#599: Does Unprocessed Red Meat Increase Diabetes Risk? – Gil Carvalho, PhD MD & Mario Kratz, PhD

Hello and welcome to Sigma Nutrition Radio. This is episode 599 of the podcast. My name is Danny Lennon. You are very welcome to the show. Today we're gonna be tackling quite a complicated but very interesting question in the field of nutrition science, looking at a exposure that can be quite tricky to get some clear evidence on for some reasons I'll explain in a moment. And trying to discuss the current evidence we have with two of the people I think do a really good job of reading and interpreting studies and coming to some fair conclusions. We're gonna be talking about the exposure of unprocessed red meat. And in particular, we're gonna be looking at two main areas of outcomes, one that relate to diabetes risk so that could include anything from instant resistance, glucose tolerance or actual diabetes onset. And then we'll be looking at some of the cardiovascular outcomes so particularly development of cardiovascular disease. What we're not gonna be looking at is any of the other outcomes and of course there are many that we could look at. Most notably cancer and I think there's a few reasons that we'll be discuss, one that there's maybe a bit more clear evidence in that area. Second, you'll note the exposure we're talking about is specifically unprocessed red meat as opposed to processed meat. I think the evidence is much clearer in relation to processed meat and the increases in risk we see with a variety of chronic diseases. And so we've covered that on some other episodes where there's much more debate within nutrition science and just more broadly of people worrying about health effects relates to unprocessed meat. Now like I said, there's a lot of different outcomes we could look at. In another episode, we're gonna look specifically at some of the cancer outcomes because that's a whole field of evidence that needs to be looked at in and of itself. But today the reason we're looking at the areas related to type 2 diabetes and cardiovascular disease is that the evidence is a bit less clear. And so for that reason, there's a variety of different interpretations of that evidence that different people come to. In particular, what is probably most difficult is when we take this exposure of unprocessed meat and we put a couple of caveats on that which we'll discuss in this episode. For one, placing that in the context of an otherwise healthy diet, so that includes plenty of fiber and fruits of vegetables, et cetera. And then the second is if we account for the saturated fat component of that unprocessed meat by focusing in on leaner cuts of meat, then again, that adds another bit of difficulty in trying to get some clear and consistent effects that will inform a strong conclusion. And so there's plenty to work through. And so to help me do that, I'm gonna be talking with Dr. Mario Crats and Dr. Gil Carvalho, both of whom have been on the podcast previously, looking at a number of different topics. Both of them are really well-established scientific communicators in the field of nutrition and often discuss interpretations of studies and evidence, et cetera. Dr. Crats runs nourished by science, which is a YouTube channel as well as a free website that goes through a lot of this evidence-based information. And previous to that, he was an associate professor in the Cancer Prevention Program at the Fred Hutch Cancer Center in Seattle. He was also involved in the Department of Epidemiology and Medicine at the University of Washington. And during his tenure in academia, he conducted a lot of clinical research specifically on interactions between diet and chronic diseases with a major focus of a much of his work being on type two diabetes and cardiovascular disease. And so that will become particularly useful as we start working through some of the studies we have to date. Dr. Gil Carvalho also has a YouTube channel where he discusses the evidence relating to a whole range of health topics, much of that including nutrition. His YouTube channel is Nutrition Made Simple and he is previously being on the podcast as well. Remember, if you are a Sigma Nutrition Premium subscriber, you'll be getting detailed study notes to accompany this episode. You'll also have a key idea segment that will play after the episode finishes. And you'll also get a full edited transcript to this episode. If you are listening on the free public feed of the podcast and you are interested in getting these extra educational resources to really help you get more out of your podcast listening, then check out what we offer with a Sigma Nutrition Premium subscription. It gives you these resources that I've just mentioned as well as a premium episode every month. I'll put a link to that in the description box where you're listening right now. Additionally, I will also link up to the studies that we mentioned in this episode. So if you click through to the episode page, all of those will be pulled up as well as where you can find our two guests on the internet and all the other resources related to this discussion. So that will be in the description box where you're listening, click through the episode page and that will all be there. But for now, please enjoy this conversation that I'm having with Dr. Gilkara Value and Dr. Mario Crap. (upbeat music) Yo, today I am joined by two guests, both of which are being on the podcast before and are two of the people who I perhaps most respect in terms of how they interpret nutrition science research. And so that's why I've asked them to come and help me with a topic where there's plenty of different opinions, various different position points around a debate. We are gonna be talking about the exposure of unprocessed red meat and we're gonna be looking at some of those outcomes. I'll talk about that in a minute, but first of all, I give an introduction to each. So first, a big welcome back to the podcast, Dr. Mario Crap. Thank you for coming and doing this. - Thanks so much, Danny. - And also to Dr. Gilkara Value, welcome back to the podcast. - Thanks, Danny. Thanks for having me. - As I've mentioned, we are gonna be talking about unprocessed red meat and specifically, we're gonna be looking at two general outcomes, noting that there are others we could look at, probably most notably cancer as an outcome. That is something we will discuss on a separate podcast because it is a particularly important outcome that relates to this exposure. But our focus is on type two diabetes risk and cardiovascular disease risk. So our focus is on these cardiometabolic outcomes today. So maybe a good way to get us started and maybe I'll ask first Gil for you to give your thoughts here. When we think of this as a exposure around unprocessed red meat, this is one where we see a variety of different interpretations, a real lack of universal consensus around how we should think of this exposure. For you, some of the reasons why you think that is, what is it that makes this particular exposure have much of this different debate and maybe lack of universal consensus? - Yeah, it's a great question. I think there's two main reasons, it's the scientific category of reasons and then there's the maybe cultural category of reasons. So I typically, I think there is legitimate areas of uncertainty, which I think are to some extent common to many single foods and the lack of long-term randomized trial data looking at specific food variation and heart outcomes. So just changing amounts of red meat in a randomized trial lasting several years and looking at actual heart attacks or strokes or type two diabetes onset. That would be gold standard level evidence. We don't have much of that, but I would say we don't have much of that for the vast majority of single foods. In fact, it's hard to think of a food that's an exception where we have great evidence at that level. So we end up having to infer from observational studies shorter-term randomized trials that look at biomarkers and then sort of dietary pattern randomized trials and mechanistic data. And so because there is this indirect aspect in this inferential aspect, there is a level of uncertainty and a level of interpretation that goes into it. So that's sort of a bird's eye view, the scientific side of the uncertainty. And then there's all kinds of cultural aspects of why this is controversial and why this is very important and very personal to different people whether red meat is diminished or increased in the diet and people have different views and very emotionally powerful views in both directions. So I think all of that comes into play when we are talking about social media and sort of the public discourse. And in a similar vein, Mario, for you, why do you find this a particularly interesting question or an exposure to look at? - Yeah, I think probably for me, the main aspect that really has fascinated me for a long time is that we have basically, particularly on social media, two diet camps that are literally opposed to each other, right? We have on the one hand carnivores or ketogenic eaters who swear that eating more meat or eating only meat as in a case of carnivores that's come with major health benefits for them. And anecdotally, many people even sharing that that has led to massive weight loss or even reversal of autoimmune disease and like these major health benefits are often shared and I think that has led to some a immediately small portion of the population but a clear vocal one on social media to be very convinced that this demonization of meat and maybe animal fats in general is just entirely wrong that it's really the other way around. And then we have people who for at the medical reasons don't consume meat, right? Who are vegans or health reasons don't consume any meat or animal products. And they similarly feel very strongly that is the source of their great health. That avoiding these foods is really what is responsible for them being so healthy. And so in this interesting setting, looking at the data and then realizing there's a lot published in this, and it's actually from a scientific perspective, most of the reasons that Gill outlined really difficult to find out the truth, right? So to stick with these particular examples, does actually eating more unprocessed red meat increase our risk of developing type 2 diabetes, or other things being equal, and does it increase our risk of having a heart attack or a stroke? So I've spent quite a lot of time with these data, and as I'm sure we'll get into, it's actually not an easy to answer question. So maybe off the back of that, Dr. Krat, why has this been something that has been viewed through that lens of type 2 diabetes risk? What have been some of the proposed hypotheses or the typical potential mechanisms that have been our starting point for thinking about unprocessed red meat as something that could impact risk? Yeah, there's definitely a number of different mechanisms through which regular consumption of unprocessed red meat could quite plausibly cause glucose intolerance, right? Or any of the major determinants of glucose intolerance, so they could cause insulin resistance, or beta cell dysfunction. And these include, and I'm just going to pick a few because an exhaustive discussion of this would probably fill the entire time we have. So when we think about red meat, an obvious thing that comes to mind right away is that it's food that's fairly rich in fat unless we pick very lean types of red meat, but we have some fat in there. And the fat tends to be about 50% or so in the form of saturated fatty acids, including long chain saturated fatty acids. And there is some literature suggesting that regular consumption of a diet rich in long chain saturated fatty acids may use in sort of a distance. A lot more to say about this. I've actually published one of the papers that has shown this. Personally, I think this is only the case. It's very high intake doses, and I'll leave it at that, but this is at least a potential plausible mechanism. Then there is, you know, when we think about how red meat often is prepared, it's very often prepared at high temperatures, right, grilled, broiled, fried, and so forth. And that leads to chemical changes in the meat, specifically the formation of a class of molecules that I actually wish we studied more. And this is what we call advanced glycation end products. So these are basically proteins or other molecules to which a glucose molecule has attached itself under the influence of the usually high heat, right? When you broil something or you grill it at very high heat. And advanced glycation end products are really interesting in the body in that they act through a receptor called rage, I think, the receptor-ferred-advanced glycation end products. And through that has a number of functions, including being pro-inflammatory and actually inflammation, even low-grade inflammation, is a very potent inducer of insular resistance. So this is certainly another potential mechanism that is not implausible. Then there is data suggesting that higher plasma concentrations of branched-chain amino acids could play a role in the development of buccos intolerance. And red meat is fairly rich in these branched-chain amino acids. Now whether the dietary intake of these branched-chain amino acids actually is a major determinant of the plasma concentration of these branched-chain amino acids, at least to my knowledge, interrupt me guys. If you feel like you have something to add here, it's not like I've just read up on all the literature in these areas, but I would say it's not entirely clear to which dietary branched-chain amino acids is the main determinant of plasma branched-chain amino acids. So that is another potential mechanism though. And maybe the last one we could mention briefly is this molecule TMAO. We could also touch on when we talk about C cardiovascular disease. So TMAO is basically a molecule that's quite interesting and how it usually is getting into our blood. If we consume meat, there are molecules in there, like carnytein, for example, that by our gut microbiota like it changed and then it changed again in our liver to, in the end, TMAO and TMAO has some, I'd say suspected glucose intolerance-inducing properties. And it also is a molecule that's potentially relevant for cardiovascular disease. For me, these are potentially four mechanisms that are, I'd say, none of which has really been disproven. They're all hypotheses at different stages in having been supported by data, but there's something to support each of them, I'd say. So from that perspective, I'd actually say it is somewhat plausible to think that consuming unprocessed red meat regularly, even though it's not a high-carb food. And many carnivores will point this out. How can red meat be a cause of diabetes if it doesn't have any carbs? Obviously, sugars and refined grains are the evil. Yeah, of course, sugar and refined grains are also not particularly healthy foods, but that's not to say that something else could act through different mechanisms to potentially induce glucose intolerance, right? So we have this biological plausibility through a few different mechanisms, which there is work that is done on this, which would speak to this mechanistic support thus far. So then from there, our next step is to look at these other lines of evidence that might get us closer to the answer that we're looking for, Dr. Caralliou, how do you think about looking at that actual outcome of let's say risk of type 2 diabetes and overall nutritional epidemiology, typically we might look at. Or if we're going down maybe some more control trials and looking at things more specifically like glucose tolerance, and then we can we see some consistency there? From an overview level, what is your sense of what we might have with this evidence base related to either type 2 diabetes risk specifically, or maybe some of the markers that you would look out for that might be informative? So I'll just summarize real quick the observational evidence and I'll let you guys weigh in on other details or weaknesses. And then if you want, we can move on to the randomized trial data that I've seen, look at biomarkers. So basically for cohort studies, what I've seen in meta-analyses is about a 10 to 30% in that range increased risk of type 2 diabetes per daily serving. And daily serving is around 85 to 100 grams, depending a bit on the publication and the definition. But so the largest and I believe the most recent meta-analyses I've seen, this is a systematic review and meta-analysis of 43 cohort studies published in 2023. And they reported a 27% increase in type 2 diabetes risk associated with the intake of unprocessed red meat per 100 grams daily. And then a 2X increase in gestational diabetes. So that's kind of the bird's eye view. Sometimes it's a little lower. Sometimes it's closer to say 15% in that range. And again, this is for daily intake. When you get to more infrequent doses weekly or something like that, then obviously the magnitude of the effect is smaller. So this is the summation of what I've seen in terms of the effect reported in observational data. I'll let you guys comment before moving on to randomized data. There's a paper in glance at diabetes and technology, similar types of figures that has a ratio as 1.1 per 100 grams per day. That was across multiple cohorts as well. Not that that's the definite amounts or certainly fits in with the same sizes you're seeing. Mario, from your reading of the epidemiology of this area, what is your sense of the overall? Well, I'd actually like to have a little bit of a discussion about this because I have made some really curious observations. So I have had it really over reading meta-analyses, which is very much to my detriment to anyone who follows my YouTube channel, because I'm very slow at producing, because I read a lot of papers. And so I actually read many papers in this area from the big cohorts from all around the world. And one thing that I noticed is, first of all, the big QS cohorts, the nurses health study, the nurses health study, too, and the health professionals follow up study. They have published numerous papers, basically, on the same cohort. So one concern with this is that there is one potential, major issue with these data. And that when the study was run, red meat was not encouraged by the dietary guidelines, right? And so what we find pretty distantly is that people who consume the most red meat tend to have an unhealthy lifestyle, exercises less take less multivitamins, all of these things. These are potential concerns, I'd say. But the bigger one is that-- and I find this very curious, particularly in this cohort, but in a few other cohorts as well-- people who consume the most red meat, including unprocessed red meat, usually have, by far, the highest total energy intake. In some cohorts, 50% more than those in quintile or quartile one. And that's a big concern, because the way these authors did this is that they usually parameterize red meat intake by looking at servings, say, per day or servings per week. I think that's a result of this, because obviously, anything that will have calories-- if you eat more of it, it will contribute to a higher calorie intake, right? So I think, personally, a better approach would have been to do a nutrient density adjustment there, if it's so obvious. In other words, red meat intake as a percentage of energy or as servings per thousand calories or something like that. I think this is a concern, because in these analyses, when we look at the model one, usually, so model one usually being adjusted only for age and sex, we often see a hazard ratio, say, for quintile five, compared to quintile one of 1.8, 1.7, or something like that. And then we adjust for a number of potential covariates, confounders, and other covariates. And then it shrinks to something like 1.1, right? So that's a major, major reduction. And what that, a worry that this triggers is that very obviously red meat intake is very strongly associated with other variables that also are determinants of incident type two diabetes. Such that when we adjust for them, we have this very dramatic attenuation of the strength of the association. Now we could argue, well, that's all fine, that isn't that normal. Well, the concern here is that if the exposure variable red meat intake is so strongly associated with a total calorie intake, that's a concern because we cannot really adjust well for total calorie intake. That total calorie intake, as assessed by a food frequency questionnaire, these cohorts is highly unreliable and highly biased. And we know this from a number of studies, including from the Harvard group that have published this. And I'm really close to this type of line of research because at the Fred Hutch-Kentzen where our last work, there was a whole research group that actually studied this and they developed a biomarker that was basically also a calibrated energy intake that used the food frequency questionnaire derived energy data, but calibrated it for a number of factors. That was based on the small W-labeled water study that they did in this group of the cohort. Pretty elegant work. And I'm happy to share that with you guys or the audience if you're interested in. But basically what they showed is that if they look at a cohort, such as, say, in their case, the Women's Health Initiative, they also do find fairly substantial associations between red meat intake and, for example, type 2 diabetes, those heart disease, cancer and so forth. When they adjusted for the food frequency questionnaire reported total energy intake that is attenuated but is still significant. But when they adjusted for the calibrated measure, most of these associations become null. And so that's a concern to me because we're not having the calibrated energy expenditure measurement for these other cohorts. And when I see the degree to which the exposure here is associated with total energy expenditure that raises some concerns. Now, that's at me. These reported associations are incorrect. No, but it reduces my confidence in these observed associations. And it makes me want to see other data to make me come to the conclusion that red meat is potentially effective. It could be useful because it's not so. Does that make sense for you guys? Really interesting. And I'm sure Gail maybe some thoughts on that. The first thing that was coming to mind is one thing that you've touched on. I think is pretty consistent across all the outcomes as we do see those differences depending on where that cohort is. And for a variety of reasons, for example, in those new S cohorts we have a different type of exposure. Generally, the total amounts tend to be higher for red meat. That can be useful for that exposure contrast. But the other aspect that I think relates maybe to what you said, Mario, is that we have a specific type of baseline diet here that needs to be taken into account, particularly when we're doing an adjustment model. So if we are doing an adjustment for total calories, then the baseline diet of that specific cohort becomes important in the same way we would think of a comparator food in an RCT. And so this is where, again, it relates to those factors you said up what tends to get grouped in with those who have the highest in takes of red meat. Also factors in when we do an adjustment for things like total calories. But what you know there is really interesting about how we think about the adjustment for calories and are we actually doing it appropriately? Do you know if you've been thought so came to mind as the Mario's gone through that? The way that Mario phrase it in the end, the way he summarized it, that these caveats lower his confidence, I would completely subscribe to that as well. And I think the BMI and the calories are such a dominant factor when it comes to diabetes that unless we have other lines of evidence that corroborate, there is always a significant uncertainty. Personally, I've always been skeptical of the causal link between red meat and type 2 diabetes. I've always held back in my content from making strong claims precisely because I was always a little skeptical that there was a true link there. I think the observational data is suggestive, but there's plenty of caveats. Things that make me be open to it are one, the plausibility that Mario covered and two, some of the randomized trial data, although there I think there's plenty of caveats as well. So I'm not sold in either direction and I wouldn't be surprised if it's a confounder or if it's a small factor in specific contexts, but in the grand scheme, not a major piece of the puzzle that wouldn't surprise me either. So maybe we can touch on some of the randomized trial data that I've seen unless you guys want to move more of the caveats and confounders of observational data in which they're coming. Maybe I could add one thing that's related to what I just said and that is that, so we could just leave it at that and say, well, okay, there could be some limitation here, but we could also choose to try to find cohorts in which there is not this type of strong association between other confounders, non-healthy lifestyle, higher calorie intake and red meat consumption. And there are some, mostly Asian actually. And what we're definitely finding is that the relationship between red meat, unprocessed red meat and incident type 2 diabetes is either weaker or absent in those cohorts. So I actually have a few here. There's one Japanese cohort where you find an association in Maine, but not women. There's one cohort in China where you find an association that's weak, but only in urban settings, but not in rural settings. That always, I think, gives me pause because if we really think that the biological mechanics of here and they actually normalize it to per 50 grams of red meat intake per day. So it wasn't that the spread, for example, of red meat intake was lower or smaller in the rural areas. That basically was totally null in the rural areas and in the urban areas where there was more of this strong association between red meat intake and socioeconomic factors, like income, you know, their UCD association. It makes me wonder, and I do think we should ask these questions, right? So I think one point I want to make here, as someone who's passionate about methods in nutrition to your audience as well. And I know you both know this, but one thing I see a lot on social media is when people discuss this, they always conclude with, oh, this is total rubbish. We should just discard it and not ever look at it. And then they pick something else that has the desired results. It also is maybe not a fire quality, but then they'll totally jump on board with that, right? Really, I think everyone, as we're doing this, we should really try hard to stay neutral and really just assess this and say, well, this has some value, but it also has these issues. Because everything we look at, including the randomized control trials, we're going to look at later, including the randomized control trials that I've ever done, they all have limitations. And I think we'll do the best and have the best chance of coming to a conclusion if we try to take in the cumulative evidence with a neutral mind. And that's very hard. But I really encourage everyone to try to have that, including in a case like this, where potentially some passions are flying in terms of your own experiences or your own philosophical ideas. One thing on those cohorts, kung, not sure specifically the one you mentioned, but thinking about this as an exposure with a number of the Asian cohorts. Typically, we do see lower overall index of red meat compared to certainly the US cohorts, but even maybe to some European populations. And so when it comes to a lot of the epidemiology, even the highest quartile or quintile tends to be relatively not that high, let's say compared to what we might see. And so that may be one of the explanations why we'll see maybe a lack of effect there. And we're just not getting to either an appropriate contrast in the exposure, or maybe we're just not even reaching a high enough total amount, past a certain threshold where we start to see those. Very fair point. We remember specifically both of these cohorts, Japanese and Chinese, I looked at the intake in the lowest three quartiles, the intake was very, very low. The median intake in quartile three was something like 50 grams a day. And in the highest quartile, the median intake was 85 grams per day, where in the US, typically US cohort, you would see more like 120 or 140 grams per day. So yes, it's true. We also see maybe some suggestive threshold effect there in these Asian cohorts where basically no increase in risk at all in the lower three quartiles. And then it only jumps in the fourth quartile if there's well. So that would suggest, as you say, that most of the population eats very small quantities of red meat, and it doesn't really have any impact there. And then only at the very highest intake level for that population, which is still maybe what we would see an average year in Western countries. Do we start to see that evidence or not? And that may be partly the reason why we're not seeing that there. So I think absolutely a good point that you're making there. And I certainly agree we need to do a lot of things. And there's a lot of utility in continuing to look at the epidemiology on this question. And it shouldn't be discarded by anyone. One of the things we can do is to see how consistent or not that might be with some of those other lines of evidence, most notably here, some of the randomized control trials looking at intermediate markers. And so, Gil, do you want to start us walking through for you some of the most informative piece of that I've kind of performed your opinion on this? Yeah. And as Mario was speaking, I was also remembering that I've seen some poor studies that try to address this question of the confounders, not just by doing the multivariate analysis, which isn't maybe a little less reliable, but by doing these baseline stratifications, which can be a bit more compelling. And I remember that the NIH, AERP, who Horde did that, and they had a group where they called them the healthier group. And basically they had a healthier diet, normal DMI, didn't smoke, never or mild alcohol drink. So basically they tried to select the population where they minimized these confounders. of the healthy user bias. And in general, they found a stronger association there between the red meat and some of the mortality outcomes that they reported. A couple of care yards is, I don't remember, I was trying to look this up, but I don't remember its unprocessed meat specifically or total red meat for that analysis. And I don't remember if it's specifically with the diabetes outcome or if it's general mortality. I actually have it right here, Gild. There is one from also from the Harvard group on NHS NHS 2 and the HBFS. I can't find it here right now either, but I do remember reading that as well, where they aimed to use the alternative healthy eating index to categorize people. And then within each of the categories by alternative healthy eating index, they look whether the association holds. Because that type of analysis would cause stratified analysis, right? As you say, where we basically not adjusting for these confounders, but we're basically trying to categorize people and then look in each. And they found the same thing basically the relationship holds remarkably strongly in all of the different categories based on the alternative healthy eating index. I'm happy by the way to share all of the references with you, Danny, if you want to post them in the show notes for anyone else, then you can do that. Yeah, good memory there, Gild. I want to bring up another point that maybe of academic interest, but maybe exemplifies why this is such a hard question. And that is the question. Should we even adjust for BMI in these type of analyses? Now, everyone will say, well, of course, we should because BMI is a major determinant of type 2 diabetes risk, but at the same time, if in a cohort we see that people eating more red meat have a much higher calorie intake and also higher BMI, then I think it's at least in the realm of possibility that for whatever reason, whether it's the red meat or the diet it came with, right? They may have gained more weight and they may be heavier and higher body weight is partly a mediating factor. Now, no one, no one thinks that red meat by itself determines BMI. So not adjusting for BMI seems like a mistake, but all I'm saying here is that it is well possible that in those analyses on these models where we're adjusting for BMI, we're partly over adjusting a little bit, right? And that in fact, the actual hazard ratio may be higher. So I think we need to consider all of these points in all directions and not just say, oh, it could be this, and let's just do away with that is probably irrelevant. So I'd agree, looking at all of the different data from all around the world, I'd say there were some concerns that make it unlikely for me to be highly confident in this, but I would say if I had to come to some kind of conclusion just based on the observational data, I would say yeah, the observational data do suggest a small increase in risk in people eating the most unprocessed red meat. And maybe that's a point we will come back towards the end when we're thinking practically from a recommendation point of view, that's where we maybe see a divergence between really trying to infer some causality around the particular exposure and outcome, where we may be do some of these adjustments versus generally thinking for people who on average in the population who tend to eat more or less of a certain type of food, would that be doing something better or worse related to their overall health? We'll certainly visit that idea. So maybe let's see what we can get from the RCTs then and pull this together on these diabetes outcomes. Briefly, another point of context, which I'm sorry if this is bogging down progress here, but I think it's important because this question is bound to come up. And that is that we are discussing the possibility that red meat has a causal effect in time to diabetes onset, but even if that's true, even if it is cause and effect, I think we'd all agree that it's probably a smaller piece of the puzzle of what causes diabetes. In other words, you could have a diet with plenty of red meat and see an even an improvement. You could see an improvement in your diabetes or glucose metabolism metrics or even potentially put your diabetes in remission. If some of the other moving parts are being calibrated, so for example, if you're losing weight, if you're losing body fat and ectopic fat and visceral fat, the red meat, I think, would be based on the evidence we have a minor player, like a small button compared to the big button of calories and BMI. So just to address this question that is bound to pop up, which is, well, but there's all these people that go on red meat diets and they improve their metrics. So when you guys are talking about his nonsense, we are talking about the possibility that it's a multifactorial disease, which is almost all of these chronic diseases are and that if there is a causal effect there, it would be one factor and potentially a minor factor compared to calories and BMI. There's no contradiction. In terms of the randomized trials, there aren't that many, but I've seen maybe three or four, a couple from the Harvard School of Public Health and a couple from the Iranian group and they are similar in the design and in the outcomes. The little patients here are get the de-tempically worked whole of months. Basically, they are replacement trials, so they tend to replace some of the red meat and this is centering on on-process red meat being replaced with some form of plant protein, typically. So legumes, soybeans, chickpeas, things like that. And particularly it is not in prior substitution. It's three days a week replacing, I believe, one or two servings a day, three days a week. And what most of these trials that I've seen report is that when this replacement is done over these couple months, they see a reduction of LDL, that's less surprising, LDL cholesterol, that's one of the most consistent observations. But the also see some metrics of glucose metabolism improving modestly. So home IR fasting glucose and then inflammatory markers, serious active protein interlooking fix that kind of thing. In both of the groups have reported those observations and the last caveat I wanted to point out is that these populations are not healthy individuals that usually have either metabolic syndrome or rectuidities already. So it's not looking at onset, it's looking at improvement of these metrics. And what strength here is that they are isochaloric, when isonitrogenous, at least some of them are weight naplens, who had perfect weight lost, and they're just trying to isolate this in fact of these foods, but keeping protein. And in some cases macronutrients constant. So I think it's interesting data. Again, a point of concern there, I think it's limited to this population. We don't know if this would be reproduced in a healthier population or not. But another thing that I think is an interesting data point. So I have a bit of an advantage, you know, maybe I should have volunteered to bring this up first because I just did this research and there's about 10 randomized control trials that I could find. Among the 10 studies that I found, there's any number actually they differ each in a very profound number of ways, differed by funding. There's actually a few funded by the National Candidate of the Association, I know that's sometimes the point of contention. Then there's some in generally healthy omnivores, that's randomized control trial, for example, by the Christopher Gardner group, well designed, and then there's some as Gil suggested in people who are already having either metabolic syndrome or suffering from pre-dibodies. So interestingly what I tried to do here is look at what was the comparison group and what we can say generally is when the comparison was read meat versus say poultry or a red meat versus carbs or red meat, here is one just a general diet that was rich in red meat versus low in red meat. So it wasn't a comparison with one particular food, but people just still devoid with whatever they usually eat, right? Then there's one where one diet was very rich in lean red meat, but low in dairy compared to one that was rich in dairy and then one that was containing neither lean red meat nor dairy. And so there's any number of these types of comparison that actually most of these do not show any impact of that switch on measures of glucose tolerance. So be glucose tolerance itself or fasting glucose or HP1C if the trial was long enough or measures of insulin sensitivity or beta cell function. In some studies actually use pretty sophisticated measures like a hyperlipid, insulinemic, uglistic reclamp, when you did three hour, frequently sampled or glucose tolerance tests to really look at the area under the glucose and insulin curves. And I came away actually with from this mostly thinking, well, this doesn't look at all. In many of these studies it was six servings a day, sometimes daily, you know, 150 grams would be five ounces of unprocessed red meat. Sometimes usually average I say was a baby for eight weeks was the intervention period. Many of these were crossover design studies which is very nice for glucose tolerance. So you can direct the compare to a beyond meat meat alternative compared to poultry compared to soy or non-soy degoose compared to plant proteo foods is generally speaking compared to carbs compared to just a diet rich in red meat versus a diet low in red meat then compared to a diet high in dairy or a diet that's both low in dairy and red meat and compared to a diet that's rich in soy. So these were basically all, I think this one the soy was I know these first ones except for the soy. These were all basically now. So no effect. And there's only few so basically there's one that compared red meat against whole soy in the form of soy nuts like roasted soy nuts that people had to snack on and one in which people were given fatty fish and another in which the red meat was compared to carbs. These three all show either a little bit or pretty substantially improved metrics of glucose on your faces. So the soy study I'd say is probably the best one here published in the American Journal of Clinical Nutrition shows that people who eat the soy nuts this was also again a crossover design study. They had a home of three point as compared to 3.9 on the red meat diet. That's quite meaningful, right? So they're not just by inside, they're not getting cured, they're not getting into the normal range for home, which would be somewhere around 1, below 1.5 maybe, but their insulin sensitivity quite meaningful, they improved. And actually, the way I interpreted this, cumulatively, is to suggest that it's not necessarily that the red meat here is the bad actor, but that the things that people eat instead in these studies, they're simply the fatty fish and the soy, probably has some very beneficial impact. And I'm saying this simply because if these other studies the red meat didn't do anything, right? It didn't actually worsen glucose tolerance or insulin resistance or beta cell function. But so that's one, but there's one important other thing to say about this compared to the observational data. And that is, when we look at the observational data, the people ate red meat regularly, when we look at the table one in a typical study that describes the distribution of other dietary and lifestyle factors across categories of red meat intake. What we find is that in observational studies, particularly in the United States, higher red meat intake tends to be associated with, as we said, higher energy intake and other unhealthy behaviors, right? Often lower vegetable intake, lower fruit intake, lower legume intake, lower micro, multi-butter-mini-take, lower exercise, most smoking, it's very common that you'll see that. And people eat some kind of mixture of red meat, including potentially fatty red meat, and they eat meat possibly in the context of, say, fast food meals, hot dogs, burgers with fries and soda and so forth, right? This is, I think, a very much in contrast to these randomized control trials because, for ethical reasons, these investigators almost always tested diets that were generally designed to be healthy, rich in fruits and vegetables, often also containing facial legumes and nuts and seeds and so forth. And then the beef or red meat that was added, usually was very lean type of red meat, like pork chops or very lean cuts of beef. And I think that's potentially very informative here, because if we reconcile all of this evidence, what we can maybe say is that, yes, consuming a lot of red meat in this regular form where people commonly get it, right, at the steakhouse with fries and a beer, in the fast food place with fries and soda, hot dog in a ballpark, right? It's a fatty cut of meat, it's, and it's eaten with other foods that may not be particularly beneficial. That is probably a very different story compared to, in these studies, where people really made an effort to provide, you know, whole grains, fruits, vegetables, nuts and seeds, and then a piece of lean red meat, where then it didn't really affect glucose from your stasis negatively. So my conclusion that I came away with from the randomized control trial data is to say, yeah, you know, we didn't really test the hypotheses based on the observational data because the diets we gave the people, even though they were very high in red meat, they didn't have the other characteristics of the red meat that red meat eaters have in the wild, right, in these free living cohorts. And so we basically tested healthier versions of red meat which diets, does that make sense? And those healthier versions of red meat rich diets didn't seem overwhelmingly to have negative impacts on glucose homeostasis. I was going to ask from your summation, I wasn't clear if those trials where you didn't see an effect if some of them were also in people with diabetes or were thin. I know that the one from Gardner, I think there were college students. So I think it's two different questions, right? One thing is, can you take healthy people and worse and your glucose metabular will work a couple of months? Another is, can you take people with diabetes and metabolic syndrome and make that better? Slightly different questions, but I think what we've seen, the data we've seen is, I don't see a contradiction, unless I miss something where in the trials where people start out unhealthy, we see at least in some a difference in the ones where they're healthy, it's really a mixture, right? It's just 10 different studies that I've seen and they include anything from healthy young people all the way to people with type 2 diabetes. Most of the studies use people who had the metabolic syndrome or prediabetes or either of the two. And that makes sense, right? Because in a four-term study, you want to be able to see changes in either direction, right? So if you give people red meat or soy, you want to be able to actually say if you suspect that the soy is going to improve glucose tolerance, as it does in just one study, you want to be able to have room for improvement, right? Personally, I do think this makes a lot of sense to study this and people who already have some degree of glucose tolerance, but you're certainly correct that that is somewhat different from the observational studies and in that, the observational studies by definition start with a cohort of people who are free of these types of conditions, and then they just try to identify those who develop them during the polar period, right? In general, though, when we think about diabetes, I think diabetes has this one big benefit, and that is that we can measure the clinically relevant variables directly. We can measure glucose tolerance. We can measure insulin resistance and beta-self-function, right? These are not just biomarkers for the disease. They are basically measurements of the disease itself, right? So I think that's a real benefit for diabetes. And for me, that strengthens the degree to which these, mostly null randomized controlled trials are meaningfully informative, in that they don't suggest to me that in the case of diabetes, that in be it healthy people or people with beginning or late stage glucose tolerance, that eating even large amounts of this lean red meat in the context of an otherwise healthy diet, has any negative impact on glucose tolerance. Now, people may say, well, how can you say that? Shouldn't we, if there's even just one study that shows that eating these soy nuts is better, shouldn't we all just be replacing all of our beef with soy nuts, right? And I'd say, yeah, if you like soy nuts, you know, feel free to eat soy nuts. But I think we also have a responsibility, a scientist, to recognize that soy nuts are eaten very differently from beef, right? They eat as a snack. Of course, there's other soy products, but they were not tested here. Actually, they did test a soy protein isolate, which didn't have the same benefits of soy nuts. The point I'm making is we have a responsibility that if we tell people eat less of a certain whole nutritious food, which meat undoubtedly is, right? It has a lot of micronutrients, and it does help people not have anemia and copper deficiency, and zinc deficiency and so forth, right? And it's something people like, and it's something that is versatile and part of cultures in any way. So I think for me personally, if I were to tell someone, don't eat this, or eat very little of it, I would have to see very strong reason for me to do that, right? And if, in the case like this, what I see is compared to most other foods in these studies, be it poultry, be it carbs, be it in many cases, other plant protein foods. Even here, there's also studies with soy and non-soiligoon that were null. So in other words, they had the same impact of glucose tones. That for me didn't reach the threshold where I feel like I need to tell people that to minimize the risk of diabetes, they should lower the intake of red meat. The one thing I would say is eat the type of red meat as it was tested in these studies, which is probably leaner cuts of red meat and eat it in the context of an otherwise healthy diet with lots of other plant foods. So that's the conclusion that I came to from all of this. And that still leaves the question, of course, how do we reconcile this with the observational data? And my attempt to try to reconcile it is to say, in the observational data, we didn't test people, we didn't have people who ate mostly, selectively, lean red meat in the context of otherwise planned, vegetable-rich diet with nuts and seeds and so forth. But we basically had people eating just random cuts of red meat, which are more fatty in combination with any selection of potentially less healthy foods. And that may have been a contracting factor in the small effect size that we're seeing in these observational studies. That's my best attempt of reconciling these data. But what kind of conclusion did you guys come to? There's two things I'll briefly comment on because I think this will also come up with the Cardiovascular Disease discussion. One is this really important point of thinking about the comparator food or just this concept of compared to what. And of course, that always makes a big difference. And as we've seen here, depending on what the comparator food is in these trials, we see difference in outcomes. There's then also an important aspect that Mary, you've raised of how people should really think about their and useful comparisons that it is something that would be replacing that other food. Now, what we could probably say then is based on what we discussed. We at least have some of these potential improvements when that comparator was in that case of the trial around soy nuts. We could maybe then say, well, we need some other trials looking at other typical types of replacements, other types of, let's say, soy products or so on or plant-based proteins that were tested yet. But potentially could have a benefit. But again, in a lieu of that, we can't really come to too much conclusions. And then finally, one thing I'll put out there, I tend to see differences in how people interpret evidence based on these things. And by that, I mean, in the case where we do see, let's say, a meeting for improvement with some of these substitutions, whether it's an epidemiology of plant-based proteins in place of red meat or in some of these trials. For some people, that interpretation will be given that that in itself is an indicator that the red meat is actually increasing risk because there's something we could do to reduce that risk, let's say, because we could see an improvement buttoning that in. Other people would take a different view and say, well, that's maybe we're seeing an additional benefit in those cases. But that's not the case. the same thing is saying that having a bit or some or get more red meat is actually driving negative impacts. And so there's tends to be different interpretations. I see people make I was trying to figure out if there is a way to reconcile everything, but I still wasn't clear if some of the the trials that you've seen. I know you said that some are healthy individual some are metabolic syndrome were related in the ones specifically on individuals that have some of these conditions metabolic syndrome or type 2 diabetes that have these metrics abnormal and where the red meat is being replaced with a quality plant protein. You've seen a heterogeneity where some do not show the improvements on the point of protein is that right? Yeah, so there is here for example one study by Hill and colleagues published in the American Journal of Plasticity 2015. They compared red meat, a diet, a red meat and one that was identical based on the dash diet basically that was now rich in plant proteins. If for people who were overweight or obese with the metabolic syndrome, pretty good sample size, parallel design. So they basically saw no differential impact at all on any measure of glucose homeostasis. So that would be fasting glucose. But basically my summary here was there was no differential impact of these two diets on any measure of glucose homeostasis. And that's in contrast to the study also published in AJC and by Azat-Bacht in 2007 where they compared red meat versus soy protein isolate versus the soy nuts. Here they saw pretty clearly better glucose tolerance and better insulin sensitivity on the soy nuts specifically compared to the red meat with soy protein somewhere in between for most of the measures. So this was also in people with the metabolic syndrome. So you know, I'd say there's somewhat comparable, somewhat comparable exposures but still quite different results. What I'm open to though is that there may be some replacements like in this case the soy nuts and I wish there were other studies with soy products that are maybe more comparable to meat in terms of how it's used let's say tofu. I'm open to the idea that that's slightly better. Particularly think about these soy nuts. They're basically not just a protein food but they're also a rich source of phytochemicals and of fiber. And it's certainly plausible that that could be better than consuming say, a pork chop or a small piece of steak. I say don't think it's totally conclusive but you know from the data as I said that I've seen I would say mostly it doesn't seem to matter and maybe this is worth pointing out you know, Gil as well. There are several studies in which people were given really, really high doses of red meat like five ounces every day for example for eight weeks and here's one study from Germany published in Diabeticologia for example 2015, Noatme. They even combined a high red meat intake with a diet low in fiber and free of coffee. So they really wanted to make it bad for people with glucose intolerance and they compared that to one that was very rich in cereal fiber and coffee and low in red meat and they did a hyperinsulin make you can see the clamp. They did a whole bunch of other tests. Absolutely no impact whatsoever on any measure of glucose homeostasis. So to me that actually is reassuring that there isn't something that's particularly strongly glucose intolerance inducing or insulin resistance inducing when we're eating even high concentrations of red meat. And so there's several studies like these again I'm happy to share these another study here that also used 150 grams per day also seeing no impact at all. This is also in people with pre-diabetes or the metabolic syndrome, crossover design study. Was the comparator? This is one group that did two studies say at one compared to carbs and one compared to poultry and in both studies it was null basically null impact whatsoever on any measure of glucose homeostasis. So you know what that but that needs to be is that if the observational data actually were reflective of the cause of effect relationship between red meat and glucose intolerance then in a study like this we're particularly actually in people who already have some beginning glucose intolerance and we're giving them lots of high dose of the red meat at least in some of these studies we should see an actual increase in say the area under the curved glucose or in a metric like insulin resistance, home IR or a reduction in insulin sensitivity is what I would expect and that would certainly increase my confidence in these observational data. And the fact that in basically seven of 10 or eight of 10 we're basically not seeing any of that and we're only seeing that with the soy nuts and fatty fish we're seeing improvements in glucose tolerance and insulin sensitivity. I personally would attribute to specific benefits of these foods that were tested there against red meat because if it was the red meat that actually is the culprit here that causes this differential effect we should see the similar effect in these seven or eight other randomized controlled trials right. Hot saying for that used chickpeas and lentils and things like that it was like you were normal. I think the word wanting in on is that the comparator is absolutely crucial and I think what you're saying is absolutely possible and even likely that maybe it isn't the red meat itself causing the problem is that some of the replacements may be a bit kinder on the glucose metabolism maybe in specific contexts like someone who already has diabetes metabolic syndrome other replacements don't seem to have that benefit. So yeah this would absolutely come down to how the diets are put together because yeah if you're if you're swapping the red meat with full tree maybe you don't see any any difference and what's in specific populations may be swapping chickpeas doesn't prove modestly your metrics. I think that that might reconcile the evidence but I would agree that we don't have convincing evidence that it's the red meat specifically causing these issues and I think this is always an issue with these replacements. Is it one food? I mean it's almost it's almost how do you look at that Mario sometimes this seems to me like almost like semantics because is it one food that's bad or is it the replacement that's better at the end of the day it's worse than the replacement it's almost almost a semantical argument when we're talking about always replacements in terms of dietary recommendations. Well absolutely I think everything that has calories we should think about in terms of replacements right because we got to eat something and we can't just continue eating like we do and just add a whole bunch of other stuff so anytime you increase your consumption of some food you'd have to decrease something else that has calories or you're you're going to be forced to change another variable that's your energy intake. So no yet I think this is entirely correct. One caveat here to what you concluded is the cumulative evidence I would agree it doesn't suggest that red meat is directly like a huge culprit in inducing insulin resistance or inducing beta cell dysfunction or glucose intolerance but there may be other foods that better than red meat I'm totally again for that conclusion and there I would say well you don't actually have to choose right you can have an occasional dinner with maybe some lean red meat and you can also have some soy nuts and fatty fish occasionally right so I would say I do that the other thing I would say from these studies we cannot say anything about you know fatty red meat eaten in an unhealthy dietary pattern because they're all basically used a healthy dietary pattern like the dash diet and incorporated very lean meat and some of these studies actually even though they they gave people six servings of red meat per day the saturated fat intake was like six or seven percent of energy so really they made an extraordinary effort to keep the fat content from the beef low otherwise this wouldn't be possible. So I think in the context of that I think we can come to the conclusion that you know if you succeed in having some lean red meat occasionally in the context of an otherwise healthy diet it doesn't seem to be uniquely likely to trigger glucose intolerance and so on and so on and so it will be my conclusion. Here you listen maybe this helps to kind of summarize based on at least some of these trials I don't know I'm still not clear if you want directly contradicts this but at least in four trials off the top of my head when some of the red meat is replaced isoclorically with different legumes could be lentils or could be soy specifically poppy regions were typed with diabetes or metabolic syndrome in these trials makes the improvements it's have only more freeze would this would you recommend that replacement specifically to that population or are you not convinced actually so there's at least two trials here has an seroshtami and hill in both they compare red meat versus soy and non-soil legumes in one study and plant protein foods in the other and they so not differential changes in any measure of glucose homeostasis so in other words what I'm suggesting is that what you're arguing namely that if we replace red meat with legumes it's always beneficial I don't think that supported by all of the data there's some data where this is the case but not all of the so the effect I think isn't consistently there as far as I can see it you could be that there's some difference in the population or in the exposure or it could be that it could be that it's inconsistent all the way to it not being a real effect and it came out came out in a couple of trials because I mean that's theoretically possible could also be that some studies are underpowered but these studies actually have well powered here one has 62 in the protocol analysis the other 64 one has two arms the other three so you know 64 with three arms it's possible they could be underpowered or that the dose wasn't adequate it's possible certainly those are the abatics right the the hill paper there hill hill with people with metabolic syndrome overweight two in a piece with metabolic syndrome the other has an seroshtami type 2 diabetes, but not insulin. I agree that lower is overall confidence. It's heterogeneity. Either we can find an expert in the methodology. If we can't, we see if it comes out both ways. And so we're not sure if it's contest dependent or what's going on. I agree with that. Exactly. That's what I came away with. And I'm sure we will return to that one more of maybe getting to how people should view this or what people should do. Would they likely experience a benefit if they do make some of these substitutions we've discussed? We'll return to that. I am keen to bring in the outcome of cardiovascular disease because there is obviously an overlap with a number of the issues that we've already talked about. The complexity of trying to come to some of the understanding here, the importance of the comparator foods, how we maybe try and reconcile the epidemiology in this area versus maybe some of the trials we have. Here, we have one case where, at least mechanistically, there's maybe obviously that we maybe don't need to discuss too much because the whole topic around the actual saturated fat that if someone is consuming a lot of fappy cuts of red meat, that is providing a lot of saturated fat to the point where LDL or APOB is going really high. I think most people would be very accepting that that in itself is increasing risk of the provision of atherosclerosis and ASCVD. And so that point there that relates to the saturated fat content, which maybe we don't have to spend too much time on, but maybe where there's this gets really interesting and I think it's something you had mentioned to me before a kill was around. If we take the saturated fat aspect of increasing LDL cholesterol out of that equation, that is worth then thinking about maybe leaner cuts of meats or we're thinking about overall dietary pattern that includes red meat, but isn't that high in saturated fat? Would we still see actual impacts on cardiovascular disease risk? Are there some reasons to believe or potential mechanisms by which that could actually be having a impact on risk beyond just providing saturated fat to the diet? Yeah, so I've seen a couple of studies that tried to address that I believe it was NIH, AERP, one of the examples that tried to do that. Now these are with adjustments. So there are some caveats as well. I'm trying to find the specifics here. Two that I've seen, one is out of the nurse's health study and health professionals follow up study and they initially saw an association between red meat and take up. I believe this was total red meat and cardiovascular death of 16% per dose. And then when they adjusted for saturated fat and dietary cholesterol, it was reduced, but not entirely. It came from 16 to 12. So it's a great value to suggest that partial media factor, saturated fat and dietary cholesterol. They also adjusted for hemirons separately and it came down from 16 to 11. So at face value, it's adjusted both our mechanisms but both our part of the picture and the NIH, AERP, they did not. It is a just for saturated fat. It was only hemirin, which was also a partial effect. So that is one approach that I've seen. And then there's from randomized trials, there's this trial from the Kraus lab where they saw differences in able-be between meat protein and non-meat protein, even when they tried to hold saturated fat constant. There was still a residual effect of the non-meat protein in lowering able-be modestly. So from what I've seen, saturated fat is likely to be a factor but probably not the entire phenomenon. Yeah, I think it's hard to know, right? I have seen several randomized controlled trials in which they again gave people a very healthy diet. Again, I think there's two studies from the same lab, I'm blankly read now in the name, but they basically used a dash diet and then they basically either headed with one serving of red meat per day as part of the diet or a plant protein replacement. And they chose very lean beef, though, for this. And they kept saturated fat and take very low. I remember I think at 6% of energy or so. And in that context, these diets didn't differentially affect this fasting lipid profile. Now, I think that's reassuring, but at the same time, we don't know because we don't have any long-term randomized controlled trials with actual outcomes data. We don't know whether there could be other effects. For example, through TMAO or the Hemiarine, that over time could increase the risk even in that setting. And as you're suggesting, our best guess, I think in this case, we have to guess, right? Because we don't have a kind of outcome. We could measure an randomized controlled trial that directly tells us what's going on. Our best guess would be that even if we did that, there could be a small increase in ASCBD risk in people consuming even lean red meat. So we have to have the, admittedly imperfect data we have. And those would provide us with a best estimate and that best estimate would be that even if we prioritize lean red meat and even if we consume it in the context of an otherwise very healthy fibro-rich diet, rich in unsaturated fatty acids say, there could be residual impacts of that red meat on, even if our APOB and LDL cholesterol remain very low, there could be residual impacts through other mechanisms on atherosclerosis and cardiovascular disease. That would be my concrete. Whether that's a big effect or not, I would think it's not huge, but it's probably a mildly increased risk in people who consume red meat regularly. - I think this relates to that point, but the comparator. And we know some of those other foods that have shown some of these benefits. So there's a meta-analysis of RCTs, for I think from the Harvard group that maybe you look at some of the differences on these cardiovascular outcomes, stratified by different types of substitutions for the red meat, for things like poultry or fish or carbohydrates, there was no real difference, but where we saw differences was for what they turned high quality plant proteins, which was legumes, nuts and soy. Again, if we were taking that into account, that would be a situation where maybe it is those things that are driving some of that risk reduction because we know for a variety of reasons the impact of, assuming soy proteins or legumes or fiber per se has this notable benefit for some of these outcomes. Well, what do we do with this? What do we do in terms of recommending whether someone limits swaps out changes or not? It is one thing and then a separate thing is the actual anti-academic, are you even pinning down the magnitude of effect of red meat itself? - To summarize my view, unlike the diabetes angle, which I've always been a bit skeptical of that causal relationship, I think there are concerns there, but I think the jury is out. I think with the red meat and cardiovascular disease, my level of confidence is a bit higher because the mechanistic mediating factors are better established, but I think that pertains mostly to fatty red meats. Once you get to lean cuts of red meat, I think the uncertainty goes up a lot. And I would say even more specifically, fatty cuts of red meat that are sufficient to elevate able to be significantly. There, I think my concern would be substantial. But when we get to these more specific situations, okay, it's a super lean cut and the able-be does not go up in this individual for whatever reason. There, I think the uncertainty is much higher. I think it's certainly possible, like Mari was saying, that there's residual mechanisms, whether it's a hemire or a TMAO. I'm a pretty skeptical of a TMAO in general as a causal factor, but that's kind of a detail. But that would be certainly one recommendation I would have is for someone who is opting to include some red meat to favor leaner cuts. - If we try and compare that to the epidemiology in this area for the CBD outcomes, it's something that again, relates to a similar story. We saw earlier that for those US cohorts, where we have those higher groups being at 120, 140, 150 grams plus as that high, there we tend to see relatively consistent findings, I would say. Again, there's a lot of nuance within that, but with a general blanket statement for these CBD outcomes, where we see not so clear cut is maybe, for example, those Asian populations or even some of the European populations where the exposure contrast is much smaller, and then the high group tends to be a lower, which then throws up another question. Is it because we don't have sufficient contrast or maybe there's a threshold effect of where we start to see negative impacts, or maybe the aspect we mentioned earlier or the baseline diet, but we do see differences depending on where that cohort is, geographically, what our baseline diet is, and the total amounts. I think where we see the highest amounts, as at least where we probably see a bit more consistency in the epidemiology, I don't know what your thoughts on this, Mario, what you take from the CBD data. - Yeah, no, exactly the same as what you guys concluded. So I would also say that I have higher confidence that in the absence of really strong, rather, masculine, trial data, because we only have randomized control trial data on one mechanism potentially linking the exposure to the endpoint, right? In this case, the pathrofile, and that's actually the randomized control trials that have been done, as you said, used mostly lean red meat, and depending on what they compared it against, there was really no big impact, or no impact at all on LDL cholesterol or APOB, but nevertheless, the consistency among the observational data, I think, is in my opinion, I might a bit more convincing here. There's also nice papers that have stratified analyses, for example, by fruit and vegetable intake and so forth. There's one food suite in a recall where they looked at people eating a very healthy diet characterized by a high intake of fruit and vegetables to see does that hold that relationship hold as well in that population and that seems to be the case. And so I'd say yes, there's uncertainty with both of these literatures, but I would personally, where I would fall is to say red meat per se is not a major concern for type 2 diabetes, most likely as long as we prioritize lean meat and keep it in consumer in the context of an otherwise healthy diet. But I'd say for cardiovascular disease, red meat, even unprocessed red meat, is probably a concern if it's fatty and maybe less likely so if it's less fatty. So that's maybe how I would word it, right? Whether the lean red meat is really not a concern at all, I think for debate. I think if you asked 100 scientists this area, it would go anywhere from yes, it's a major concern to some people saying it's not a concern. I'm falling somewhere in between and I would say, to me, that is enough reason to say, well, maybe we don't have to have red meat every day, even if we prioritize lean red meat. And that also then creates space for some of these other healthier foods that definitely also have health benefits where we can have occasional meals with fatty fish and occasional meals with some legumes and some tofu and whatever else we like. That's my view where I would fall. But to maybe round all this, when it comes to actually how we should think about recommendations and again, a huge caveat here for people listening to the others and have an address, all potential health effects. So most notably cancer as an outcome, which is ever important, I want to consider for this exposure. But given that we're talking about the two exposures we've looked at today, in terms of recommendations at a general level and the current evidence based that we have, where along the spectrum do you guys fall of thinking about red meat as something that we should either consume freely without too much worry, something that should be delimited to some degree, something that the recommendation should be mainly avoid or restrict as much as possible, along that old spectrum of possibilities, where do you think the current evidence base for these outcomes might lend at least you individually to fall on those recommendations? - So I would actually say that as long as we prioritize lean cuts of red meat and otherwise, either healthy plant-rich diet, moderate intake from my perspective is absolutely fine, but it's not a must. Like it's also fine if people just have occasionally or never, but I do certainly think that for the purposes that we've discussed here, I think if someone had a small piece of red meat, say two times a week, three times a week, maybe even four times a week, I wouldn't have major concerns with it. And I think that is with regard to type two diabetes and also with regard to cardiovascular. So I wouldn't encourage extremely high intakes, but I also don't think we need to tell people to strongly diminute it, for example. I think a moderate intake seems just right to me. - Yeah, my takeaway from the evidence is a bit more conservative based on what I've seen. And because some of these cohort studies looking at cardiovascular disease did some baseline stratifications, looking at people who consume more fruits and vegetables than are healthier in general. And the association with cardiovascular disease often survives, and in some cases even seems stronger. And because we basically see a near linear relationship with no clear dose of intake, where we don't see an association, I would be a little more concerned, telling people, for example, that three, four times a week is definitely fine. I think the increase in risk there is likely small statistically, but I think it's entirely possible, specifically with cardiovascular disease. Diabetes, I'm more, I think things are more up in the air, but with cardiovascular disease, I think there's a likelihood there, particularly with satire meats, and if there's an elevation of it, it will be that we will have some cardiovascular events that would be preventable. And I think at the individual level, that risk is modest. If you're talking about a couple of servings a week, by the time we were talking about one serving a day, I think the risk is modest still, maybe in the order of 10%, which is modest if your lifetime risk of having an event is, say, 20% it goes to 22%, so it's certainly not a big increase. It's a very small part of the overall dietary pattern, but a little bit more concerned with recommending three, four times a week liberally. I think if you add the caveats of lean cuts, and there's no increase in April B, I am definitely less concerned, but the end of the day, what do I tell people that are close to me where there's no, I only want the best for them. I recommend limiting, right, not being a regular staple of the diet. I don't think the red meat needs to be eliminated from the diet, but I would recommend infrequent consumption, maybe once a week or under that would be a range where I'd feel safer. - And I think as we've maybe alluded to a few points, it would be great to continue to see more high quality data on specific research questions that we've outlined today, where it would be nice to have some clearer picture and some clearer answer, but hopefully this has served to everyone, listening as a great overview what the literature generally is showing right now, what we have available, maybe some what way we can start to interpret this and what this might particularly mean for these outcomes that we've discussed today. So with that one's a Dr. Ramyobrats and Dr. Gil Karalliou, thank you so much both of you to giving up all this time to come and talk to me and then for the great insights that you've both shared. I've really enjoyed this and I really appreciate what you've done. - Thanks so much, Dan. - Thanks guys, just a second. (upbeat music) (upbeat music) - Thanks so much for listening into today's episode. Before you go, I just wanted to remind you about Sigmunditrishm premium, our subscription for those of you podcast listeners who want to significantly deepen your understanding of nutrition science and become truly confident in your knowledge. So what's the idea of this subscription? Essentially, it was created with the goal of allowing you to more deeply understand the material you're hearing on the podcast episodes themselves to be able to retain more of that after you finished listening or reading through the notes and then be able to easily and efficiently revise over that so that in the future you can be able to remember that information, to be able to reuse it, to be able to create your own content or ideas using things that you have learned. And how do we go about this? Well, there's a few different ways, but at the core of the subscription is our detailed study notes that you get to each episode where you get a beautiful PDF that is full of all useful descriptions, background context, diagrams, charts, et cetera, to allow you to more deeply understand some of the concepts that were mentioned throughout that particular episode as well as then linking them back to previous episodes. You also get these segments at the end of each episode called our key ideas segment where I recap certain key ideas. You get episode transcripts, you get then a number of premium only episodes. So you have your own premium podcast feed that appears on whatever app you already use and you get these extra premium only episodes. Some of them might be ask me anything sessions where we answer your questions that you submit it directly or they could be a variety of other episodes that you may have seen previews to in the public feed. So for full details on this, then check out the link in the description box wherever you're currently listening right now or just go to sigma nutrition.com and you can see all the details there. And of course, your support is what keeps sigma nutrition going. We don't run ads, we don't sell supplements, anything like that. So your support is what allows me to continue to do this. So thank you for that. I hope you do come back for the next episode regardless And until then, have a great week. Stay safe and take care.