly healthy population.

And as a result, there's been a shift in recent years to moving towards phasing out ethnicity as one of the key variables within these reference equations and moving to what we now term as race-neutral reference equations, trying to address some of those factors. And also, for some of the patients or participants who may not quite neatly fit in with one of the ethnicity category categories that were present previously.

MIC CAVAZZINI:                To put it simply, is that to say that those scaling factors or whatever for ethnicity, it's probable that rather than genetics being the driver, it's probably social factors, socioeconomic factors.

DANIEL TAN:      Absolutely. So as you're likely aware, lung function is determined by multiple things, genetics, environmental, social factors, and a lot of these factors which we may not necessarily be able to readily capture within a study. And so, moving towards a race-neutral approach has been one of the strategies that societies have put in place to try and address these inequities.

PAUL KING:        I think the thing is the whole thing of lung function and equations aren't perfect. But it does seem to be that different races have slightly different body builds and therefore probably do have different lung functions. And traditionally the Caucasians, have been felt to have relatively bigger lungs than other races. This has been potentially an issue as people from an African-American background in particular were felt to have roughly a 10 to 15 per cent lower lung volumes.

And the implication being from that is that African Americans may have been classified as having lower lung function when that wasn't actually the case. And therefore, they weren't diagnosed formally as having lung disease when they could have been. And rather than being racially-based, it's based really on their social and environmental factors. So, there's been a shift, but it hasn't been confirmed definitively and it may be that we need to have a more nuanced approach between individual patients and individual conditions, that needs to be assessed in more detail.

MIC CAVAZZINI:                And so to assess the impact of this change in equations in the Australian setting, the Monash Lung Sleep and Allergy and Immunology Lung Function Laboratory conducted this observational study using real-world data collected between 2014 and 2024. You included 6,100 adults between the ages of 40 and 70 who were referred with respiratory symptoms of clinical or a clinical suspicion of obstructive airway disease.

Now, there's a whole method section which I didn't understand, not being a clinician, about how spirometry was used according to the guidelines. Helen, can you take us through those measurements particularly the definition of obstructive versus restrictive conditions? Will these be familiar to any generalist listening or is there a bit of specialist knowledge there to share?

DANIEL TAN:      And perhaps Helen, you could touch briefly on what Z scores are as well and how we classified into obstructive and restrictive defects.

HELEN LUO: Yeah,. So essentially the way that we categorized these ventilatory defects into obstruction and restriction and mixed was through the values of FEV1 and FVC. Where the definition of an obstructive defect was defined as an FEV1 over FVC below the lower limit of normal. Restrictive was defined as an FVC below the lower limit of normal. And a mixed ventilatory defect was defined as both a reduced FVV1 over FVC and a reduced FVC below the lower limit of normal.

And the lower limit of normal was equivalent to the fifth centile. And the severity of these ventilatory defects was categorized into mild, moderate and severe based on these Z scores, which is essentially a reflection of how far your test results deviate from the predicted value of a person of your age, height, sex and plus minus ethnic group, depending on which equation that you're using.

MIC CAVAZZINI:                One question that came to my mind was when we're talking about the race-specific or race-neutral equations, is it just the cutoffs for severity that differ, or is there another sort of factor that multiplies through the algorithm that affects the output?

HELEN LUO:        Yeah, so with the variables of age, height, sex and race, we're trying to get a predicted value and with that predicted value we'll compare it against the actual patient's I guess lung function value and with that they can be categorized into each of these ventilatory defects and also categorized into a severity grading.

MIC CAVAZZINI:                Yeah, okay, so now getting on to the results. The main finding from this shift into race-neutral equations was that, in fact, there is a high level of concordance between the race-specific and race-neutral equations. 93% of the sample did remain in the same classification when race-neutral equations were adopted. And the 7% that were reclassified largely represented a decrease in the prevalence of ventilatory defects, particularly of the restrictive type. So, do you want to unpack where those came from?

HELEN LUO:        Yeah, so talking about the seven percent that have been reclassified, to break kind of break that down overall it was 1% that went from obstruction to normal, 4% that went from restriction to normal, and 1% from mixed to normal. And only 1% went from normal to obstruction.

To break that down further into each of the race categories. Starting with Caucasians, the biggest difference that we saw was from the prevalence of restriction from 8% to 4%, which is essentially a half of the original prevalence. Amongst Southeast Asians, the prevalence of obstruction went from 15% to 10%, which is around a third decrease. And the biggest change that we found in people categorized as Other was from restriction with a decrease from twenty-one percent to seventeen percent.

And if we're looking at the severity gradings of those that were classified as having a ventilatory defect, for those people that had obstruction, 19% of these people went to a lower severity grading, and the majority of these people were Caucasians and other. And from the people who were classified as having restrictive defect, 47% percent were reclassified into a lower severity grading, and this change was mostly seen on Caucasians as well.

MIC CAVAZZINI: Yeah so, 7 % might not sound huge on its own right. But given that around 3 million Australians live with obstructive airway disease, we're potentially talking about 150 to 200,000 people being reclassified to a less severe category. Do we believe that this would lead to less overtreatment or more undertreatment?

DANIEL TAN: That's a really good question and I think it can go both ways. one of the important things to realize here is that we're talking about nominal changes in a disease classification. So, we've not changed anything for these patients. They're just being reclassified using a different diagnostic standard and different reference equation.

So that to have such significant changes in the prevalence and severity with just a change in our interpretive strategy is considerable. That will have an impact on multiple factors, not just the diagnosis and the severity classification of these conditions, but naturally because we use lung function to help determine and guide our treatment strategies, also how we manage a very large proportion of these patients. It's relevant to note that a lot of the misclassification or change occurs typically around the boundaries of what we consider obstructive and restrictive. So it's individuals who are very close to or just under that boundary. and so these are the individuals who are most significantly impacted.

MIC CAVAZZINI:                And your paper does corroborate what's been found from studies in the US and the UK. You know, a multinational cohort study of nearly 400,000 adults from those countries showed indeed a reduced prevalence of non-obstructive intellectual defects in Caucasians by 69%, reduced prevalence of obstruction in Asians and other by 28%. If the premise for moving to race-neutral equations was to avoid bias against biased diagnosis in non-Caucasian patients, I found it interesting that actually the largest shift in classifications was in Caucasians and not those other racial groups.

DANIEL TAN:      Paul or Helen, did either of you want to try that one?

HELEN LUO: I’m happy for one of you two to answer that one.

PAUL KING:        Well theoretically we've had the initial problem well African Americans are excluded because their lung function was put artificially too low. Now assuming—it depends whether how much is due to social background, how much is due to race, but if it was mainly due to race, it may be that the people from a Caucasian background have an artificially inflated lung function now. So, therefore it won't be picked up when it's low because the whole threshold will have moved upwards. But that's not accounting for social factors, but you know, that that is something which could be borne into account.

I think one other thing to come back to. So when maybe that 87% of the people are reclassified, it should be remembered that most of the people that we have here are normal. So therefore, the actual reclassification of disease is bigger, it'll be six to sevenfold larger. Because approximately you know 75% of the population we screen were normal. So therefore, when you look at 25%, it actually is a third of them are altered so it's actually has a much bigger effect on our target population.

MIC CAVAZZINI:                Is this now standard in Australia that we are using the race-neutral equations

DANIEL TAN:      Currently Australia's in a bit of a transition period. there's a lot of uncertainty, particularly with these studies which have shown the significant impact that this change has on how we diagnose and classify lung disease. And many sites have not yet transitioned over. I think we haven't got a very good understanding yet of where the entire country is up to in in terms of this. We know some sites have moved over, some sites haven't. And that's actually the focus of one of the next studies that we're planning to do, which is a survey of of different lung function laboratories nationwide, which Helen will be leading to find out what different sites' attitudes and current practices are with regards to using these reference equations.

MIC CAVAZZINI:                And you described in the discussion a couple of the systemic knock-on effects where this adopted widely. So, the higher prevalence of restrictive defects in Southeast Asians could lead to more individuals being referred for complex lung function testing and chest imaging for potential interstitial disease. Lower predicted FVC values in Caucasians could increase the portion of idiopathic pulmonary fibrosis patients meeting criteria for anti-fibrotic therapy. How profound would these impacts be on the system capacity and costs and that kind of thing? And would that be a desirable outcome in terms of detecting the accurate diagnoses?

PAUL KING:        I think it could be pretty significant. As mentioned in the discussion that the low values in Caucasians could increase the proportion of them who have a diagnosis of idiopathic pulmonary fibrosis and alters, the treatment we give them, and the treatment's quite a significant treatment with some side effects, so that's a kind of big deal.

But once again we do tend to judge these patients with their other clinical features as well and their x-rays, et cetera. So, just to give you an example, we used to classify DLCO by another way, which is previous reference equations, and anyone who's overweight and female used to be classified as having a low DLCO. But we'd have a look at these patients and recognise that phenotype and then sort of account for it. But I think potentially it could significantly increase it, but it depend on who is reviewing the patient.

MIC CAVAZZINI:                You've already mentioned the known systemic differences between Caucasian and Black American patients. There was a mention in the introduction about how little has been investigated in Indigenous patients, in Australian Aboriginal and Torres Strait Islander patients. But with the limitations of your cohort, there were just there were just three categories Caucasian, Southeast Asian and other. How problematic was that binned other group for really teasing apart the racial differences, ethnic differences.

DANIEL TAN:      I might take this one. So, I think ultimately a as you'd mentioned, Mic, there's been a number of these studies done elsewhere and I don't think the findings from ours are necessarily entirely novel compared to what's been shown in other studies. What I think is important though is that the implications of these new reference equations will be very different in different populations, different countries with different epidemiology. And so, understanding the specific impact of these changes within each country and within each population is very important.

Ultimately, we work with the community and the population that we serve. And in in our particular catchment, which is a relatively large of over 1.5 million people, it predominantly consisted of these three main categories: Ao Caucasians, Southeast Asians, and Other. And I think it's not necessarily a weakness of the study that we didn't capture the other categories. It's it's just the makeup of our of the of the population that we serve.

MIC CAVAZZINI:                Yeah, and I mean, what should listeners take from the outcomes of this paper? It a confirmation that there should be wider adoption of race-neutral equations? Or…

HELEN LUO: I think with this study we're just trying to highlight where exactly the changes will be if we do end up shifting to these race neutral equations and for clinicians to be able to make an informed assessment of the patient in front of them based on these results. I guess an important next step going on from this study would be looking at more longitudinal data of these patients that have been reclassified and seeing I guess how they do in terms of their symptoms, the treatments they've got in and how this shift actually impacts patients longitudinally.

MIC CAVAZZINI:                Yeah, the paper wasn't trying to say this equation is better than that equation, but if we adopt the recommendations, this is the kind of shift we're likely to see. Keep an eye on this patient group and assess them with all your other clinical judgment.

HELEN LUO:        Yeah. Yeah, exactly.

MIC CAVAZZINI:                That is everything that I had sketched out. Have I missed any important aspects of the paper that you want to pick up on?

PAUL KING:        I think you’ve done a great job, Mic.

MIC CAVAZZINI:                I winged it, okay?

PAUL KING:        Once again, just to say just to say that that the population with disease it will definitely alter how they're treated. And we don't really know at this stage that that's right or wrong.

MIC CAVAZZINI:                Right, hence the need for longitudinal. So yeah, to zero in on that. It will affect the way some patients are treated. As you mentioned before, these are patients who are already on the margins of the categories of disease classification. So, is it an open question as to whether they needed the more intensive treatment or less intensive treatment?

DANIEL TAN: Yes. if I could add one last thing. Helen got asked a very similar question when she presented this at a conference about the implications of the work and I thought her answer was a very good one. Which is we don't particularly favour one equation or the other. It's important to understand the impact of them. But probably what's most important is that there's consensus and uniformity in how these reference equations are applied throughout Australia. And I think having an understanding of the current attitudes and implementation of these equations will be a really important next step.

MIC CAVAZZINI: Many thanks to Daniel Tan, Helen Luo and Paul King, for being so patient in explaining their important study to me. You can read their paper “Implications of race-neutral equations on interpretation of lung function in Australia” at the website of the Internal Medicine Journal, which all RACP members have free access to. The College is also proud to offer the Journal of Paediatrics and Child Health and the Occupational Medicine Journal, which are edited thanks to the generous commitment of members. I’ll provide a link to the other literature mentioned in today’s story at our own web page, racp.edu.au/podcast then click on the episode. There’s also a transcript if you need to check back on anything, and a comments section where you can keep the conversation about disease classification going.

To explore the other episodes in the [IMJ On-Air] series, click on the Episode Browser tab at the left of the website and then on the drop down menu for “podcast themes”. This also allows to search for episodes relating to specific domains of the Professional Practice Framework including cultural safety and clinical ethics that are mandated areas for continuing professional development. Please send me any podcast ideas or feedback via the email address podcast@racp.edu.au. The theme tunes you’ve heard today from View Points and Valante were licenced through Epidemic Sound. This podcast was produced on the lands of the Gadigal clan of the Yura nation. I pay respect to their elders past and present. I’m Mic Cavazzini. Thanks for listening.