C (3:38)

So, Beverly, I think most people understand obesity as just excess weight, and that is historically, traditionally how we define it. We use a metric called the body mass index, or BMI, to decide where obesity starts. And so normal weight is considered 18.5 up to 24.9. Overweight is the category of 25 to 29.9 BMI. And then obesity occurs at a BMI of 30 or greater. So that's been our historical definition. And I will say that there's a lot of discussion nowadays as to how that should be challenged and improved upon because obesity is so much more complex than just excess weight.

B (4:28)

Thanks so much. Now, I think we can all agree that historically, we in the medical profession have really foisted the responsibility for healthy weight control right on the shoulders of our patients. And that's something that we lament and try to move on from. And it was really considered a matter of personal responsibility rather than a biologically based issue that required medical intervention. But as advances were made in understanding physiology of weight control, the pathophysiology of obesity, I think we finally moved beyond the eat less, exercise more platitudes that we would launch at our patients. So it's also been very exciting to see how research has been progressing to the clinical stage and the eventual expansion of our obesity modifying treatments that are available. We've really needed, though, more guidance for obesity screening and treatment selection so that we can choose what's appropriate and directed for our individual patients, as well as guidelines for everyone as a whole. So the wide adoption and implementation of obesity screening practices is a good thing. Right? We're using bmi. We're seeing then the electronic health record. It's flagging us, so that should be good. So, Beverly, what are some of the high level advantages of bmi, why has it gained such prominence in obesity medicine? And again, this will be the intro as we discuss more of these issues later.

C (6:03)

So I start out saying that BMI has its challenges, and I think that is the primary conversation nowadays. But it's good to remind ourselves that BMI does have its advantages and there's reasons why it has persisted for decades in our clinical practice. And I think, you know, number one, it's not that bad when it comes to the data and its correlations with health. You know, we do see pretty strong correlations with higher BMIs and incidence of type 2 diabetes or hypertension or high cholesterol or heart disease. And the list goes on and on. And I think that the relationships are so strong that we tend to maybe over rely on BMI when we're looking at the individual patient. And that gets me to my second point, that BMI was originally designed as an epidemiologic screening tool to risk stratify populations. Not necessarily the individual patient you see in front of you in clinic, but really on a population level. So when we bring in this really convenient weight and height equation into our clinic, we're translating it to some extent and it's not a perfect fit. I think that the last piece as to BMI's advantage is a little bit of a chicken or egg point because BMI has been around for a long time. So all of our research, all of our knowledge, all of what we do and why we do it has been based on BMI as that metric. So moving away from BMI is very challenging because as we will see very soon, how do we interpret a body fat percentage number? How do we interpret a waist circumference? We don't know yet.

B (7:56)

So this dovetails nicely into this idea that as physicians, as medical professionals, we're trying to really focus on health for our patients. And yet in some ways, both our patients and the system seem to center weight. And BMI can further emphasize that as you talk about. Presumably then BMI is correlating though with health and making certain health assumptions. So what are the kind of assumptions that BMI is making when we classify based on it to overweight, obesity, severe obesity?

C (8:37)

There's kind of a few directions to explore there. I'll say. You know, number one, BMI looks at weight and weight encompasses a lot of pieces of us. Our body is composed of not just fat, but bone, water, muscle, organ tissue, kind of the whole gamut. Our body composition is not just fat. And when we talk about Obesity as a disease, what we really care about is the fat mass. And BMI's assumption is that as BMI increases, so does fat mass increase. Therefore, that extra fat is fit into that relationship with other diseases, comorbidities, mortality, et cetera. But we do know from seeing patients that even though we might have that relationship, it's not as bad as it seems. For example, athletes who might be doing a lot of strength training, they're having a lot more lean muscle mass, may sit higher on a BMI scale, but they don't actually have what we think of excess fat mass. There's also the question of quality of the fat. So too much fat is an issue. But also if the fat is the type that is pro inflammatory, prothrombotic, it sits around our viscera or our organs, then that type of fat, no matter how much it is, may also be contributing to metabolic diseases. So there are individuals who may have a normal bmi, so to speak, according to our scale, but they actually have an elevated fat fast that we're not capturing with that bmi.

B (10:19)

The discussion of fat and sick fat, functional fat, fat distribution, all of that is, are hot topics. And part of what makes it hot is that you already mentioned obesity is a disease. And that is something that a lot of us in obesity medicine have been trumpeting as progress beyond what we were saying before as obesity is a moral failing. But there are still some out there who would say, well, you know what, this is not quite the same as type 2 diabetes where we can really connect the dots from a pathophysiologic process to this phenotype. And so is obesity a disease that has been in the ether of the conversation, the scuttlebutt lately. There is this idea though, that whatever it is that we realize OVC to be and to define chronic illness disease, fat is an important part of that conversation. So if we think about then trying to figure out abnormalities of adipose quantity, functionality, distribution, then it stands to reason that BMI would not be perfect. You know, it does not estimate that. It does not directly measure it. It does estimate it in some surrogate ways that helps lead us to this current discussion that are there better tools out there that can easily screen, can easily quantify, can easily characterize adiposity? And so this is where they're talking about, well, what if we just directly measure it through this tool of DEXA scan so that we can look at lean body mass and we can look at fat mass. And so in this study they're talking about if we can measure and we can do it on many people. Can we apply it then in a way that is practical? Can we come up with meaningful categories that can then be used for treatment and evaluation decisions? This is the study, as stated, is seeking to provide percent body fat thresholds that equate to the overweight and obese categories. The study focus, as they state, is this analysis uses a large diverse sample of data to provide sex specific thresholds of percent body fat based on key obesity related comorbidities exhibited as metabolic syndrome. And then use of these measures of percent body fat may be more clinically relevant markers for metabolic health across the general population than body size, for example, bmi.

A (13:02)

And with that introduction, we have a good understanding of a lot of the controversies around BMI about how we classify and define and quantify obesity. And so I think it really sets up nicely why we might want some other way of defining that and understanding that. So we're going to look at the methods and as usual, we're going to go through them carefully to try to understand exactly what the authors are doing. We'll start with thinking about a couple of different aspects of study design. One of them is what the authors use here, a cross sectional study, but then also a bit more specifically, which is a correlation analysis, which is what the authors do here. First of all, they use a cross sectional study. So they use data from the nhanes, and we'll talk about that in just a minute. As a reminder, cross sectional studies get data from a population of subjects all at one period of time. So these are not longitudinal studies. They do not follow individuals over time to wait and see who is going to develop an outcome. They get all of their data at once. And then they look and they define some things, some clinical aspects as exposures and some as outcomes, and then looking to see the relationship between those two things. One thing that is relatively unique about the NHANES cross sectional studies, and I think that frequently causes confusions, is that the NHANES is a series of cross sectional studies. And so what they do is, is they are frequently repeating this analysis. And so that gives you data over time. And it's key to know it's not because they're following the same individuals over time, but it's because they're repeating the analysis over and over and over again. And so that way throughout the population in the US you can see how many of these things are changing, even if they're not following same subjects over that time period. So they are serial cross sectional studies. The analysis that's done here is primarily a correlation analysis. And a correlation analysis is really just looking at two of the aspects that you're pulling out of this here, a cross sectional study and looking to see how closely they are correlated. So if you're going to say if one of these changes, how much of that seems to be driving a change in the other one and you can get a sense of yes, this is a major driver, this is really the key aspect of it, or that seems to be a fairly minimal contribution. So that's really all that that correlation is telling you is how closely those two metrics are connected to one another. So now into this study in particular, so the inclusion criteria as we mentioned. So this is the US population that was included in the NHANES data set. Data was pulled from 99 through 2018. And as many of you all know, the NHANES provides a demographically representative sampling of the US population. It's collected continuously, but then is collated into two year data sets. They are looking at adults here and so they define that as 18 through 85. And they recorded data that, that we see here. So demographic information, as you would expect. So it's age, race, ethnicity, there's anthropometric data. So height, weight, waist circumference, bmi, which we've mentioned already, several biochemical pieces of information that are pulled, all about that metabolic syndrome that Andrea introduced us to already. We'll talk about that more later. But fasting glucose, triglycerides, HDL are included here and blood pressure is also measured and reported in this study. Beverly, another question for you. So DEXA scan is used here. That is how they determine percent by body fat. I think a lot of us as general endocrinologists will be very familiar with using DEXA for bone density. And I know you all in the obesity medicine field are fairly comfortable with using it for percent body fat. But, but help us general endocrinologists understand kind of how does that work? What are the pros of using DEXA to calculate that percent body fat? What might be some of the, the cons, the downsides to doing that?

C (16:36)

I guess on a higher level when we think about how to measure body fat, there's a few different ways. On the low complexity end, we have skin caliber tools, literal rulers that pinch your subcutaneous tissue to give you a measurement there. And that's probably our least accurate and least preferred way. And on the high end we use actually MRI machines to look at percent body fat because that really gives us a high resolution understanding of the compartment of where that fat is. Does it sit, you know, around your viscera or does it sit inside individual muscle fibers, for example? And that's mri. I would say the DEXA actually has been considered our gold standard for a long time for measuring percent body fat because it's easy to use or easy to administer rather, and it's a low amount of radiation and it still gives us information as far as compartments. So it'll give a total body fat percentage as well as perhaps visceral body fat and then other specific locations. Most clinics nowadays that offer body fat percentage measurement usually use a machine that uses bioelectrical impedance analysis and that is a different method that doesn't use radiation at all, but as in its name, electrical activity. In order to measure the different compartments, water, bone, fat. Some of the fancier machines are able to give a little estimate as to where that fat component might be. Again, does it sit around the hips versus the waist for example? But most bioelectrical impotence analyzers really just give a total body fat percentage. Those are a little bit of the pros and cons, where the DEXA is maybe more user based does give us that compartment information, whereas the BIA is the acronym for that, that BIA might be more accessible to a general clinician.

A (18:46)

Those are the inclusion criteria. The exclusion criteria, as alluded to before, is for any years where they were missing those DEXA results, they did not include data from that. So that would be the 2 year increments of 07 and 08 and then also 09 in 2010 as a reminder of how the metabolic syndrome is defined. And that requires three or more of the following. So you have to have a waist circumference greater than 40 inches for men and great than 35 inches for women. HDL has to be less than 40 for men and 50 for women. Fasting glucose 100 or above, a blood pressure above 130 over 85 and finally triglycerides that are 150 or above. Finally, we'll end with the analysis that the authors use here and, and I'll make an editorial comment is that very little of this information was in the methods itself. So I had to go through the results and actually a bit into the discussion as well to try to understand exactly what the authors were doing in in their assessment. Think I've got it figured out and so I'm going to give you how I understand it. And so again, the authors here are trying to understand some prevalences. So how they start is they start with looking at the BMI of their patient population. So they then look at the cutoffs for overweight and obesity so that bmi greater than 25 and greater than 30 and figuring out the percent of individuals who have metabolic syndrome based on each of those cutoffs. And then after that, they go back to see similar cutoffs of percent body fat based on those percentages of metabolic syndrome. And then with that, they're able to compare the prevalences of metabolic syndrome to that defined by bmi. So it was a little bit convoluted, but a little bit backwards and how they kind of come up with that. So, Beverly, need your thoughts on this. So as I was wrestling and trying to understand this a little bit more here, I found it kind of interesting because a bit of the argument we're going to get to this in the discussion is that BMI is not a great indicator of percent body fat. We need a more direct measurement. And yeah, when they come up with this, you first start with BMI and then you do. You go backwards and make calculations from that until you end up with equivalent cutoffs for percent body fat. So I was curious what your thoughts are with this method is based on cutoffs for bmi.

C (21:03)

I'm with you, Chase. I was a little confused when I read the title of this paper. I was really expecting more of that longitudinal outcomes prospective study to give us body fat, different cutoffs of body fat and incidence or prevalence of subsequent comorbidities. What they did was not that. What they did is exactly as you say. They are still anchoring their cutoffs to our current BMI thresholds. And I think ultimately what they wanted us to understand was that their criteria for obesity based on body fat should be something like that 35%. And I know there are specific ranges for men and women, but something like 35% body fat or greater should now be the definition, quote, unquote, for obesity, because 35% is associated with a 35% risk of metabolic syndrome. But they got there through the bmi, which I think makes their point moot as far as trying to get away from BMI.

A (22:15)

So we'll @ least take what the authors did here as a starting place for then where they go from there in their analysis. Once they did that analysis and they figured out okay with using those percent of how many people in each of these groups, groups have metabolic syndrome, and making those the cutoff and then comparing BMI and percent body path cutoffs with that, they then wanted to see, well, how closely are those correlated to one another. And that's where they, where they do this correlation analysis that we referred to before. Now one thing that I found was interesting is that they described this as a correlation analysis and they used a scatter plot. Andrew's going to describe this to us in a little bit in the results. Interestingly, they didn't actually do a formal assessment of that correlation. So typically you can quantify that. You can give that a number. So yeah, there's a picture, you can look, you can see the dots, you can get a sense of how much those overlap and the line that's generated from that, the linear relationship or perhaps a nonlinear relationship. But it also comes with numbers so that you can actually quantify that. Understand that a bit more. And I was surprised that there was not any number, any quantification of that correlation that was provided. The authors, as we'll see in the results, instead point out that their data, they talk about the spread of the data and that it has a nonlinear relationship, but again, don't quantify that at all. So I've gotten ahead already. So we're going to, we're going to hand it over to Andrew now that I've already talked a little bit about the results, but he's going to walk us through some of the numbers and then describe those results that I've alluded to already.

B (23:41)

Andrew, I will go ahead and present the results from the paper. So we're looking at almost 17,000 participants who are included in the final analysis. Approximately 8,700 of them are men and 8,200 of them are women. We see that the ages for the men and women are similar, so around 42 years old. And the BMI is also similar about 29 for men. And for women, the percent body fat followed expected sex differences with the average percent body fat for the cohort for men of 27.8 and 39.7 for women. And then looking at metabolic syndrome prevalence for men, it was 38.4%. So higher than what it was for women, which was 29% for the whole group. If you took the whole cohort, 34% of that group were noted to have metabolic syndrome. Looking at self reported race and ethnicity, we have about 40% of the cohort, white, 27% Hispanic, 21% non Hispanic, black, 8% non Hispanic, Asian, 4% non Hispanic, multiple and then no answer provided by 0.4% of the cohort. So Beverly, any things that stood out to you in terms of the study groups or the comparisons?

C (25:15)

Not so much. I think it's well representative of our US Population and also gets at at some of the disadvantages sometimes when we extrapolate from pharmacotherapy trials, for example, where there always tends to be more women than men and very Caucasian as well, at least here we have a little bit more of an even distribution. And the fact that the N is so large means that even if we have smaller percentage of non Hispanic black, for example, compared to non Hispanic white, then at least we're still well represented.

B (25:51)

It would be helpful to look at what kind of results were they looking at calculating overall. And so ostensibly they wanted to use metabolic syndrome as a surrogate of health impact, health risk. And so then they chose BMI definitions of overweight and obesity and said, hey, what percentage of these groups then have metabolic syndrome? And so then what they found was that those in the overweight category had a metabolic syndrome risk of 5%, so mildly elevated risk. And then they found that those in the obese category had a metabolic syndrome risk threshold of 35%, so significantly elevated. Then they took these risk thresholds and correlated them to percent body fat and then used these data to create percent body fat definitions of overweight and obesity. So they were looking to see, well, what percent body fat do you have to have to then have a 5% metabolic syndrome risk and the same for 35%. And then they'd say, then we're gonna classify that as overweight based on body fat and obese based on body fat. They wanted to compare the ability of BMI to predict body composition as defined by directly measured percent body fat, both at the group, but also at the individual level. Essentially they wanted to see, are there false positives and false negatives? Because ultimately they're trying to get at are we both over and under treating using BMI and can we treat more accurately using direct measures of body fat percentage? When they looked at metabolic syndrome and BMI, as mentioned, in the overweight category is 5% prevalence of metabolic syndrome, 35% for the obese category. And they calculated the receiver operating characteristic and area under the curve. And we could probably talk more about what that means in a bit. But for men that was 0.83, and for women that was 0.75. Then looking at the metabolic syndrome percent body fat, they said, well, for 5% metabolic syndrome risk, the threshold of body fat percentage for men would then be 25%. So if you have 25% body fat and you're a man, then you would have a 5% risk of metabolic syndrome. And for women, it'd be 36%. Then at that significantly elevated risk threshold of 35%, men would have to have a body fat percentage of 30% and women 42%. And then looking at that receiver operating characteristic and area under the curve, that would be 0.8 for men and 0.71 for women. Then they compared categories of healthy, overweight, obese for men and women, looking at BMI and percent body fat. And we can see the exact numbers. But the summary would be if you use BMI as your criteria, you would actually designate more men as healthy compared to body fat percentage designations. Overweight would be about the same and you would designate fewer men as having obesity versus percent body fat. Now, for women, if you look at it, what they'd say is that if you're using percent body fat, that fewer women would be designated as healthy, you'd have more designated as overweight, and you'd have a similar amount between BMI and body fat percentage. That would be in the obese category. Then they looked at the relationship between BMI and percent body fat and they developed the scatter plot that Chase had already referred. And if there was portrait of correlation between the two, you just see a linear straight line, but you do not see that. Instead you see what they describe as a curvilinear relationship. And they're saying that their findings show that there's a large scatter seen. And based on that, and again, without any more technical calculations, their conclusions within the results section would say that there's poor predictive value for BMI for adiposity at the individual level. So even though they seem to correlate as a group pretty well, IFAT and BMI seem to match fairly well, they would say, wait, look at this scatter plot. What you'll see is that there are all these individuals who do not quite fit as a good correlation. And that also the scatter plot of course shows that there are systemic sex differences seen. And they would talk about, you know, if you look at this scatter plot, you can see that there's a group of women with BMI under 25 yet have a high percent body fat and some over 42%. And so actually by BMI they should be low risk, but by body fat they're actually high risk. And therefore we would be under treating this group of women and underestimating their risk. Similarly, for men, you could see a group with a bmi that was over 30 and yet their percent body fat was under 30 and even under 25. And this could be a group that we would over treat.

A (31:32)

All right, we'll now go into the discussion and full marks, Andrew, for describing that. And it's always difficult to take the visual data here and try to describe that to you all. You did excellent job there. So what we're going to do is start with a couple of quotes from the authors and how they summarize their work. So a summary to start with where the authors say this analysis suggests percent body fat thresholds for patient guidance. These metrics align with overweight and obesity in a large nationally representative sample. And a second statement that I would describe or I would label as an implication from the authors where they say obesity related diseases may be be more effectively managed by finally moving away from anthrocometric estimates of adiposity to direct measurement of the fat component. And I want to get input again here from Beverly and Andrew on this. And I think it would be fair to say that the authors are at least suggesting, and they don't state this explicitly, but I think it would be fair to say they suggest that measuring percent body fat is superior to BMI in clinical care. However, I'll point out as we've mentioned already, that the study was not designed to determine that. So really what did they were focusing on how correlated are BMI and percent body fat and they point out there's some correlation but individual variability again at least based on the the visual pattern that that was seen in that scatter plot and the authors actually suggest, as Andrew's already told us, is that the receiver operating characteristics for predicting metabolic syndrome are actually fairly similar for BMI and for percent body fat. So Beverly, let's, let's start with you. I'm interested in your thoughts on on this. Again I'll say a suggestion from the author and do we think that this supports this? You alluded to this idea before already, but do we think this data actually furthers that argument?

C (33:17)

Maybe this is an overreach and not quite fitting in this particular study. I think there's a lot of value in this study of course because it does examine these thresholds for body fat percentage and where those body fat percentage thresholds would correlate with that 5 and 35% risk of metabolic syndrome as Andrew had mentioned. I think what's really interesting as a question that it brings up is that S curve they have in figure three and I know this is an audio only podcast, but if you think about any S curve really most of them have very low lying slope in the very beginning and then it becomes steeper and steeper in the middle, which goes to show that small incremental changes in whatever factor you're looking at, and in this case, body fat percentage, small increments in body fat percentage really correlates with a massive change in overall risk. So when these authors are suggesting a body fat threshold for men being 25% overweight, 30% obesity, you're looking at a 5% difference between those two numbers. And I gotta tell you, with DEXA and BIA nowadays, we can see that difference from one measurement to the next, depending on a person's water content, depending on their hydration status, actually. So I think we have to be a little bit careful with how we define those thresholds and perhaps need other measures to be more accurate into how we interpret that.

A (35:06)

All right, Andrew, thoughts from you on this suggestion from the authors and how they might have come to this conclusion.

B (35:12)

There definitely is a lot of generous interpretation of the data that they use. It is interesting to look at the va. It sounds like they want to incorporate more DEXA in general. And so then they're proposing this as a good and more accessible tool. And that's not exactly the question that you have, but it is something that I think about in terms of there. There's a desire to want to make this be a good tool. And I think that clinically we find that there are those individuals who don't quite fit into where BMI would predict risk. And trying to then use this tool as a marker makes sense, patently makes sense that we would try to get a better sense of it. But as Beverly's alluding to, and as you were, Chase, it is surprising that the overall data for the population seems similar for fat percentage versus bmi. And then I have to wonder how much better is this versus the clinical acumen of the obesity medicine specialist? So in a busy clinic where you're in primary care and you're seeing people quickly and you don't have have a lot of time to devote to obesity care, maybe using BMI a lot to make some reflexive decisions. But for those who are having a concentrated visit on obesity medicine, one wonders how much of an advantage the imaging would have over your own clinical assessment of their body habitus.

A (36:49)

The authors then go on to highlight that the relationship between BMI and percent body fat is particularly weak at upper and lower ranges of bmi. And we've talked about this already, and it was alluded to, with athletes, for example, being one such example of that. To wrap up the discussion, the authors do not list any limitations of their work but they do go on in their conclusions to make a couple of statements that I'll quote. First of all, although BMI can be helpful as a first level screening criteria, it is not an accurate method for determining body composition and in fact does not provide accurate information about fat and lean lean components. And then second, currently available methods such as DEXA provide estimates of intra abdominal fat and other information about soft tissue lean mass with emerging relationships to all cause mortality. Okay, so we're going to wrap up here and Andrew, let's start with you. I want to get your thoughts on just the quality of this report. Overall. You spent quite a bit of time going through this and really trying to understand exactly the approach that the authors used here. So what is your sense of, of the quality of this report? Overall?

B (37:52)

I was happy to see the work. There's a lot going on in terms of better characterizing the risks of weight for our patients and what are the best ways to approximate risk. Is it all weight? Is it body composition? And so it's good that these questions are being asked. I do take issue with some of the conclusions drawn being derived from derived primarily from a visual interpretation of a scatter plot and would like to see that better quantified and better discussed into how that would be implemented in clinical practice in a better way. I think that the way they designed it based off of BMI to construct the whole study was confusing and I would like to see a differently designed study to help better answer this question about how percent body fat could be used in clinical practice.

A (38:52)

Beverly, same question for you. What are your thoughts on the quality of this report overall?

C (38:56)

I think people really need to read this report in full. I do not think that body fat can replace bmi. I don't think this supports that statement at all. I don't think think we can define overweight and obesity by percent body fat instead of BMI with this report, given that this report uses BMI to anchor those definitions of obesity and overweight via body fat. So I think it's a very worthwhile read because it really demonstrates kind of the questions that we need to be struggling with as far as body fat threshold, underdosing individuals based on BMI or overdosing individuals based on bmi. But it's definitely not the study that would change clinical practice.

A (39:48)

Let's think about that last part just a little bit more about if it should change practice. And I think some, we've all stated some concerns and reservations about whether they should do this, but help us put it in the context of other discussions that are going on. Andrew, you and I touch base about a larger discussion about how do we need to define percentage body fat? How does that contribute to definitions of obesity and overweight? So help us understand that larger discussion at a high level and how maybe this information might fit into that.

B (40:18)

It's being much talked about, the recent commission that's through the Lancet diabetes and metabolic work that they're doing and trying to understand obesity in a better way to help define it for individuals and also to help us in clinical practice and to try to get around a completely weight centric way of thinking about obesity. There has been a call to then for our patients, both get weight, height and calculate bmi, but then do some other measure that would approximate body fat. And that could be, as Beverly was talking about, bioimpedance, it could be skin calipers, but it could be as simple as waist circumference and weight to hip ratios. And so that is where it is going to be interesting to see. Will this move the needle in terms of accessibility and coverage of things like adexa, which are not trivial in terms of cost and will that provide better care for our patients? Because. Because there are downsides to performing some of these other things. For some patients, doing waist circumference is uncomfortable. For some people, they're not trained well or medical assistants or even physicians about how to do that properly. And certainly the time that it would take to do a DEXA that in addition to the cost, doesn't necessarily warrant widespread implementation. However, it's certainly if we're looking at some of these patients on the margins where we're really struggling with how we should classify them, then this tool provides valuable additional information. Whether that's better than just waist circumference again remains to be seen.

A (42:07)

Beverly, last question for you. So you don't see the data here really driving a change to use of percent body fat, but give us a sense of what information we might need then to be able to really move the field in that direction. If that is in place where we're.

C (42:21)

Headed, I want to see real outcome data, real longitudinal outcome data where we enroll that one person, thousands of people, but we follow that same person over the decades, measuring their body fat, assessing their risk and incident of diseases that would be able to inform us risk thresholds with body fat percentage without relying on vmi.

A (42:51)

And with that, I would like to thank Andrew Craftson and Beverly Chang for joining me for this month's edition of Endocrine Feedback Loop. I learned a lot and know that you all did as well. Please join us again next month. And now you're in the loop. This has been Endocrine Feedback Loop. Endocrine Feedback Loop is brought to you by the Endocrine Society with Production oversight by Brandy Brady Brown and Andrew Harmon. If you want to like and subscribe, you can find us on Apple, Spotify, or wherever you get your podcasts. We'd love to hear your feedback on this episode or the podcast itself. Please email us@podcastron.org Endocrine Feedback Loop is a free service of the Endocrine Society. To learn more or to become a member, visit the society's website at www.endocrine.org.