#600: Finite Knowledge, Infinite Ignorance

Hey everyone, I just wanted to give a quick message before we get into this episode. As you'll hear, this is somewhat of a milestone for the podcast being not only the 600 episode of the podcast, not including our premium exclusive episodes, but it's also the 12 year anniversary of this podcast starting. And I just wanted to say, thank you so much to all of you who are listening right now to have listened to any time over those 12 years, whether that's from the very start, from some time over those years, or even more recently. And for your support for telling other people about the podcast, for sharing the episodes, for letting people know through a recommendation about it. And of course, to those of you who subscribe as a premium subscriber and support the show directly, that is literally what allows me to continue doing these podcasts. And I just want to say, thank you so much for that. It really, really means a lot. And through whatever way that you have helped the show either through listening, sharing or supporting directly as a premium member, it really does make a huge difference. And like I said, it what allows this show to continue to run. So thank you for that. To mark the occasion, as you'll hear in a few moments, we're going to do something a bit different. Alan and I are going to discuss some of the things that over the many years we've been doing this podcast, positions that we have maybe changed our opinion on or modified our position on slightly. These will include some things where we maybe have a different conclusion now, thanks to coming across new evidence or considering a topic in more detail. We're going to discuss what our old position was, maybe what led to that changing and our current new position on that. And then a couple of examples of things where our conclusions, maybe our not only the same, but have even strengthened and through the process of learning more and coming across more evidence saying getting better at interpreting evidence, our position as strengthened. And so I know over the course of these 12 years, I have learned a tremendous amount through the course of not only preparing for episodes, but for my interest just generally in reading and interpreting nutrition science, having the privilege really to be able to talk to the world leading experts in the field of nutrition science about their publications. I know I've learned a lot continue to learn. And so hopefully this might be informative for you to see some of the things that led to modifications in those positions. And of course, they are still open to change again in the future. You'll see that the title of this episode is something in some way inspired by a Carl pop-up quote that you'll see probably in the description box of this particular episode about the fact that really our knowledge is something that is finite and our ignorance has to remain infinite. And so for that reason, there's no shortage of things we could have discussed this episode. Without this preamble getting too long, I just want to say again, thank you so much for being a part of the podcast. This point for supporting this work and for valuing these types of discussions where we try and stay away from sensationalist stuff. We don't do this to promote certain products or supplement lines and read the goal is to try and get the best quality objective information out there, but more importantly teach people how to interact and interpret the nutrition information they're coming across and the studies people are discussing. And so with that, I want to thank you for the support because it really means a lot. So thank you and please enjoy this episode. Hello and welcome to another episode of Sigma Nutrition Radio. Well, I say another episode, but it is somewhat of a milestone where episode 600 of the podcast and also as this goes out, we will be basically spot on for the 12th anniversary of the show, which you know itself is a bit strange and wild. But to do that, we said we would do something a bit different, maybe mark this milestone episode. And to do that, of course, I am joined by someone who's been an integral part of the podcast for many years. And that's Dr. Alan Flanagan. Alan, thanks for coming and joining me. It's a pleasure. I'm honored to be here. I think episode 500 was down under, I think that feels bizarrely long ago at this point. I'll have to check that. But I do remember that in the episode, I threatened that once we got the 600, that would be the end of the podcast and there would be no more. I won't confirm or deny if that's accurate. So people will have to wait to see if after this, there is a 601 or if this indeed is the last time they ever hear from us. That's that we will leave that hanging in the air for people to do with what they please. Yes. So that is possible. So just in case that is what's going to happen, I encourage everyone to savor this episode. The concept that we came up with that we thought would be quite interesting and useful as a way of looking back on all the years that we've been doing these podcasts and talking about different concepts is to maybe reflect on over that period of time that the podcast has been running. What particular areas we may have changed our mind on particular topics or particular ways of assessing them or changes in our position around a variety of topics. And so we're going to each present a few that come to mind as ones that we have maybe altered or modified our position on, give some of the reasons why that is the case, maybe what our original position was, things that maybe influenced changing that and where we now fall on that particular topic. And then we'll also maybe reflect on something that we haven't changed our position on or if anything has got even more precise or stronger over time as we've continued to see more evidence come out on the topic, be able to sharpen skills in a particular area looking at that. And so hopefully this will serve as a good reflective episode of sorts and will maybe be useful to those of you listening. So maybe to start, I'll give you the honors of starting us off, Alan, maybe can you introduce us to what is your particular position or claim that over time you have changed your mind on over these years? Maybe you can start us off with what that is and what that initial position you had was and maybe walk us through what has happened in the time since. Yeah. So I've gone with three that I hope will as we work through them kind of illustrate for people different examples of altering a position or changing one's mind. And the three I've gone with are time restricted eating or intermittent fasting as a kind of intervention for weight loss, cardiometabolic health, etc. And that's an example of somewhere where I have changed my mind and adopted a reverse opposite position to what I had maybe five or even seven years ago. I was very much convinced that there was something to time restricted eating, but it was an intervention with some potential promise that there could be independent effects of time restricted eating on a range of markers might maybe independent of weight loss, for example. And this might have to do with obviously the timing of the meals or the condensing of the eating window, etc. And I'm now at the opposite end of that. I really think time restricted eating is a whole lot of nothing really. If I'm being completely honest for reasons off, flesh out. So that's an example of me going from I think there's something here to I don't think there's anything here really. The second one I'm going to discuss is cocoa polyphenols, specifically in relation to cognition. And that's an example of softening a position that I was once more optimistic about. It's not going to the extreme of where I am with time restricted eating. And it's still being open to some better research being conducted. But based on short term acute studies of cognition, I had very much formed an opinion that perhaps given some of the mechanistic stuff we know about polyphenols in the brain that this might be something that actually does in a diet high or rich in polyphenols, maybe have longer term protective effects. And I'll discuss the cosmos trial and why I've now solved in that position to I'm no longer really that sure or confident in whether there are any meaningful effects of cocoa polyphenols on cognition. And how that might be relevant for long term neurodegenerative risk. And the third example then I'm going to discuss is EPA Ethyl Estran. It's impacts on cardiovascular disease. And that to me, I wanted to discuss that because that to me is an example of why null findings are important in any research field. Because prior to the publication of the strength study, which did not use EPA Ethyl Estran, used a mixed EPA DHA free fatty acid form of intervention. I was of the opinion that if we were granular and scrutinized the omega-3 literature, then a lot of the conclusions people were coming to that there is no benefit for heart disease were incorrect or misleading. And I have always kind of thought that it's not necessarily accurate if you're if you're very granular about formulation and delivery method. And then the strength trial was published terminated early due to no benefit. And everyone through their hands up was like, you see there's no effect. And actually I felt that once you reconciled why strength came to no effect because of the choice of intervention. That's an example of going from. a softer position to a more confident position. So the opposite of where I've gone with Cocoa Polyphenols. So I hope that there are different illustrations of the type of changing position or altering position based on different factors within the evidence of each of these we can discuss. Yeah, and I think that's the real thing that we want to hammer home that it's not just a case of saying, oh, here's something I believe in the past. Now I don't necessarily believe it anymore. I think the real utility for people is what we often maybe try and mention the podcast is thinking about this epistem, epistemologically about how do we know things about nutrition? How do we evaluate that evidence base? And then as we get not only better at doing that, but have more data that we can actually assess those on, we can modify those positions. And so hopefully those three examples, if you say normally have these different angles that we're looking at, but all our examples of how we might try and understand evidence better. And so I think this will be really useful. Let's start with that first one, time restricted eating. Of course, this is one where you've outlined that maybe position that you had in the past and that you have basically completely gone away from that. What for you were some of the either key pieces of evidence or the key shifts that happened that started that process of you changing mind. What was most impactful in thinking, oh, maybe this is something where I need to modify this position? Yeah, I think from 2018 to 2020, we had the publication of several human interventions using different time restricted eating types of protocols, whether it was an early time restricted eating where all food was consumed, three meals were consumed. For example, between say 7 a.m. and 1 p.m. or 2 p.m. or even these kind of midday time restricted eating with the midday to maybe 8 p.m. type of window. And there were some interesting findings in a free living context. Some of them were more controlled interventions in terms of diet. But there were various findings coming out in relation to weight loss. For example, that weight loss was potentially greater in a time restricted eating intervention compared to a kind of conventional eating pattern spread across maybe 12 hours. There were some suggestions that independent of weight loss, cardiometabolic risk factors would improve. And from that, then we were starting to get claims that if we're thinking about the circadian kind of corona nutrition aspects that will maybe there was an energetic advantage to time restricted eating that in fact if you front loger, three meals in an early time restricted eating type of protocol, you might actually get greater energy expenditure and thermic effective feeding. Or even if you condense it in a midday context that increasing the fasting duration is itself something that might augment weight loss or fat loss. And I think my confidence started to be shipped in that sometime around 2020 to 2022. Initially we had that low in colleague study, which I wasn't entirely convinced by because they just didn't have a reporting on the actual kind of timing of the meals or anything within that. But it was followed by some other time restricted eating interventions. There was a paper in a Chinese intervention, G in colleagues published around maybe two years later 2021, 2022. And you were seeing essentially no meaningful difference in weight loss over longer term periods again over one year. Allude in colleagues and other time restricted eating intervention from China compared an early time restricted eating with energy restriction to a conventional no restricted on eating duration again with the same kind of magnitude of prescribed energy restriction. And and found no difference in weight loss over one year. They lost seven to eight percent of their body weight from baseline. And I guess some of the work that I did on my PhD and others very much in my mind negated the hypothesis that there was any energetic advantage to time restricted eating. And there was one paper then published a couple of years ago. You probably recall, do you remember Kevin Hall developed this kind of predictive equation for weight loss? Like if you have repeated measures of body weight, you can actually use that with and then you know the time frame over which those repeated measures are, you can use this whole predictive equation to model weight loss, model the predicted energy deficit for a given amount of weight loss. And there was a paper published a couple of years ago from Jamshed and colleagues. And it was a 2022 or 2023 paper. And it was one of these studies that again, free living context time or restricted eating compared to a conventional eating passion was one of these studies that initially in its overall analysis suggested time restricted eating led to greater weight loss. And that was independent of self-reported energy intake. And the self-reporting of energy intake in nutrition interventions is a huge issue because it's woefully inadequate. But the nice thing about this study was they then went and models using Kevin Hall's equation, the predicted energy intake and there was actually a greater difference. There was a greater reduction in predicted energy intake in the time restricted eating group. So again, you were like, okay, so the greater weight loss wasn't independent of energy intake. It was likely simply explained by the fact that time restricted eating simply may facilitate a greater reduction in energy intake. And I remain relatively unconvinced by the magnitude from a clinical meaningful, clinical meaningfulness perspective of some of the cardiometabolic improvements. And so just taken together the kind of hypothesis that while time restricted eating influences weight loss through some kind of additional, either weight, the thermogenic fairy, you know, that there's an additional benefit to a restriction on the eating window in terms of weight loss. I simply do not think the evidence supports that. I think it's basically we're coming back to differences in energy intake. If you control energy intake or have similar energy intakes between your two groups, you do not see any advantage to time restricted eating. And the magnitude of the weight loss in terms of there was a score in colleagues met an analysis that came out and they reported a statistically significant quote-unquote effect of time restricted eating. But the actual magnitude was 0.93 kilograms of weight loss and the confidence intervals range from like 0.17 to 1.7. So entirely sure of anything that we would call clinically meaningful weight loss, which is about 5% of initial body weight, 5% to 7%. And so I really from probably about 2020 started to waver 2022 onwards then started to very much think that there's nothing independent of restricted energy intake that is explaining any effects of these interventions. Does that mean that it's entirely useless? No, because clearly in a free living context there are some people that may benefit from just a restriction in their eating window. In terms of facilitating a reduction in energy intake. So I think there may be some utility there for certain people. But is there these kind of lofty broad brushes as far as an intervention for weight loss and cardiometabolic health that people should quote-unquote be doing? I just don't buy it. I don't think there's really any independent effect of time restricted eating at all. That is not explained by reductions in energy intake. And so ultimately there may be a practical benefit like you say if someone ends up reducing energy intake or we see some differences maybe in hunger across the day in depending again on the situation itself. But really when we look at any of the metabolic or weight loss outcomes, once you're accounting for differences in energy intake, there's a whole lot of nothingness is the kind of ultimate position. Yeah, basically. And I think that's relatively confirmed now. From several of the kind of intervention trials that have directly tested this from that scoring colleagues meta-analysis where the actual summary effect estimate is so unconvincing of anything. It's basically a range of noise. So I just think there is little to no evidence that eating duration, modifying the eating duration specifically by shortening it, with any shortening it to any corresponding time of day window improves cardiometabolic health independent of restricted energy intake. Perfect. So we will leave that for our first one. Maybe I should do an introduction to the three that I'm going to mention across this. The first one I was going to mention relates to probably a subset of two to three separate claims and altes out relating to protein. But broadly the concept of maybe previously overvaluing protein dosing based on theoretical maximizing of muscle proteins synthesis. And within that there are a few different things we have to tease apart. The second one I'll get on to a bit later relates to unprocessed red meat specifically. and the claim that unprocessed red meat can be included regularly within an otherwise healthy dietary pattern without any real need to worry and the activities of any particular concern. And then the third relates to dietary cholesterol and the often heard conclusion that well dietary cholesterol has no effect on blood cholesterol levels. So maybe first I'll outline that the protein one. I know you've looked at this area, so I'd love to hear your input on some of these. There's a few subset of claims that all kind of converge on this idea of, oh, we're valuing protein dosing too much at any one particular meal. One was just on the total amount of protein that might be needed for some of the outcomes right to body composition or even satiety and impacts on energy intake. I do remember a point back if we look back a number of years where I probably even would have said to people that had general heuristic to aim for two grams per kilogram of body weight of protein for just people in the general population because oh, it's gonna have this effect on satiety, this is gonna impact energy intake. It's also gonna be useful for body composition for some of these muscle and strength or the outcomes. So there's that part around the total dose that we can mention. The second was maybe even going more granular than that and thinking about per meal dose of protein and that not only is do we need to get the total amount correct, but each dose of protein should be appropriately high. Even maybe drilling more into that of looking at the protein types and saying, well, it needs this amount of loose scene within there to again, theoretically maximize muscle protein synthesis within this particular feeding. And if we're not achieving that, we're not gonna be able to maximize this anabolic effect of protein. And then tightly related to that is how many times can we do that across a 24 hour day? 'Cause the more times we maximize this MPS response then better, right? And so this distribution idea of within that total amount of protein someone's consuming, we need to distribute that relatively evenly across the day and each of those feedings needs to be above X amount grams of protein and X amount of loose scene or other particular amino acids. And obviously related to that becomes the whole conversation around protein sources because of the differences in amino acid profile between a lot of the commonly touted anal-based proteins or versus plant-based proteins. And just over time through a variety of evidence that we can maybe discuss in a moment, I've just started to see that. First, once you have an improvement of overall protein in the diet, for most people in most situations, you don't see too much differences based on a distribution that is skewed slightly or is that even across the day. And there's, even last year, there was a really nice study from Asco and colleagues that looked at even muscle protein synthesis specifically with a skewed and even distribution in people who were training. And they had people who were either omnivorous or vegan and within each of those, they also had a balanced or an even distribution or the unbalanced and basically across those four different groups you don't really see any difference. And their total protein was in around 1.1 to 1.2 grams per day. There was also a justisan paper that again looked at muscle protein synthesis, amino acid utilization. And again, once you get the sufficient intake of protein in that particular study, this was 1.5 grams per kilogram of lean body mass. Again, whether that's even or skewed, it didn't really seem to give any difference. But beyond that, it's also probably not that important for most people anyway. And then we get to the whole concept of once you get a enough total protein, again, how much do each of the protein sources matter or the amino acid profile at each feeding. Speaking of earlier, when you mentioned when we were down in Australia for that conference, this is something you actually presented on in around these differences in plant and animal based proteins. And some of these issues maybe historically with the protein literature in general, maybe this over focusing on certain scoring of proteins, again, over focusing on muscle protein synthesis. And again, in most of the data we've seen in the last few years, whether that's on hypertrophy, whether it's on strength measures, whether it's on some in older adults, we've seen again, once you get a sufficient amount of total protein, and a lot of these studies are aiming for maybe somewhere between 1.2 to 1.6 grams per day, whether that protein is coming from plant-based sources. So we have a number of dietary interventions that are plant-based. So Benjamin and Walls group have done some really nice work on this where there was mainly micro protein was the supplemental protein they were using. Comparing that to omnivorous diets, you didn't see any difference in hypertrophy, you didn't even see it in muscle protein synthesis. And then there's being last year, low-bethamurino, data you've met analysis of RCTs, looking at plant-based versus omnivorous diets for strength outcomes. And across those, I think there was eight RCTs in total, five were vegetarian, three were vegan. You don't see any differences in total strength or whether you stratify that for upper or lower body. And so again, it seems that once you're at this, a sufficient amount of total protein, some of these other things that was maybe paying too much attention to, don't really matter. Even for people who, let's say, nominally their goal is to gain muscle mass and strength, even in a lot of those situations, it doesn't matter that much. Never mind most people who want to improve. And I think probably you can make a same thing about the actual total amount of protein people would need and recently did an episode on this, but essentially this idea that for strength or body composition outcomes, there's gonna be some benefit to even going up, say to two grams per kilogram of body weight. There's actually not really good evidence for that. Most of those seem to top out around maybe that 1.5, 1.6. But I think the biggest nuance is that even at that difference between lower intakes of 1.1, 1.2 up to that 1.6, are relatively so small that for most people aren't that important. And so all of that kind of converges into similar things of I would have maybe over-emphasized the amount of protein that's needed at each meal based on those amino acid profiles or looking at its role on muscle protein synthesis, theoretically what would maximize that response in terms, as opposed to actually just thinking about the magnitude of difference we're actually seeing. And so now it's much more, not even relaxed, but just more in line with what we can confidently say from the evidence, I think, right? That's the difference. And so I don't know where your position is as felt or if it's in any way similar. - Yeah, I'd basically kind of be at the same point that you are now. I think I went on somewhat of a similar kind of evolution where in terms of just thinking about protein, where initially based on proxy outcomes, like impacts on muscle protein synthesis, I would have lent towards the, there's clearly a quote-unquote superiority to sources like way or a dairy based kind of protein or an animal protein based intervention compared to. And then there were these supplement studies published. I think it was Joy and colleagues, this is probably maybe 2015 or 16. And there was a couple of others where the initial dent in that for me was when the studies used a plant protein supplement but matched for the losing content to say away. And then so the MPS differences like fall away. And then you get beyond that and you get to the kind of total protein, daily dose issue and you get to like whole diet kind of interventions. And these kind of differences start to basically become irrelevant. And for me it was, when you did talk about this a lot, when we did that episode of Eric. For me it was the clear gap between muscle protein synthesis is a proxy marker or fractional synthesis rate, which are markers or measures that may not actually capture any physiologically relevant outcomes for what we want in terms of protein. If we're thinking about this from a sports or recreationally active person's perspective, recovery, et cetera, those measures just aren't really relevant relative to something like lean body mass accretion or exercise adaptations. And so once I started to see some of those outcomes that would compare, you know, a vegan diet or a vegetarian diet to an omnivorous diet, looking at more kind of relevant outcomes in terms of strength or exercise. And it just all fell away. Once you were in these ranges and then like you say, I probably would have linked more the higher end of those ranges, certainly based on I think analysis like the 2022 meta analysis that looked as dose of protein. Can't remember who the lead author was, honest. - There's a Rob and Morton paper from 2018 that kind of got a bit of notoriety. Eric was actually one of the lead authors on that, but Eric has been quite critical of some of their statistical analysis since then. That was the one that kind of was the 1.6 grams per kilo kind of regression. - Right, that's the one that's in my head. - And that was the one that kind of got a lot of sighting. But more on the research stuff. - And we discussed that. - They've highlighted that are probably better quality ones are inarounds that kind of figure of 1.51.6. There was a newness paper and a Tagawa and colleagues one might be better quality. Maybe if it's 2022, it could be the new news Tagawa. - Yeah. But it was just to echo your point of ed, the magnitude of the actual difference between some of those lower, like 1.2 to 1.6. being relatively questionable in terms of unless you're like at elite sport level where the minutiae matter like for your average person who is exercising to retain muscle mass or build muscle mass and be healthy and or looking even thinking about it from an aging kind of standpoint, healthy aging standpoint, all of those minutiae on type of protein, source of protein, excessure, really did fall away from me as well over the last couple of years and it leads you to thankfully a much more kind of flexible place and I certainly wouldn't, I think even it's translated for me in my own diet, like that hang up I probably would have had in the back of my mind a few years ago even as even Moran the 2022 mark about if I was eating more kind of corn products or like plant-based needle turnitives or whatever and it was like is it really doing? Is it really giving me what I need and yeah I just don't have I don't have any of those hang ups in my head anymore. Yeah I think that's a really useful thing from the practically speaking because look if someone is having 1.6 or two grams or whatever it is for protein it's not dangerous in any way and it certainly you can make a case that again theoretically could we point to some of this data and say if you have a goal for body composition or strength that maybe that 1.6 is better than 1.2? Sure but like you know the magnitude is very small that's probably not perceivable for most people but I think the misinterpretation most people then have is if they are consuming less than that quote-unquote optimal amount and let's say people are talking about 1.6 I was per kilo then they think if I'm having a lot less than that I'm somehow not going to be make any progress I'm not going to have to recover or I'm going to be losing muscle which is just absurd right that most of that is coming from the stimulus you're doing from your exercise and even if someone was consuming like at the RDA and they were doing a sufficient training stimulus maybe I think it's fair to assume that you could probably get a better long term response to having a bit more than the RDA but even if they're consuming that 0.8 grams per kilo it's possible for them to build muscle and build strength etc if all the other things are in place yeah I think that was one of the big and I don't read I don't know anything about sports or exercise research at all but I guess so I'm taking this kind of interpretation on trust from people I trust that do but that was another thing that hit home and it was like actually the biggest impact on any of these certainly proxy measures and muscle protein synthesis fractional is resistance exercise like that is by far the biggest factor in influencing these kinds of outcomes and of course long term longer term adaptations that that someone is desiring off if they are pursuing some for in training program so that to me was also like oh that's the thing that really matters here is like just make sure you're doing exercise or resistance training in this particular context and the rest of the kind of details in terms of protein type and dose and spacing and all of that is just far in the rearview mirror in terms of kind of priority but the final thing I'll say because it echoes something you said earlier is that once you do pull away from this kind of overvaluing or this too much focus on protein and particularly protein dose per meal because you think if I don't do this I'm doing myself some degree of harm or I'm not going to be able to progress is that once all of that stuff falls away like you said and you're not as worried about having really high intakes for protein that the per meal dose or the sore so I like these things that aren't that important I have found at least that it becomes so much easier for a general dietary pattern to be probably way healthier because you're not made you're not over indexing on protein when you're thinking about okay what I'm going to have for this snack or this lunch or after training and instead of saying oh fuck I only had 10 grams of protein in this particular snack I'm going to have a protein bar instead of saying well I can also have some nuts maybe doesn't have any protein but probably from a long-term dietary pattern standpoint you're actually having a better quality diet or these sorts of things. Yeah you're not just you're not just people aren't just like feeling like they have to have a chicken breast it's like compensated yeah it's it definitely does yeah it just makes makes diet a lot less rigid in that sense and I do I remember the heady days of the long 2010s when it was I wouldn't leave you wouldn't leave the house without a shake oh yeah with some way and at least you go two hours before you start to atrophy that was absolutely tablize into a weekly yeah when is that that MPS curve when is that going to drop and when can I try to get my MPS again I need to spike my MPS that's my obsession by diet so with that's another of a protein let's go on to your second one you mentioned animal which is around cocoa polyphenols and some of the cognitive outcomes and of course we have a full episode going into some of this previously so can you maybe remind people of with some of the literature maybe what was some of that emerging data what position maybe did you have previously and then since then what have been some of the things that have maybe slightly altered your view in terms of the confidence you might have on those particular outcomes that you outlined earlier going back a lot of years there human research really only starts coming out 2018 onwards some of the polyphenol research goes back a bit further in terms of contemporary research influencing my thinking and you can find studies back in kind of 2014-15 and I found I've always found polyphenols fascinating because of the fact that they are non-nutrition of compounds like they're not essential in the way that vitamins and minerals are you won't die without them they have been I think we've mentioned this term before in the podcast some polyphenol researchers have termed them lifespan essentially they might not be essential for your day-to-day capacity to not die but in terms of what they do in the body overall health status they're contributing in meaningful ways precisely because they're non-nutrition of compounds their behavior in the body is more as a kind of xenobiotic in terms of mechanisms of action they look more like drugs than they do like nutrients interacting with the kind of microbiome the production of polyphenol metabolites that remain in circulation they're impact on different signaling pathways and the promotion of certain processes whether for cardiovascular health for example in endothelial function but my interest specifically in the cognition aspect was I guess twofold one it's just interesting mechanisms like in terms of some of the mechanistic work or some of the animal model work you do see some really interesting mechanisms in terms of polyphenols impacting or lowering neuroinflammation promoting neurogenesis and in the context of neurodegenerative disease which is a major issue as far as our kind of lifestyle or chronic health conditions in society go it's one of the most pressing concerns for which we really have almost no meaningful pharmaceutical interventions so there is this emphasis now with rising dimension Alzheimer's rates on still on potential earlier intervention lifestyle interventions potential impacts of diet and we have all these long-term associations and epidemiology better difficult to tease apart particularly because the age of onset is later and so what was happening in the mid 2010s was a lot of interesting short acute human intervention studies looking at different polyphenol compounds you blueberry anthocyanins were a big focus green tea categans and cocoa polyphenols and what was interesting about these endothelavinols specifically was these short-term studies you could see these kind of acute impacts on specific domains of cognition, visual spatial tasks processing speed, acute memory often those kind of markers rather than for global cognition and they were often like I said acute studies often single-day studies, high dose polyphenols looking at someone's cognitive function across a range of testing thereafter over the course of the day or supplementing them over a short-term period of maybe a week to a month maximum and all of this was making me think we're seeing some interesting effects on cognition this and my thinking at the time was if you added this up yes these are acute studies but surely if you sustained a dietary pattern like this or sustained intake of these compounds over time would that up to would that add up to a degree of meaningful protection against cognitive decline and in 2022 from 2022 onwards we've had several publications from a large study the cocoa supplement and multivitamin outcome study acronym is the Cosmos trial and this was a trial with a bit 21 and a half thousand participants over about four years testing the effects it was a two by two factorial design so there was either cocoa flavonols emplacivo a cocoa flavonol so active cocoa and a placebo multivitamin or active cocoa and an active multivitamin and then you could have placebo cocoa and active multivitamin and placebo cocoa and please the SIBO multivitamin, so a full placebo group. And what was interesting, there's now been several cognit of publications specifically. The primary outcome of the overall trial was cardiovascular disease. And that did actually find a 27% lower risk of cardiovascular disease mortality, 16% lower risk of major cardiovascular disease against events with the cocoa polyphenol supplement group, the cocoa flavonal supplement group. And then we had the publication in 2022 of one of the cognitive outcome studies, which looked at, and what was nice about it is there was repeated measures of cognition. So you had repeated measures over a three year period, and 70%, 77% of the participants completed the cognitive assessments in all three years of follow up. So you've got repeated annual measures over a three year period. So very much getting us away from those acute studies where he gives someone a dose of cocoa flavonols and you look at their verbal and visual memory. And there is basically no significant difference between the active and placebo groups for cocoa flavonols across either global cognition or specific domains of cognition, specifically episodic memory. What was interesting was that the multivitamin group did show a significant difference in terms of the global cognition scores. What the cocoa group, which based on all of these acute studies that preceded this that I mentioned, this to me was a relatively surprising five, was counter to the hypothesis that I would have been working with coming into the publication of the cosmos trial. And it had a three year follow up period, and it was to say, I guess, surprising. And there were various substudies then published from that. There was the Cosmos Mind Study, which is the one that I just mentioned that found no difference in global cognitive scores or specific domains of cognition over repeated annual examinations over a three year period. And then there was the Cosmos Web Study, which was a web based, a cognitive testing. And then there was another study that was published in 2024, which was a clinical sub cohort. So these were people that attended for in-person testing themselves. And what was interesting, I think, about that particular study is that it also found no effect on overall kind of cognitive measures, which looked at global cognition again, and then episodic memory and executive function. So that's an individual who were looking, who were attending for in-person cognitive scores. But one of the other studies from the three cognitive substudies in Cosmos actually have repeated measures. And this is why I articulated earlier that my position on this has softened, but I have open questions. So it's not, I now think there is absolutely no effect of cocoa, flavonol's on cognition. That's not the case, and I don't think we can conclude that. Because one of the issues we have beaten to death on this podcast over the last five years or so, which I really hope our listeners are now well-accuned to, is nutrition intervention trials largely do a hopeless job of accounting for baseline nutritional status. If you are conducting an intervention of a nutrient or a bioactive food component, you cannot expect to see an effect if people already have sufficient levels of that nutrient or bioactive food component. And the Cosmos web substudy was a study where they measured urinary measures of a metabolite called GVLM. This is a direct reflection of cocoa flavonol intake. And what they actually found was that in individuals with the lowest level at baseline, there was a significant improvement in memory performance and that was correlated with the increase in their urinary levels of this cocoa flavonol metabolite. IE as a reflection of increased intake from baseline from the start of the intervention. And the Cosmos studies that I've mentioned that have no effect on global cognition didn't have repeated measures of GVLM. They essentially only had baseline measures. And they did consider that biomarker, which is a validated biomarker of flavonol intake. But they ultimately did not have repeated measures. So we weren't able to see, did people actually increase this in a way that might have a meaningful impact or were they deficient at baseline. So in that study, the effect of cocoa flavonols in the other Cosmos in the web and silery substudy was primarily evident in people with the lowest baseline levels who exhibited a more pronounced increase in levels of this metabolite of cocoa flavonols over one year. So what have I taken from this? Well, I've gone from being relatively optimistic that there was a genuine signal in the noise for the impacts of cocoa flavonols on aspects of cognition that could be relevant for long-term brain health and neurodegenerative risk. I think there's enormous disconnect and discordance in our evidence for neurodegenerative disease generally. That is a methodological reflection of the diseases that we're talking about in question, not necessarily nutrition, it applies to a range of exposures. It's very difficult to study how something you do in your 40s influences an outcome that largely starts to occur in people's seventh or eighth decade of life. I don't know that will ever cross that bridge either. So we're left with trying to do more interventions like Cosmos that aren't just acute single meal test studies, but Cosmos and its publications took me from that position of being relatively optimistic that there was a genuine signal in the noise to now not being at all confident that there is, but still because of the study that looked at baseline levels of a relevant validated biomarker, the future of nutrition science for me is biomarkers. And the more that we can get established biomarkers of an exposure to relate that, the more that we can actually get away from self reported dietary intake in nutrition interventions and improve the actual ability to say that an intervention has been adhered to, that study still leaves me with the open question because my hypothesis always with any dietary exposure is that you will only see an effect of this intervention if it is in people with low status at baseline of that exposure you're interested and of that treatment that intervention you're interested in. The fact that that was found in one of the Cosmos substudies still leaves me to think that maybe in people who have very low intakes of polyphenols, they can benefit in terms of their cognition if that is increased, but that is a big if based on this because it's a subgroup analysis. So you need to know, unfortunately, expensively go and some people in this field need to go and try and find funding to say we're going to specifically do something like a depletion type of trial, which you can do for polyphenols because they're not essential for life. And you can get people and stick them for a month on a diet that almost eliminates all polyphenols or something like that. I'm just, I'm spitballing if you know what I mean, hypothetically in terms of research design, but this is something, like I said, that Cosmos has made me go from cautiously, has made me go from optimistic to not optimistic, but cautiously still thinking there could be something if the correct type of study is actually done that addresses some of these issues. - Yeah, so we've, like you say, that's a, okay, with the frame that you've had that initial positivity and optimism around the evidence that was there, thinking that, okay, this could go and get then reinforced when we see some of this kind of longer term data, that doesn't pan out so that kind of undercuts this potential confidence, but doesn't negate that there could still be a benefit knowing that we need to test this specific context out where it looks like there could be a bit more promise. I, in those who have baseline levels, I've intake that are relatively low, they would be the ones that might receive this benefit. That is to be tested, but at this moment, we can't have any degree of confidence that there would be, hence this kind of modification of your position. - Yeah, exactly. And like I said, the reason that I still hold out some very cautious, very cautious openness to their potentially being a signal in the noise is because the increasing use of biomarkers in nutrition interventions is relatively emerging methodology. The validation of the biomarker that I mentioned for Cocoa Flavanoals, GDLM, only occurred in the last five years. And we had a really nice analysis that came out of the Cosmos study, I think it was published last year, maybe your 2024, which looked at biomarkers of Cocoa Flavanoals, including this one. And the degree of using biomarkers, rather than self-reported intake, using biomarkers to determine adherence levels in the participants in the trial and what we found. the people with the highest level of adherence, and they looked at the cardiovascular outcome specifically. It strengthened all of those outcomes from the primary trial. So we need to stop fumbling around in the dark, which is what. And people, it's funny to me, we still criticize. The mainstay criticism of nutrition is epidemiology, oh, it's epidemiology, epidemiologists do a far better job of assessing dietary intake than trialists do in nutrition. Far non dietary assessment in nutrition interventions is trash. And this is something that I am encouraged about. The ability now with more advanced methods and mass spec and this kind of thing to be able to validate biomarkers of different aspects of diet and nutrition to then use those. Your designing studies going forward, like people should be thinking, right, we're not going to just give a 24 hour week call every month in the intervention, an average that and call that a dietary assessment, like use biomarkers to do. And yeah, this would mean a more expensive study, but like I would rather see bigger, fully funded research projects in nutrition than an increasing proliferation of small crappy trials with self-reported dietary measures. So the biomarker aspect of cosmos is what keeps me with the cautious openness that there may actually be something here. That in itself could be a topic for another discussion. We may have another time point of some of these methods within nutrition trials. For now, let's keep this thing moving. I'll maybe mention one again, we've discussed this a bit and it's one of those areas similar to what you've just discussed there where there is still open debate and ambiguity because there is people who interpret this evidence base slightly differently and come to slightly different conclusions, even when looking at the kind of same evidence. There are maybe some different reasons for that. It also very much depends on the outcome, but essentially the update of a position that I was going to mention was related to unprocessed red meat intake in the context of a pre-existing healthy dietary pattern. In that maybe certainly if we go back a decade, I would have been fairly okay to tell someone, if they look in the context of an overall healthy diet where you're consuming enough fiber and has all the good things that we typically recommend, if you have unprocessed red meat, particularly if that's a leaner cut of red meat, you can probably include that regularly without any real downside. It's going to provide you with protein. There's going to be some micronutrition and it's nothing ready to worry about. This now to a point where this is still a big debate within nutrition, like I said, where people have different perspectives on it, I think where we can probably have the clearest delineation here is when we look at different outcomes. So I think if we were to think of the big chronic diseases where we might see associations here or the biological plausibility, we have cancer, cardiovascular disease, type 2 diabetes in relation to this exposure. And again, if you have a different opinion, I'd love to hear on any of this, with relation to cancer as an outcome, and specifically if we focus on colorectal cancer, we can probably have more consistency in that evidence base for this exposure of unprocessed red meat. Maybe next to that, I think cardiovascular disease and there's some things that all maybe tease apart there as well. Maybe a bit less than for type 2 diabetes and there's a bit more open questions. But when we look at some of the, certainly the epidemiology in this area, tied in with a lot of the mechanistic data that we have that would speak to some of the effects that could happen from, let's say, regular consumption of unprocessed red meat at certain dose thresholds, there is enough for me now to think that it's probably not a good idea to freely and liberally consume this as X amount of dose. And again, we have to define that depending on the population and the outcome we're looking at most notably around colon cancer, but even in the context of cardiovascular disease, the obvious one that we can maybe put aside straight away is if we're talking about so much consuming a lot of fatty cuts of unprocessed red meat that is contributing a lot of saturated fat and in turn elevating LDL cholesterol or APOB, that straight away is something that is increasing cardiovascular disease risk and that is something we don't maybe need to spend too much time on. Maybe where there's a bit more of the interest ends up being in this area where we're specifically talking about, let's say leaner cuts of meat or in a context where the overall dietary pattern is relatively though unsaturated fat. So below the typical thresholds we mentioned in addition to having an overall healthy dietary pattern, how much of an issue is that? And I think the main, the kind of summarized my kind of position now is that there's enough along those certainly the biological plausibility and the mechanisms and for example on the CVD stuff, I think it was the wrong crowds paper, Dr. Gil Carvaliou mentioned recently this as well, that even when you have that in the context of again low saturated fat overall intake and you take away the impacts on LDL cholesterol or APOB, there is enough impact on some of these cardiovascular outcomes that is still left over, there's still residual effect even on things potentially like APOB that is not being generated because of the saturated fat content per se. There's other mechanisms at play. Certainly, then when we think about colorectal cancer, there's a number of mechanisms that we could talk about. But then as we've discussed the epidemiology in the area, certainly once you go to those higher intakes, particularly in US based cohorts, compared to lower, we start to see pretty consistent increases in risk and they're not something that I would willingly ignore on the account of well, maybe we don't have that same degree of risk if you consume enough fiber or if you consume enough fruits and vegetables because it's very difficult to account for that, which again is a fair statement. It is in the epidemiology. It completely account for some of those factors, but I think to say that is sufficient than to say there's no need to worry about it is something I wouldn't do. I think it's certainly prudent to limit intake on the base of the day that we do have in the area. That's maybe the kind of change I would have there. But I would love to hear your view. Again, we've talked a bit about unprocessed red meat previously, but for you, when you're coming to conclusions on this, where do you end up falling around unprocessed red meat in those specific contexts of overall healthy health, to apply it, but all the things people typically mention, these will offset potential downsides. Yeah, if we can find a just to the epidemiology for now, I became relatively sensitive to the kind of dose of the exposure that we were talking about in epidemiology. What made me live to this was seeing consistent associations in a classic high versus low comparison, for example, in something like the health professionals follow-up study or the nurses health study or some of the epic cohorts where you've got high higher intake, kind of quartile or quintile to be able to compare to. Then you would see the JACC cohort from Japan or some of these East Asian cohorts, no association, consistent, absolutely no signal in the noise for "high red meat intake". Again, another principle we've probably hopefully beaten into everyone of our listeners' heads at this point is you can't just talk about something being high or low in the abstract in nutrition. In any context, epidemiology or interventions, you have to define what that means. Then you go and look at the Japanese cohorts, for example, or some of the other East Asian cohorts, and you see that the "high" "quote and quote" group are consuming 72 grams of red meat, of one processed red meat a day. You go over to some of the North American cohorts and the "high" group is somewhere in the region of about 200 grams a day, or over at least 170, 180 plus given underestimations in dietary intake and epidemiology, we would infer that that bracket of over 180 grams a day is probably higher than that. You start to pick into, beyond the definition of "high and low", what these actual exposure contrasts are in terms of gram per day intake, and a consistent picture emerges in epidemiology, where levels of around 100 grams or less per day are consistently not associated with risk at all, levels of over 150, 60 are. So we can relatively make that conclusion, certainly from the epidemiology that, yes, certainly higher daily amounts do not appear to be conducive to lower risk of several health outcomes, cardiometabolic and cancer-related. What I feel like people are not familiar with or don't know, because we get this kind of trope that, well, there aren't control trials of red meat intake and relevant. Obviously, there are going to be surrogate endpoints, intermediate endpoints on red meat and any of these things that could further inform on that. And there are. There is a body of really nicely conducted, fully controlled feeding trials by Sheila Bingham and colleagues at the. Medical Research Council, human research unit in Cambridge. This is going back to the early 2000s, late '90s or early 2000s, but they did specific to colorectal cancer, for example. They were looking at home processed meat intake and the endogenous formation of N-vitrosocompens, which are strongly implicated in colong cancer, colorectal cancer, carcinogenesis. And there are some really interesting findings in some of their works, specifically one study that I remember early 2000s had a dose for response across levels of red meat intake from 60 grams a day up to 600 grams a day. And where you started to see significant increases in the endogenous formation of these carcinogenic compens, which by the way from their other research, we know is mediated by hemeyern specifically because people say, well, its processed meat has all the nitrosamines and these kind of compens preformed, but unprocessed meat doesn't. Well, it's the actual hemeyern component of unprocessed meat that then interacts endogenously to produce these compens. Their dose response study show that becomes present over about 120 grams if I remember correctly. So from 120 up to 600. So that's largely congruent with the kind of risk thresholds that we see in epidemiology. So like under that 120 gram dent, they're low intake of 60 between 0 and 60, there was no meaningful difference. So I am in agreement. I think that the idea that is still very prevalent and particularly I think in the health and fitness space where if you have a good diet pattern, you can have your daily dose of steak and it's no worry whatsoever. I don't think that's correct. And I think that we're not only relying on epidemiology to support that claim at this point, but we're also able to point to controlled feeding studies that are relatively congruent in saying that doses of 100 grams or less are probably not an issue, but it's not a large amount of unprocessed meat. But over that, there are relevant changes in physiological markers of colorectal cancer and indeed some cardiovascular markers as well that would give some pulse ability to a higher risk of some of those actual disease endpoints where that dietary pattern to be sustained. I think maybe one other thing that ties into this and it's not directly, let's say, measuring those impacts of red meat directly per se and by that I simply mean that this concept of replacement or the comparator food, depending on what type of study design we're talking about, that we've mentioned a lot here has an implication on what the actual outcome is like we to be for someone. And so for someone who is trying to make health conscious decisions, it's not only about, okay, at certain doses of red meat per se, we might see some of these negatives directly attributed to that. There's also the tradeoff of changing that exposure or changing someone's intake and if we can select a certain replacement food that would then bring a greater degree of disease reduction, that is having an impact on someone's overall health. I think it was in the episode that we did around industry funding. I think we mentioned the Lopez-Morraine O'Baper where they looked at red meat and again, you can see whether you find differences in some of those outcomes or negative outcomes are not a lot depends on what is the actual comparison being made. And so again, that adds in another layer for people that again, yeah, if we compare it to maybe eggs directly, maybe we're not going to detect that in certain studies, but if we compare that to other types of foods that we are positively contributing to the overall dietary pattern, we might see more beneficial outcomes as well. And so again, not directly looking at those thresholds of red meat and what they're doing, but a factor of substitution. Yeah, I think the substitution effects have become a lot clearer as well with some of the feeding, with some of the kind of interventions that have looked at like red meat versus white meat versus a plant meat alternative. It's just quite clear that there are impacts on some of these markers that are not like you said earlier, entirely explaining, for example, by saturated fat content. And then it's clear from a kind of long-term health outcomes standpoint that yeah, there is still net benefit to someone replacing a portion of the red meat that's in their diet with lentils or something like that. So yeah, the substitution factor I think is also really crucial. Maybe a topic for us another day and I think it would be worth us to go through another episode actually, specifically drilling it on the count throughout outcomes because there's a lot of nice evidence to discuss there. But for now, let's move on to your kind of third and final update, so to speak, Alan, and maybe remind us of what that was and that initial position you have. And then let's go through what has kind of changed. We've talked about it a lot, which is omega-3 fatty acids and cardiovascular disease. There are still more times than not I see takes along the lines of these are not recommended. There's no effect of omega-3's on relevant cardiovascular endpoints. And I've never quite bought that argument for reasons that I'll explain. And what changed my mind on this was actually the publication of a trial that found absolutely no difference in cardiovascular risk over a four year period trial was terminated early at the acronym strength. So the strength trial. And that I'll explain why a trial that found no significant difference in the impacts of omega-3 supplementation on cardiovascular hard cardiovascular endpoints strengthens my position. There's a benefit to omega-3's for cardiovascular disease. So what's really crucial in this literature is several moving parts that varies and that largely explains why different attempts at evidence synthesis and meta-analyses often came up with relatively weak effect estimates or kind of unconvincing kind of slash more evidence-needed conclusions. And it's dose. It's also delivery method. And it is within the delivery method the specific formulation. By formulation, most papers in this area refer to formulation as simply the relationship and respective amounts of EPA and DHA. That's not how I refer to formulation in this context. I refer to formulation as the actual structure of the delivery method. There was a meta-analyses from Frank Hoon colleagues when 2020-2021 that was one of the better considerations of potential dose response. And that had kind of found that we'll over kind of 850 milligrams a day, but it hadn't stratified by whether it was EPA or DHA. And so that is relevant. But more important than anything is delivery method. There's three main delivery that's that formulations that you could have. If you go into your local health store and you buy an omega-3 supplement, it's likely in triglyceride form. Triglycerides are big, bulky structures. You're largely going to get, for example, in your little capsule of fish oil, you're going to get something like maybe 500 milligrams or 300 to 500 milligrams of the EPA and maybe 200 milligrams of DHA. You'll get around maybe 8, 7 to 800 milligrams, maybe 900 milligrams of total omega-3s in a 1 gram of fish oil kind of dose. They're like, think we can confidently say relatively useless as a supplement for cardiovascular disease. It's not what we're talking about when we're talking about omega-3 for cardiovascular disease. We're talking about prescription formulas that are not considered dietary supplements because they have actual oversight within the framework that drugs are developed in, for example. And we have three of those formulations. One is a combined EPA and DHA in a form known as ethyl ester. The second is only EPA, no DHA, but also in that ethyl ester delivery method. The third then is carboxylic acid combined EPA and DHA. I'll explain briefly what those are. An ethyl ester is a form of omega-3 fish oil where it's formed when the long chain fatty acids, the EPA and DHA are bound, IE, they're esterified, to ethanol rather than glycerol. So triglyceride is three fatty acids bound to a glycerol. So that's where we get that term triglyceride. Ethyl ester, it's bound to, it's esterified to ethanol. The reason that this is good is it results in a much more concentrated level of the fatty acids than a triglyceride form supplement. And the other important thing is their absorption, although they're much more concentrated, is actually still similar to long chain fatty acids. In test on the absorption, packaged into chylomicrons, transport throughout the bodies through the lymphatic system via an unsubsequent breakdown into free fatty acids. Carboxylic acid formulations are free fatty acids. They're non-esterefied. They are not bound to either glycerol which would form a triglyceride or ethanol which would form an ethyl ester. absorbed with greater efficiency and on paper they're supposed to have they have i.e., their hypothesized to have significant degrees of bioavailability than ethylester. They don't require breakdown by lipases, and this is assumed to mean that they're better, right? When it comes to ethylester, if you trace a lineage of studies that have used ethylester formulations of EPA, a kind of clear chain of an evidential picture emerges. We can go back to the gizzy prevention only for whatever it is in Italian, gizzy, pee, trial, secondary prevention trial, gave participants either a placebo, they'd had a myocardial infarction within the preceding 60 days, randomized them to a placebo or a combined EPA-DHA ethylester, and there was significantly lower risk of further cardiovascular events in the ethylester EPA-DHA group. Now, why do we get to EPA rather than the combination of EPA and DHA? We have researched on by Philip Calder, this is going back to the early 2000s as well, published that looked at uptake into arterial plaque of EPA and DHA, and what those studies showed was rapid incorporation of EPA into arterial plaque directly into plaque, and that was happening within about three weeks of beginning to supplement. The uptake of DHA into arterial plaque is minimal, minimal, and then we have subsequent studies, imaging studies, and otherwise more recent evaporate cherry, a body of evidence out of Japan as well, where it's not just that EPA is incorporated into plaque, it's having direct effects on plaque, it's improving plaque morphology, it's improving plaque stability. If LDL is lowered with a statin, that EPA is helping to actually attenuate and regress the volume of plaque in the artery, and then reduce it comes along, which was a trial that tested EPA-only ethyl ester in a dose of four grams a day, 25% lower risk of cardiovascular disease events, and everyone's thinking, "Okay, we've got a clearer signal in the noise here." And then the strength trial is published two years later, and strength used a carboxylic acid, free fatty acid form of EPA. It found no difference over the course of its follow-up. And as I said earlier, it was terminated early due to lack of benefit. But here's the thing, the strength trial, and people would stack strength against reduce it and see, "Ah-ha, we have a conflict." It's like, "No, you don't. You have two completely different interventions." At every level of the factors that I just mentioned, you have one that used EPA-only. EPA is the operative fatty acid we are interested in for cardiovascular prevention, because of the aforementioned incorporation into plaque stabilization of plaque regression of plaque, etc. Direct impacts on plaque morphology and stability. So, DHA is not doing much in this picture. So that's the first difference. Reduce it EPA-only, strength EPA and DHA. Then you have the formulation. You have the actual quantity. As a result, you've got four grams of purified EPA in the reducer trial. But because you've added DHA to the formulation in strength, you have a lower absolute intake. It's a four gram per day total dose. And then you've got the delivery method related to the formulation, which is ethyl-estroform. The form that has previous, a lineage of research, showing that that is beneficial, including an open label trial in Japan, the Gellist trial, versus this free fatty acid form. And what's interesting is even though the free fatty acid form was hypothesized to be more bioavailable, in fact, the achieved plasma levels of EPA in strength were not close to what was achieved in reducer. So, in mixing the mixed formulation and strength, you have lower absolute levels of EPA compared to EPA-only intervention trials. And you have a lower uptake and you have a less of a deep of an increase in plasma EPA levels, despite the fact that your product is supposedly more bioavailable. And in reducer, plasma EPA levels strongly correlated with reducer risk of cardiovascular events. In the strength study, there was no such correlation. And so strength to me was a trial that was not evidence that reduce it was like a false positive or any of the other previous evidence using ethyl-ester were false positives. It was a trial conducted on a false premise. The prevence was that a carboxylic free fatty acid formulation would be more efficacious based on two factors. One, their absorption doesn't require a full fact containing meal or anything like that, and two, the greater bioavailability. And those two do not hold. The hypothesized absorption might not have been particularly desirable because what I think actually happened in strength from a metabolic perspective is that we know that omega-3 fatty acid, if you take higher omega-3 fatty acid supplementation, you will often see greater omega-3 oxidation in the liver. And I think that's basically what happened. I think you just had oxidation of these preferential oxidation of free fatty acids omega-3s in the strength trial. They never made it to the arteries uptake in the arteries in the way that you see in ethyl-ester. And so by binding it, you have that absorption similar to triglycerides, potentially a factor in their ultimate incorporation into arterial plaque. Strength was a wasted opportunity that field should have, in my opinion, sought to replicate reduce it with an EPA-only ethyl-ester. They didn't. They went a different direction, confused everybody, but in fact to me, as I'm saying, strength to me was actual evidence in favor of the lineage of research that I'm talking about for ethyl-ester formulations of EPA. I think we can relatively discount DHA because of the incorporation into plaque. I think delivery method, the formulation clearly matters. And strength got it wrong. And that's on them in terms of their design. But it's an example of where everyone else was thinking, this just adds to the confusion and this is evidence that we just kind of, there's a conflict so to speak, quote-unquote, on the evidence. It, to me, once you started to dig into the differences in the trials, it actually became clear that it kind of reinforced my perspective that with EPA enriched purified EPA ethyl-ester, that is the intervention. That is it. Like, forget testing other things now. Continue to progress our evidence base with that as the drug, so to speak, of choice. And I think that's very clear now. Yeah, I mean, as we've mentioned before, often things that are seemingly irreconcilable at first, and that seem to put out confusion, actually, as you dig through them and look at them in the appropriate granularity, they start becoming reconcilable once we start noticing some of these things. And so based on what you said, though, there's obviously all the takeaways specifically about that omega-3 evidence base. But more broadly, what are those kind of main couple of lessons that you leave people, because you mentioned around, sometimes we can fall into this nihilism, right? I've like, oh, well, this just doesn't have any effect, right? Then you talk about this importance of how we actually should think about null findings. And there's a kind of lesson in that as well. And then obviously, all of that relates to the specificity of any research question. This becomes a perfect example of that. So for you is that kind of one of those lessons in around just how we think about null findings generally within science as opposed to defaulting to nihilism or just throwing our hands up when something happens. Yeah, that's exactly it. It's the kind of superficial evaluation. Because when people talk about this evidence, they simply use the term omega-3, right? To articulate to people how ridiculous that is, given the importance of things like delivery method formulation, the choice of fatty acid, etc. Imagine we just always discuss cardiovascular drugs. That's all we talk about, right? And we say, well, actually, when you look at the research on cardiovascular drugs, it's really inconsistent. I can make that statement because, for example, CETP inhibitors have been relatively underwhelming compared to statins. So someone would obviously interject and say, this is ridiculous. You're just making a category error. You're discussing this overarching umbrella term without saying that actually statins are really bad. This same principle applies here, but people don't think like that because it's nutrition, because it's nutrients. You cannot just talk about omega-3s in cardiovascular disease. There's no granularity there. There's no specificity to the type of fatty acid that is a delivery method, etc. So for me, this is when an old trial comes out, you have to get into the weeds with it to really understand why is this study different? And when you do that, with the strength trial compared to EPA ethyl-Ester research in this area, strength goes from being just an "noel finding" or a conflict in the evidence. It's not a conflict. Strength is a perfectly explainable outlier trial, not a standalone falsification of the omega-3 cardiovascular disease hypothesis, and too many people treat it as the Lacher. It's an explainable outlier trial that proceeded from a false premise and conducted an intervention. It's an expensive waste of money because it was thinking too mechanistically about fatty acid metabolism and actually not mechanistically enough because maybe someone could have spotted that this might just go directly into liver fatty acid oxidation. And to me it is a really good example, as I said, of the value of null trials because in in the formulation and the delivery methods of the intervention and strength, it bolstered the my thinking that actually there really is something with EPA ethyl ester trials and that is not random and it's actually reproducible and it also importantly made me more focus on because in strength there was still a triglyceride lowering effect of the intervention. The reason I emphasize all of the plaque related impacts of EPA incorporation into plaque from ethyl ester supplements. All of the EPA incorporation studies as far as I can see, sorry the plaque studies that have measured that have used in ethyl ester forums as far as I can see. The reason I stress that is because I actually think the lower risk of cardiovascular disease has as much if not more to do with the direct impact of EPA on plaque morphology instability as it does with the triglyceride lowering effect for which they're primarily indicated and that is a hypothesis that I will stand over based on the imaging studies that we have to date. So it was like, oh, this is actually a great example, not a falsification of a hypothesis, but of a null finding that shows you that the direction you are looking in is actually I think now I became, as I said, this is my example of something that I was like cautiously optimistic but think that in my head I was like people just aren't doing due diligence enough to reconcile the kind of conflicts in this evidence and strength actually cemented me much more into I think there is a real benefit here and that this is support, I mean cardiovascular treatment guidelines have ethyl ester EPA indicated now for people who have residual high triglycerides following a statin reduction of their LDL, but I hope as some of the imaging studies evolve that actually evolves to potentially just a juvent use based on the potential for improved plaque related like morphology instability and lower risk of hopefully rupture into the line and enhancing plaque regression and people who get very low LDL levels. Hopefully this serves as a way to be even more precise with those future research questions because it actually is informative if it's taken in the way that you've outlined. So let's keep moving, let me get to the third specific update that I was going to make and this is one where a few years back anyway at least we've done a full episode around dietary cholesterol and much of that has actually been quite informative in what I'm going to mention here now, particularly some of the old trials that you had raised in that, but in general the previous position that I would have been more partial to is one that I think I don't know about you, but I see it's probably quite a prevailing view even amongst people who are normally relatively well informed around these topics, it has become assumed to generally be true or as a heuristic that dietary cholesterol has a little or no impact on blood lipids and therefore cardiovascular disease because it doesn't have this a blood lipid level or blood lipid increasing effect that may have been presumed previously and then hence there's all the other conversations about changes in guidelines and so on. And obviously there's a part of this that relates to that blood cholesterol regulation that's going on the liver and all these different mechanisms that we would have talked about previously that indeed if we are looking at the dietary components that influence blood cholesterol levels or LDL cholesterol levels within the blood, there are other factors that are going to much more strong we have an impact on that notably saturated fat or maybe more specifically that ratio of polyunsaturated to saturated fat in the diet and maybe dietary cholesterol comes far away beyond that to the point where like I said that kind of prevailing view was this is so inconsequential, it's not something that needs to be considered and then that eventually over time gets morphed into dietary cholesterol has no impact on blood cholesterol levels which from a at literal standpoint just isn't actually the case and so there's probably two position or two aspects that are needed to nuance this one that I think that we've mentioned previously and I think you've actually written really well about is that the adding of more dietary cholesterol in the diet the impact on blood cholesterol levels is going to be dependent on that p to s ratio or we can also think about the total amount of saturated fat and so if we don't factor that in we can't really think about the impact of dietary cholesterol and then the second area we can look at is the effect of a per unit addition of dietary cholesterol on blood cholesterol levels and that's going to be different based on the baseline level of intake so if someone has a low or a high baseline level of intake their response we're going to see on their blood cholesterol levels is different and again this is one that isn't maybe a pre-shared or discussed as much and everything gets lumped in to say it has no effect and so I think some of the main work that you would have discussed running to the p to s ratio would have been around the Sean Felden colleague study from the early 1980s which is really nice work and I'll let maybe you elaborate more on some of the details if you wish in a moment but essentially we had this situation set up of comparing that impact of dietary cholesterol they from their baseline of 300 milligrams per day to either a higher intake of 750 or 1500 milligrams per day and so you had that in the context of four different p to s ratios so going from low so from 0.25 all up to a high p to s ratio 2.5 and so notably for people the higher that p to s ratio means it's higher in polyunsaturated fat and lower in saturated fat and so the one that's going to be the highest amount of saturated fat and relatively low in polyunsaturated fat i.e. the most likely to raise blood lipids is going to be a low p to s ratio and so in this intervention they have them on the 300 milligrams and then they have them put into one of these high cholesterol diets with the addition of eggs in the context of different p to s ratios and ultimately what you see there is with a situation of a low p to s ratio in other words high saturated fat low polyunsaturated fat you see a really meaningfully big jump in LDL cholesterol when they are going to those higher dietary cholesterol intakes whereas you don't really see much of that LDL cholesterol change in the context of a low saturated fat high polyunsaturated fat intake i.e. a high p to s ratio and so this was giving us that indication that background context of the amount of saturated fat and polyunsaturated fat in the diet is important and there's probably some other feeding trials that we can discuss that show similar things the second element that wanted to raise here and we can talk a bit about this as well is like i said that baseline level of intake which earlier in the podcast you've already referenced as being important for any nutrient we're considering and here it seems to show that blood cholesterol response to dietary cholesterol it's actually really strong at those kind of really low intakes so if we take the lowest intake we can get and then progressively over time that the per unit increase is getting lower and lower so if someone thinks of this as that curve going from let's say a theoretical zero intake up to like a hundred milligrams per day you would probably see this noticeable jump in LDL cholesterol that 100 to 300 milligrams per day you still continues to see a rise but it's a smaller additional rise in terms of the steepness of that curve and then in that maybe 300 to 500 600 range we're maybe seeing very small further changes and maybe going beyond that the curve just doesn't show any increasing effect and so we have this kind of depending on someone's baseline intake we would have this type of relationship or hyperbolic relationship as we might determine mathematically and so this also makes the context of if we're saying dietary cholesterol has no impact on blood cholesterol levels well again it depends on what is someone starting at and the issue why maybe this maybe you can speak better this that I but where we might see this is if we try and think of some of the areas where particularly in epidemiology where we are thinking out where would we see those really low levels of baseline intake it's going to be very hard to get that in the general population and so we're maybe not being able to detect that as much even though we know that relationship exists and then obviously the other important part of this pragmatic perspective is where we tend to have dietary cholesterol rich foods we tend to have those that tend to have saturated fat normally as well and so having these hypothetical situations where would have very. low saturated fat, for example, and very high dietary cholesterol, we can do experimentally, but maybe isn't directly correlated to the types of high cholesterol diets people are consuming and saying, well, let's just have loads of eggs and bacon or whatever. So for me, that kind of change is being well, dietary cholesterol does have an impact on blood cholesterol levels. The magnitude of that impact depends on those factors, I the PETS ratio, as well as the baseline level of dietary cholesterol, and then we're going to see those bigger jumps when levels are very low and adding in dietary cholesterol. I don't know if there's any other parts of this whole kind of conversation that you think are worth adding or other pieces of nuance that you would fill in some of the gaps. Yeah, there's a couple that come to mind because this is still, well, I obviously agree with the conclusion that like yes, dietary cholesterol clearly does have an impact on blood cholesterol and LDL levels. I for an area where we have lots and lots of really controlled human feeding studies, lots of them are now dated and you're going back to like the 60s and 70s and 80s and even to the 50s with Keezy and the metabolic ward stuff that he was doing. And to me there are some gaps that were kind of never filled in the evidence and they temper for me even some of the kind of more recent synthesis of evidence that you get. Like there's that Vincent and colleagues paper which is quite highly cited as oh look we've got this we've got this curve that goes from 0 to 400 or something like a 300 and above that there's this plateau. So yeah, going from 0 to dietary cholesterol to say 200 you'd see this big increase but going from say 200 to 400 you'd see very little and that's really relevant in studies that might suggest there's no meaningful impact even if it's even if there's a small magnitude but there's also the relationship with fat and total fat and a lot of the synthesis of this are taking primary trials that have not well accounted for all of those moving parts. So there's always a degree potentially in my head there's always a degree of artifact to some of these curves that I don't know that they're as represented of all the time of the true relationship such that it is one thing on that is for example total fat content. So we typically now we correctly think that baseline dietary cholesterol is important but baseline total fat may also be in addition to fatty acid composition important and I'm thinking specifically back to Kees's research where intakes because he did consider down to 0 up to about 6 to 700 milligrams a day and basically found that changes in blood cholesterol levels were independent of changes in cholesterol intake over that range from 0 to 6 700 milligrams per day but that conclusion was based on adding the very high dietary cholesterol intakes to diets that had less than 15 grams of fat. So that's always been fascinating to me it's not something that's been well filled in in that literature and we also have a very tightly regulated system of cholesterol homeostasis which is very different from person to person. So in some of the feeding studies conducted in the kind of 60s and 70s for example there's a really nice study that was done by Nessel and Poiser in 1976 or 77 and there was basically controlled feeding intakes of different levels of dietary cholesterol and looking at the relationship between endogenous cholesterol synthesis and increased blood cholesterol levels. So there's three different aspects of cholesterol here they have the cholesterol coming in through the diet and looking at fatty acid composition then you've got cholesterol being synthesized endogenously which is the majority of our daily cholesterol circulating levels and then you had changes in blood cholesterol levels and what was interesting is that in individuals who showed little to no change in their blood cholesterol levels despite high exogenous dietary cholesterol intake those were people who had very efficient compensation such that they would suppress endogenous hepatic cholesterol synthesis in response to endogenous exogenous dietary cholesterol intake. As a result you get little meaningful change in their circulation cholesterol levels. Individuals who see increases in their blood cholesterol levels are those who fail to down-regulate endogenous cholesterol synthesis in response to exogenous dietary cholesterol intake and that is a factor that is modified by genetics of course but also modified then by the fatty acid composition of the diet the ptoes ratio as you mentioned particularly. So I've just always retained this kind of interest this is one of these areas where I still I obviously accept that dietary cholesterol does have an impact on cholesterol and blood cholesterol. There is interesting and this is from animal models looking at LDL receptor activity in the context of different dietary fatty acid ratios adding dietary cholesterol. So what we see in terms of the attenuating effect of a high polyunsaturated fat to low saturated fat ratio attenuating the impacts of high dietary cholesterol intake on subsequent blood cholesterol levels is explained in part by some of these experimental models that basically showed that the higher you get the polyunsaturated fat content in the context of a certain amount of dietary cholesterol you retain a degree of expression of the LDL receptor but it's not entirely complete if you removed all cholesterol from the diet of an animal that LDL receptor activity would return to maximal. So it's not a full compensation but it does explain what we see in humans. The reason and still unconvinced of an independent causal relationship with long-term cardiovascular disease is because of these individual factors that I've described in terms of the relationship between endogenous synthesis, exogenous intake and the modifying effect. So I see cholesterol, dietary cholesterol now, not necessarily as independently causal but as itself a modified a factor that is modified in the diet as an exposure. So it's like conditionally associated in certain contexts and I think controlled feeding studies bring us to that conclusion. All of which is to say for me this remains one of my open conundrums that I have no hard position on in nutrition. I will not go hard to one way or the other on this because as voluminous as our evidence base is there are open questions that were never quite adequately answered and I yeah like I said I do not swing one way or the other too far on this. I come to no hard conclusions on this question and that for me is still a kind of an open enough question. Because that is worth noting right that like I said while there is the position that dietary cholesterol has zero impact on blood cholesterol levels which as we've noted it isn't technically correct. There's maybe some that would go far further than being in this position of softening that and would go all the way to well the best intake of dietary cholesterol has to be the closest one could get it to zero based on some of these same reasonings whereas again like you said how far someone needs how much evidence directly supports that there would be the type of benefits in terms of health outcomes to focusing specifically on dietary cholesterol to a minimal point. Again it's a separate thing to the point of saying well it has a relationship but it does have an has an impact on these levels. It's different than to go to the point of like you say making this every strong central theme of let's get this to zero. Yeah yeah and this is really important because the studies that largely are relied on when people say oh well there's this change from zero that's quite a large magnitude but then it's the change from say three or four hundred that is much less and that matters. They're based on models changes of dietary cholesterol and direct interventions that tinker with these moving parts don't always necessarily confirm that like there was a study published last year by Carter and colleagues crossover trial so really nice study that basically compared a high cholesterol low saturated fat diet so 6% saturated fat 600 milligrams a day of dietary cholesterol a low cholesterol high saturated fat diet so 12% saturated fat and your average population intake of 300 milligrams of dietary cholesterol and then the company high cholesterol high saturated fat 12% saturated fat 600 milligrams dietary cholesterol and what was really interesting is that in the diet with high cholesterol but low saturated fat which in terms of their achieved intake was 8% you actually had a small decrease in LDL cholesterol despite that intake of 600 milligrams a day so that brings us back to this effect of dietary cholesterol is really not truly independent but it's always been going back to Kees's research the effect of dietary cholesterol is conditional upon other moving parts in the diet Fadiasa composition being probably the most important of those. And in that study, predictably the combination, the high saturated fat, high cholesterol, to high it resulted in the greatest increase in LDL cholesterol. You know, I reminded keys some this up. There's a quote from keys in one of his reviews. It's a 1965 review. And he wrote, this is the quote, "For the purpose of controlling the serum level of cholesterol, "dietary cholesterol should not be completely ignored, "but attention to this factor alone accomplishes little." And for me, that's always been like, "That's it, that's it, he's nailed it again." It shouldn't be ignored. And I always would always emphasize in individuals at high cardiovascular risk who need to maximize LDL cholesterol lowering. I do think it's a good idea if they have a high cholesterol intake to obviously lower that. But my priority is always polyunsaturated fats replacing saturated fats. And then additional lowering of dietary cholesterol may be required in some people after that step, if necessary. And again, I think a point we've made before is if they started that first step that you just mentioned of actively trying to lower saturated fat replaced that with polyunsaturated fats, those types of food choices. In general, not only are you going to have the reduction on LDL cholesterol, but in general, it will probably change dietary cholesterol intakes as well. You would imagine depending on what types of foods people are consuming to have these higher saturated fat intakes. The other point, I think I can't remember the author's name, but there's a couple of the controlled feeding trials that you had mentioned in our previous episode on dietary cholesterol, where rather than doing a modeling of effects, they actually looked at reductions. I think it was from 340 down to 140 milligrams per day. And the degree of LDL reduction from that drop was something like 0.1 millimoles per liter. So for example, we've been in the US, it's like four milligrams per deciliter. And so again, an effect, but again, placing that in the context of, as you said, how much focus you want when we're seeing these types of changes? And again, it depends on who we're talking about, the level of risk, et cetera, but just kind of mind there. Yeah. And it's like I said, although I come down, I'm really unconvinced. This is just my personal take, but I personally don't think that the language of causation for dietary cholesterol specifically is appropriate. I don't think we can say that there's independent causal effects on cardiovascular disease. And the epidemiology is very population dependent. And because we don't see that cross-population replication consistently, I think that's another whole, even at the epidemiology level. But I'm not saying that it isn't either, if that makes sense. I'm very much, I just don't use, I think people will get into mechanistically describing like cholesterol absorption and stuff like that. And yes, we do get, but even that is, is itself conditional on the balance of steriles in the diet. And people often forget, I think we alluded to it before, but in the 50s, you had two hypotheses. You had the fatty acid hypothesis, which was Keese's pursuage, which is that the fatty acid composition of the diet impacts blood cholesterol levels. You also had the sterile hypothesis, because people looking at the polyunsaturated fat-rich oils used in fatty acid research were like, these are also rich in phytosterols. And so you had this whole sterile hypothesis that actually it was the displacement of cholesterol by other steriles. So you have all of these moving parts, in terms of fatty acid composition, steriles, in individuals' genetic ability for modifying endogenous cholesterol synthesis. You can see controlled feeding studies in the 60s where they shovel dietary cholesterol into people and saw almost no change in blood cholesterol levels, because they were down-regulating dodging this cholesterol synthesis, and they were recycling and eliminating the circulation cholesterol that was present at the level of the intestines and hepatic inter-hopatic circulation, such that there was almost little to no meaningful retention of the cholesterol that was consumed in the diet. All of this is fascinating stuff, and all of this to me tempers any sort of hard causal conclusions that people come to. I think that is the perfect way to sum that up. We will probably revisit that again. We have another thing to add to the list for a future episode, I think, because like you said, there is some refastening stuff in this area. As we start getting towards the end of the realm, let's turn our attention to what you had highlighted as something where you have not changed your position based on more time, more evidence, more understanding of a particular topic, and instead maybe the opposite. You mentioned sodium, and so maybe again, can you recap for us what that kind of position is, and then where you are now with it, given what has happened over the last number of years? Yeah, so I have, I think probably banged on the drone that sodium is what we think it is for a few years now in terms of a leading dietary nutrient of concern for cardiovascular health. But of course, that has been something that has itself been contented and disputed. And that has largely been based on the evidence that has come out from kind of several analyses, meta-analyses, usually of cohort studies that suggested the mythical "J-shaped curve". And this essentially is a contention that low sodium intakes are associated with risk of cardiovascular disease in addition to high sodium intakes. And actually the sweet spot for health, the lowest risk, is actually at a moderate sodium intake, somewhere in the region of 7, 8, 9 grams of salt per day. And that's the sweet spot because that's where we see lower risk in some of these analyses. And this has always been a reflection of the methods used to measure sodium and studies that have used repeated measures of, if the first thing is you need 24-hour urinary sodium and potassium collections. The second thing is that you really need repeated measures of that in order to attenuate the measurement error from inter, from within person day to day variation in sodium excretion. And we know that studies using just a single 24-hour measure, produced by us risk estimates, studies using a spot measure where just a single urine sample is taken, and these predictive equations are used to then predict their 24-hour sodium excretion are even worse spot measures are horrible in terms of their validity. They're just awful in terms of their predictive value. They're not predictive, they're not predictive equations because they don't predict 24-hour intake. So they're just equations and they're awful. And yes, people could make this kind of case that, well, the actual outcome, yes, sure, lower sodium, lower high pretentia, lower blood pressure. But that's only an issue in people with high blood pressure. And really, we don't have good outcome data. And there was a 2011 meta analysis by Prath McGregor and Finghay. And that looked at longer term outcomes of some trials, like trials of hypertension prevention and the tone trial. And it did show around an 18% lower risk of hard cardiovascular endpoints. But as a meta analysis, it wasn't particularly overwhelmingly convincing because it was based on a relatively small pool of, I think, three or four studies. And not a huge, not a huge, like most of the studies in this area were still looking at intermediate endpoints in terms of blood pressure. Then we have more research that shows actually no lowering sodium that isn't just beneficial for blood pressure in hypertensive individuals that basically lowers blood pressure across all individuals independent of their hypertensive status. But then it's still coming back to this question of outcomes. And for me, what nailed this kind of home was two studies published in the same year. The first was the SAS study, which was a large sodium substitute, salt substitute intervention trial in China. And the SAS trial was basically an intervention which, for salt substitutes, they're basically a usual table salt, is 100% sodium chloride. And these are 75% sodium chloride, 25% potassium chloride. And they had 21,000 participants followed over about five years. And like, it was relatively ridiculous about the SAS study is just the amount of outcome events. It was something like 4,000 participants died during the trial, various cardio. They were individuals at high cardiovascular risk at baseline, high stroke risk specifically. So that was a trial where you had measures of sodium excretion and potassium excretion, measures of blood pressure and stroke risk. All of those added up in the salt substitute intervention, 14% lower risk of stroke and lower urinary sodium excretion and all this kind of stuff. And then that was the kind of the first study that started to shift this. There actually is about a really large long term trial with hard endpoints published in the New England Journal of Medicine in 2021. But the real clincher for me was a meta-analysis, also published in 2012. 21 in the New England Journal of Medicine from MA and colleagues. And this basically included individual participant data from six long-term studies, four were prospective cohorts, but they had repeated measures of 24-hour urinary sodium collections, and two were one was the trials of hypertension prevention one and the trial of hypertension prevention two. So they had long-term follow-up data in 10,000 participants, and they had crucially for the analysis between two to seven repeated 24-hour urinary sodium collections in the participants. And what we ended up seeing was in the high versus low analysis, a 60% higher risk of cardiovascular disease in the higher sodium excretion category per 1,000 milligram per day increase in sodium. There was an 18% higher risk of cardiovascular disease events. And this, for me, given the sensitivity we have to have for method of measuring sodium, this really solidified my position because there's two factors that emerge. One is that not a single study, not one that has ever used repeated 24-hour urinary sodium excretions to model to to assess risk, has ever found anything other than linear relationship between sodium intake and cardiovascular disease, no J-shaped curve, no R-shaped curve, no S-shaped curve, no curve at all of any description. And so in the absence of that, I don't think that we can at all stand over any of the kind of U-shaped curve analyses because of their flawed basis in methodology of measuring sodium. And until a repeated measures 24-hour analysis comes along showing anything but a linear relationship, this is the evidence that we have. And for me, it's very much solidified that the relationship between sodium and cardiovascular disease is a linear relationship. And that in all individuals for their long-term health and cardiovascular health specifically, lower sodium intake is preferable to higher. And I think what would be particularly instructive for people, if just after listening to what you said, would actually be to go and maybe listen to the previous episode on sodium that you and I did several years ago now, because it not only does it touch on some of these really important aspects around measurement of sodium, the importance of these multiple measures and all these things that you've just discussed, but also to see that in that time, the consistency with some of these conclusions that you came to based on this work that has came out because I believe that was probably around, I think probably around 2020-ish, when we discussed all this stuff, whether it's a J-shaped curve and sodium, and that would be really in form, I think, for people to go. If they are interested in diving deep in on around this relationship with sodium and some of these outcomes, also, if they've interested in the sastrial, there's an episode with Professor Bruce Neal, who ended up that if people want to get more of the details on that. But yeah, I think that is a particularly excellent example around sodium and these outcomes where your position is stronger than ever based on the amount of work and across all these different lines of evidence that you've discussed. The perfect example, we are just coming up on probably the limits of what people are able to focus on. So I'm not going to belabor this too long. I'll mention it, but we don't really have to discuss it because it's a very obvious one. For something I haven't changed my mind on, I thought of what was something that has been relatively consistent even since the podcast started, and that was maybe energy balance as being that energy in versus energy out is going to tell us about essentially the energy storage within the body as the name suggests, as opposed to conflicting hypotheses that had been put up over time that would suggest the energy balance model is completely wrong and we need something else. And I think since the podcast started as time came over, we had the Kevin Hall studies in the metabolic war that looked at stuff. We had other types of trials like that compare different macronutrient distributions within energy match conditions. We've had more and more. We've had all the different hypotheses or predictions of those hypotheses from let's say these alternative hypotheses that were put forward. We've had their predictions tested in a range of different studies now, all of which have converged to essentially falsify that position really to the point where the energy balance model as is currently described. And I think there's maybe been some nuance added in some of those publications over the last few years is something that I think I strengthened my confidence in that position as opposed to changed. But again, we don't need to get into that too much because I don't know if it's a particularly interesting topic compared to some of the more nutrition specific things we've discussed and for a sake of time. But for me, that was the first one that that came to mind. Yeah, I think that's at this point, very Hall's work on this has been so convincing from multiple angles. We mentioned the predictive equation earlier. How well that holds up. The idea of some thermogenic fairy in the context of low carbohydrate diets is just untenable, I think even to touch on some of the kind of aspects of energy metabolism, I think we've discussed before that energy expenditure is a manipulable component that the thermic effective feeding is a manipulable component of energy balance is relatively just fanciful in terms of its thinking and based itself on a kind of flow of methodological approach to measuring thermic effective food. There was just, it was so clearly a mismatch, I think, between discourse that was grounded in evidence versus discourse that was grounded in, I guess, a degree of ideology if we're talking about kind of the Ludwig-Kara-Ebeling perspective on low carb diets and energy expenditure. And I think that the work that he did to build up a body of robust evidence that was able to say that the kind of the energy balance model, shall we say, of obesity is the model, like it is related to energy balance and not related to substrate utilization or oxidation or any of these kind of fancy little terms that bounced around the research for a while. And I will list for people a few of the different episodes that I've talked about that because there's a nice progression of as more work came out and we discussed individual studies related to that over the years and how that continuously evolved to again more strength than this particular model of obesity. So those will be linked up in the description box. But I think in terms of this episode that does us, hopefully everyone is still here with us after all this time and has enjoyed this as a change of pace of looking at these different areas and maybe reflecting on some of those and hopefully it has been thought provoking for those of you listening. And also a big thank you to everyone who has listened at any time whether you are a relatively new listener or whether you've been here for the past decade plus. Thank you for supporting the podcast. It's very much appreciated. I know it's maybe not the typical type of nutrition podcast and hopefully we have quite a unique audience. So I'm very pleased to have and thank you for sticking with the podcast and hopefully you're continuing to enjoy it. Certainly, well, actually, certainly not this we can probably say is the last ever episode of this podcast will not be back. We will be sticking to our promise of 600 and that is it. So thank you everyone for listening to this point. You will not hear from us again. It's been a good run. Go go and enjoy yourselves. Don't think about nutrition. We'll come back with a different topic sometime. That is the end of Sigma Nutrition Radio. Thanks everyone for listening on behalf of Dr Alan Flanagan and I. We shoot a good rest of your day and good rest of your lives because we will not be back. Take care. Everyone, thanks for listening into this episode and again, let me say thank you so much for supporting the podcast throughout its 12 years of existence to this point and for allowing it to keep running by supporting through whatever means you do. If you are interested in being a Sigma Nutrition Premium subscriber, then the information is linked up in the description box where you're listening right now. For this particular episode, you'll get quite a hefty set of detailed study notes covering all the concepts that Alan and I have discussed. You also get edited transcripts to our episodes. In addition to that, there's an ability to ask questions which we cover and ask me anything episodes. There's also a premium exclusive episode each month that you receive on the private premium feed of the podcast in whatever app you choose to listen to your podcast on normally. It is the way to directly support this work. If it seems of interest, if you're intrigued, then I think it might be something that is of use to you. If you do continue to listen to a lot of podcasts for educational purposes, then that Sigma Nutrition Premium is linked in the description or over on sigminutrition.com. and check it out, see if it's for you. If not, I hope you do join me again for the next episode of the podcast. And until then, thank you so much for being a part of these 600 episodes, and I'll talk to you soon.