Debate: Seed Oils & Heart Disease - with Tucker Goodrich & Matthew Nagra, ND | The Proof

a 35 reduction in risk of all-cause mortality and a 33 reduction in risk of fatal atherosclerotic cardiovascular disease is a small benefit well it also showed a large increase in cancer in the intervention group towards the end of it right you don't get to just ignore that welcome to the proof podcast a space for science-based conversation exploring the health and longevity benefits that come with mastering nutrition physical exercise mindfulness recovery sleep and alignment facts nuance and trustworthy recommendations minus the hyperbole hello my friends welcome back great to be here with you i'm your host simon hill i'm a qualified physiotherapist and nutritionist with an undergraduate science degree and a master's in the science of human nutrition in my career i've worked with many people including elite professional athletes to improve their health performance and longevity and i'm currently involved in research with a group of nutrition scientists in australia looking at dietary patterns and mental health today i welcome dr matthew nagra and tucker goodrich for a conversation or debate on seed oils specifically omega-6 rich seed oils so why have this debate well while the guidelines clearly do recommend consuming sources of unsaturated fats including vegetable and seed oils instead of saturated fat rich foods such as butter there is a very vocal minority that claim omega-6 rich seed oils are harmful especially for cardiovascular health while it's true this is a minority position and many advise me not to host this debate i'm a firm believer in allowing people to express their ideas and then letting the listeners you come to your own conclusion if someone thinks these omega-6 seed oils are detrimental to health i want to hear the evidence that they have to support their argument and i want to hear their positions challenged from someone who is across all of the research i also want to hear the opposite someone challenging the consensus ultimately each of us can then come to our own conclusions as to where we land on a particular topic in this case on omega-6 rich seed oils such as sunflower safflower corn peanut canola soybean etc so that's the background to this conversation and my intention here as this conversation goes for four hours i'll spare you from a long introduction please do enjoy and i'll catch you on the other side hello gents welcome to the proof glad to be here happy to be here simon thank you yes i've been uh very much looking forward to this one so i'm glad that we're finally able to uh sit down together and and do it today we have set aside roughly two hours for a conversation or debate uh whatever you wish to call it centered on seed oils i'd i'd like to start here by having the two of you briefly introduce yourselves to those listening and then before we hear your opening statements i'll quickly go over the details of the debate so that everyone is clear as to the topic your positions and the the ins and outs of the way in which this conversation will be conducted or how we intend for it to be conducted as agreed upon by the three of us in the weeks leading up to today so tucker if you could please kick things off here by introducing yourself to the audience that would be great uh my name is tucker goodrich i'm been called lately an independent researcher of health issues my professional background was in technology in the finance industry on wall street where i ran some big complicated systems um in the financial industry and got sick and then started trying to figure out what to do about being quite sick and i won't bore you all with the details of that but i started using the same tools i used professionally to try and figure out what was going on in my health and then once i fixed all those problems started trying to help other people and popularizing some of the research that had uh that i had discovered thank you uh matt over to you yeah i'm uh i'm a naturopathic doctor in vancouver bc i definitely have a big focus on nutrition in my practice but also outside of that i just really enjoy researching nutrition learning more about it i think it's a fascinating field and i think that's something all three of us can agree upon for sure um and you know that's why i decided to reach out to tucker and then obviously thanks to you for uh agreeing to sit down i know a lot of the times um it's easy to just you know avoid uh discussing those with those with opposing views but i definitely think it's it's a good way to move forward right i'm sure there'll be a lot of things that we agree upon and there's obviously a few things we disagree upon and i'm just looking to to learn a bit here and then hopefully uh share my perspective as well okay great so i'll go through a quick bit of housekeeping here to to run through the the agreed upon terms of this conversation and the debate topic along with your positions and then we can get into things so it's been agreed that this will be an open conversational style debate that you will avoid talking over one another this makes it uh for an easier listen for those tuning in that you will answer questions directly so no changing topics or avoiding answering when a simple question is asked that you will give the other person a chance to respond as opposed to rambling and making several consecutive claims at once that there will be no personal insults and i think this is a really important one and i don't expect that we will see any with either of you from what i uh know uh but you know we're here to discuss science and address arguments claims etc so please know ad hominem attacks and as the moderator my job is to stay impartial and ensure that these agreed upon guidelines are maintained throughout the conversation to the best of my ability if someone is not adhering to these guidelines then i will step in to get things back on track but my hope is that the both of both of you make this easy for me and and i have little if anything to say in addition to this we have allowed two hours for this conversation if we get to the end and both of you think we need more time then we can decide uh then and there and uh if we want to continue on we've also set aside five minutes each for opening statements and then we will try and keep it to roughly three to four minutes each going back and forth to allow each other time to respond before moving to other claims there's there's not going to be a timer there so that will be up to me to make a discretionary call on and finally there will be an opportunity at the end for a two minute closing statement from each of you how does that all sound to to both of you are there any objections or comments before we move to the debate topic and your respective positions just thank you simon for volunteering to moderate this yeah that's another thing we should definitely thank simon for those listening um we had a hard time scheduling everything and simon kind of jumped in uh when we were you know we figured out dates to work for tucker and i but we couldn't figure out a date that worked for a moderator so yeah thanks simon for hopping on it my pleasure i think we need more of these uh conversations so i'm happy to do it okay so let's move to the debate topic the the debate topic that has been agreed upon is it is more reasonable to believe that seed oils are beneficial rather than harmful for coronary heart disease risk i'll read that out again it is more reasonable to believe that seed oils are beneficial rather than harmful for coronary heart disease risk and tucker's position is that he disagrees with that statement matt's position is that he agrees with it and to be super clear we have defined seed oils as oils made from seeds not the fleshy part of the fruit so this would exclude olive palm avocado or coconut oils so seed oils refers to oils such as canola soybean sunflower etc gents is there is there any objections here nope just to note uh high linoleic is really what we're talking about since there are some newer varieties of oils that have been reformulated sure matt you're happy with that i'm not sure yeah no no okay uh it's also worth noting for the listeners that all references that tucker and matt will be referring to in this conversation have been provided to one another ahead of time references were first provided on them on march 8th and then they had an opportunity to review them and provide extra references to assist with their rebuttal by march 18th which was five days prior to today's recording i think there was a little delay on on the second set of references but overall uh they both did a really good job in a limited amount of time to share their references which will will no doubt make for a more productive conversation than just flying blind okay cool well let's move into the the opening statements i believe matt you're going to kick things off and then uh we'll hear from tucker okay so i'm just gonna share my screen first and then i can get started and just as a reminder i will be timing these so we can try and stick closely to the five minutes okay so just start any time sure i guess yeah so yeah i just wanted to start with you know this meta-analysis on uh polyunsaturated fats specifically replacing saturated fat done by hooper in 2020 and they found that uh increasing polyunsaturated fat in place of saturated fat uh reduced risk of cardiovascular events and there was a non-statistically significant reduction in coronary heart disease events as well um of course these were all randomized controlled trials similarly the american heart association presidential advisory paper on dietary fats which specifically chose trials that were not compounded by trans fats at least not to a large degree had sustained adherence for at least two years and proved adherence by measuring biomarkers of intake and they also found that there was a reduction in risk when replacing saturated fat with polyunsaturated fats now i understand there might be some questions or concerns around the finished trial but even if we remove that as i've shown here on the bottom is still a statistically significant result moving over to this one and this is actually a reference that tucker had cited but i thought some of the results here were interesting if we look at the results for heart attack whether fatal or non-fatal we see a borderline statistically significant reduction in risk with higher omega-6 intake and this is largely coming in the form of linoleic acid again and if we add in the oslo diet heart trial as i have on the bottom here it is statistically significant now without it we can leave it out it's still moving in that direction um and the reason that they excluded it in the first place was because omega-3 intake went up to a degree as well but one trial i think that deserves specific mention is the la veterans administration hospital study so this was a double-blind randomized controlled trial where participants were randomized to one dining hall or another and uh you know half of them were getting standard fare essentially a higher saturated uh animal fat diet and the other group were having foods that were meant to mimic the standard fare but were higher in linoleic acid and low in saturated fat so they'd take like defatted milk and fill it with vegetable oil and try to mimic the you know texture taste and all of that that way and ultimately we see a reduction in risk of fatal atherosclerotic events in the experimental group and specifically see benefit amongst those with high cholesterol at baseline so these would be the people who would be more likely to benefit from reducing their cholesterol levels so it does make sense and moving out of the randomized trials into observational research we see consistency there too so this meta-analysis of prospective cohorts found that increasing linoleic acid intake was associated with a lower risk of coronary heart disease events and mortality and they also performed an analysis uh where you report where they looked at replacement of either saturated fat or carbohydrates and similarly linoleic acid came out on top for those as well uh and moving outside of looking at you know polyunsaturated fats or linoleic acid if we look at actual oil consumption uh replacing one tablespoon of butter a day from this large prospective cohort with canola oil was associated with a seven percent lower risk of dying from cardiovascular disease and we actually saw a non-statistically significant reduction with corn oil as well around four percent reduction so um we're seeing a lot of consistency amongst the different types of studies and markers there and you know another way to look at vegetable oil intake is to look at biomarker levels so if you look at tissue levels of say linoleic acid it can be a proxy for long-term intake and we see that overall higher tissue levels of linoleic acid are associated with a lower risk of cardiovascular disease lower risk of dying from cardiovascular disease and again a non-statistically significant reduction in total coronary heart disease risk and this is similar to what we see with what's called mendelian randomization so these are people with genetically elevated um or genetically higher versus genetically lower linoleic acid intakes over the course of their lifetime and or sorry levels not intake over the course of their lifetime and higher linoleic acid again associated with a lower risk of coronary artery disease and the reason i present that sort of data is i'm looking at the data that makes up the top of the evidence hierarchy here what would typically be considered the highest quality source of data looking at biomarkers or intake and we see them all pointing in the same direction and i have no doubt that both tucker and i are going to have evidence to back up each of our claims but we're going to have to look at whose evidence is more applicable to the question that we're asking uh and better able to answer the question that we're asking and that is what effect do seed oils have on our coronary heart disease risk and i think that's about it for my intake i sped through that but hard in five minutes yeah you got there just in time but thank you for that so let's let's hand the the baton over to tucker okay let me just get my timer up um unfortunately i have a few more slides um if i'll probably put this up on my blog if anybody wants to come back and uh try and catch up on what i'm going over here just bear with me yeah and i'm sure we'll get into the details and stuff throughout anyway this is just a quick kind of intro so how i look at it but that's a good point that the the there will be links in the show notes to any resources from both tucker and matt it is more reasonable to believe that seed oils are beneficial rather than harmful for coronary heart disease risk i disagree with this and i will note that i usually refer to it as cvd so it basically works out to the same thing at current rates this is a new disease and it's important to note that this is a study from 1948 looking at the increase of heart disease in the united states they found a rapid increase across all states that they looked at another study looking at a rapid increase in coronary cardiovascular disease in the uk again rapid increase in cardiovascular disease and this is an ex intensive look at uh three populations white americans japanese people living in uh japan and america and african americans uh living in and then versus africans and what they found was that the latter two populations had high rates of heart disease relatively uh versus very low rates in their native populations this is important because it was a test of is this a genetic disease or is it an environmental disease and their conclusion was that especially for instance in japanese americans where they saw a rapid increase in heart disease in people who moved to the united states that it was clearly an environmental disease and not a genetic disease there we go so what changed during this time this is strictly looking at the united states we see a rapid decrease in the consumption of animal fats and an increase in things like shortening and margarine specifically soybean oil and canola oil along with various other seed oils saw a rapid increase over the course of the 20th century in the united states and then later around the world this led to a rapid increase in linoleic acid during the course of the 20th century the change in soybean oil consumption in the united states according to the stunt study was about a thousand fold and was far and away the biggest change in the food environment in the united states over that period seven countries study which was a seminal observational study uh conducted from 1958 to 1995 looked at some of these populations including the japanese and found interestingly that the best explanation was the oxidizability of ldl particles it was not ldl per se because what they found was a wide discrepancy between ldl levels in different populations and the cardiovascular diseases in those different populations so they had to find something else to explain it and they figured it was probably diet things like smoking specifically they talked about oxidized ldl and they noted that what causes our ldl to oxidize is polyunsaturated fatty acids in 1979 um it was found that ldl does not cause the first step of atherosclerosis it's only modified ldl in 1989 they determined that it was specifically the peroxidation of polyunsaturated fatty acids in the lipids and specifically linoleic acid which was breaking down in ldl and turning the ldl toxic now human rcts on fats and cardiovascular disease there are actually only three that specifically look at an increase in linoleic acid um all three of them showed harm um the three that showed a bit of benefit did a blended intervention of omega-6 and omega-3 fats and the latter two that showed a larger benefit included longer chain omega-3 fats mostly from fish oil the best that one can say about these results from hooper 2018 is low quality evidence that increased intake of omega-6 fats may make little or no difference to all cause mortality or cardiovascular disease events one means no effect for a risk ratio now this is after they inappropriately excluded the largest trial which showed harm that's the minnesota coronary experiment i might just interject to let you know it's nearly a five minute yes thank you i've got a timer in front of me only one study lowered linoleic acid and increased la they showed a 70 reduction in events and today oxidized ldl is the explanation for cardiovascular disease this is from a 2020 consensus statement from the european atherosclerosis society and this is a recent paper from harvard university looking at the benefits of olive oil and finding that it's the susceptibility less susceptibility to oxidation which of course the fat susceptible to oxidation is linoleic acid and that is it and then i may we'll go with these later but that's my conclusion is that there is no valid data showing benefit in human only in harms only harms and it's been addressed repeatedly in human rcts the pathway to harm from linoleic acid is not credibly disputed in experimental evidence okay um do you mind a little over there that's fine oh good um now is this like open conversation part i guess at this point yeah so this is the open conversation parts of it let me get back to the screen oh you stopped sharing for me yeah this is okay just just to remind you guys here so this is just back and forth conversational we'll try and keep it to three or four minutes give give each other time to respond and and try not to make too many claims all in one sort of strength yeah of course um and yeah so talk about mine yeah go ahead oh yeah i just have a couple questions for you just to better understand um you know how you came to those conclusions um so i'm wondering like are you able to describe for me the epistemic framework that you employ to arrive at your conclusions i read the papers what do you do well no so so that's not obviously i do that too but what i mean is like how do you evaluate the research like how do you evaluate what has or what is better able to inform our opinion on these matters uh versus uh because obviously we can throw research back and forth i'm just trying to get like what it is that your framework is clearly the highest level of evidence that we have are human interventions right i agree some of the res some of the papers that you showed confounded different variables right which has been common in research in this area so for instance the oslo study increased linoleic acid but it also increased omega-3 fats specifically via fish oil or i think it was also fish in that intervention so what one has to look at is a consistent pattern of evidence across a body of research try and figure out what the confounders are and then draw your conclusion yeah yeah no so that's yeah that's just kind of what i was um wondering um and then and i definitely want to talk about a lot of these individual trials but the other thing i'm interested in i'm willing i'll give you my answer as well but what would you need to see to change your view on this topic i would like to see some in i would like to see some evidence like any evidence i mean they're they're like quality high quality evidence yeah and i mean as i was getting into we've got a lot of epidemiological data a lot of it's based on really poor evidence and a lot of it's not evidence at all you know we've got the nurses health study which claims to show benefit from this and it's a completely basic garbage data set as i went over in those two slides that i skimmed at the end they're claiming that they discarded any um ffqs where the range of calories consumed was implausible and they defined plausibility on the low side as 800 calories for men or 600 calories for women over a 30-year study so that's an implausible data set and unfortunately they included data that was clearly erroneous and not something that people could survive on and it calls the entire body of evidence into question okay the nhanes study shows the exact same thing the level of implausible data included in the nhanes epidemiological study is from 50 to 80 percent now that's not saying it's accurate that's just saying you know that's not saying the remainder is accurate that's just saying it's plausible so if we're going to start looking at data that at its face is you know absurd then we shouldn't spend much time on it right people can't live on 800 calories a day for 30 years or 600 calories a day i'm sorry 801 calories a day and 601 calories a day for men and women respectively over the course of 30 years and there's with data problems like that there's no way to correct for that right they claim that they adjust for it they claim that they toss the data out but data errors like that shows that they have a fundamental process error and that calls every piece of data in their data set into question so their conclusions can't be relied upon okay so so yeah like i agree that there are going to be um whether it be even implausible intakes or they're going to be outliers within the data set 100 agree with that what we're looking at with epidemiology though is we're looking at are those who say based on say food frequency questionnaires they eat the most of x y and z and less of other things do they tend to aggregate in that group versus those in say the lowest quartile or quintile or whatever it happens to be and that the same the same goes for other exposures like other potential confounders we're essentially just trying to tease out on average is that what we see in the highest intake group now but you can't do that if you don't know what their intake is and okay and what i want to point out and if i can just share my screen here um what i want to point out is that we have multiple analyses actually looking at the concordance rate between epidemiology and randomized controlled trial data so if if it is the case that nutritional epidemiology is completely debunked doesn't doesn't reflect um or it doesn't actually translate to outcomes whatsoever but we have this uh comparison of over 950 randomized trials over 750 cohorts assessing 97 diet disease outcomes or if we limit to meta-analyses 83 diet disease outcomes and they're concordant in 69 percent of the the cases clearly it's right a lot more often than it's wrong and similarly what that shows us is that it's always wrong what do you mean i mean i looked over a couple of your studies for starters the data specifically for linoleic acid was you know first off they never validate it against actual data right they never go there listen they're never going and measuring what people take every day say an award where they're measuring their food right the most that they do is take ffqs and then say okay we're doing some statistical leisure domain on the data to set to you know try and show that it's internally consistent but they're never actually correlating it to real world intakes right they're not measuring people's foods and then correlating that to food frequency questionnaires i mean the nhanes doesn't do that the harvard data set doesn't do that the epic data set doesn't do that the pure data set doesn't do that right they are taking their starting point is a questionnaire now i think this is a tangent this is irrelevant to what i was saying this is crucial because if we can't rely on the data we can't rely on their conclusions but that's not that's not i'm not talking about the methodologies used in the um epidemiology what i'm talking about is the results that we get from epidemiology that are using these methods that you're saying don't work they actually are concordant with the results we see from randomized controlled trials the majority of the time that is what this says and it's not the only analysis i think that is really key this is one analysis over 950 trials over 750 cohorts we have another analysis uh this one was somewhat less it was like 40 something diet disease outcomes um hundreds of uh cohorts and randomized control trials i can't remember the exact number there um then we have another one from morphe 34 associations again two-thirds of the time the results are concordant and so this is even let's say let's see yes but unfortunately the fact that they're concordant two-thirds of the time doesn't tell you anything about whether a specific study is going to a specific epidemiological study is going to be validated in an rct and in this specific case we have three rcts that shows harm so it's really irrelevant what the epidemiology says because we have the actual studies well no so would you agree then based on that that and we'll get to specific uh outcomes in a second here but would you agree that the majority of the time randomized controlled trials produced i mean sorry uh nutritional epidemiology produces results that are concordant with randomized controlled trials it's irrelevant no but that's that's that you're talking about the question you're going that's not it's not i'm talking about the quality of uh of if you agree that randomizing fold trials are the highest tier of evidence as you said human intervention drops and uh epidemiology is concordant most of the time with randomized trials like all i'm asking is do you agree with that statement it's an irrelevant statement that it's just not going to go down this tangent i mean i don't we agreed that we would answer questions yeah relevant questions we also agreed we wouldn't go off on tangents and going off on you know it's i think it's fair to observe that epidemiology is unreliable and often produces unreliable outcomes i mean but again like that's not what i'm asking here tucker if we look at linoleic acid specifically if you want to highlight that specific marker here linoleic acid cardiovascular mortality concordant between rcts and epi right there so if you want to say that oh we're not interested we're not looking at the specific exposure of interest we're seeing it right here in this paper so again i'll just have they're looking they're looking at you're looking at studies here i mean they mentioned the hooper paper the hooper paper in correctly excluded the biggest linoleic acid rct so it's again it's garbage data so we'll we'll talk i plan to talk about mce but again my question is and this is i think the third or fourth time is do you agree based on the data the three analyses i've just shared and there's actually a fourth but it was i figured it was too late to share it by the time i came across it um there are three analyses here all showing that the majority of the time uh nutritional epidemiology is concordant with the randomized controlled trial data do you agree with that is that a like just a simple yes or no no i don't you don't well then why do we see a 69 percent concordance consistently roughly uh you know give or take a few percent depending on the cause this is it's just a meta analysis without going through and auditing the thing but you had two weeks to this nobody could audit this in two weeks well no i mean we're just talking about linoleic acid really you could have highlighted yes and we if we look at the high quality data that we have the human rcts we see harm consistently so at that point it doesn't really matter i mean the hierarchy of evidence you showed puts observational studies below rcts yeah the way the scientific process works is that you do an observational study and then if you find an interesting outcome you attempt to confirm it with experiments and that's exactly what's happened in this field of research right and even the study that you agreed that i produced hooper 2018 that you agreed was a valid study says they show no benefit and that's no benefit after they exclude the bulk of the data we'll we'll get to that again i i absolutely plan to address those but the fact of the matter is on the whole now if if you don't have an objection to the methodologies used here i mean in a debate setting i think i've already said that i an epi an objection to the methodologies used here you're looking at observational studies no that's that's not that's not actually what it's it's not it has nothing to do i'm not talking even one bit about the concordance between observational studies and experimental studies exactly yes and and i'm just asking the fact that there's an overall statistical number of correlation based on this study doesn't tell you anything about a result in a given field sure so it's irrelevant no no it's not irrelevant i'm just asking if the majority of the time we're looking at 87 if we look at the meta analyses or or 90 or so 83 or 97 diet disease outcomes so so many different exposures in them in 69 of those cases we're seeing concordance so so my we're not discussing six we're not discussing those cases we're discussing one topic here i understand so let's focus our discussion on this particular topic no but the the problem is that you're essentially saying that nutritional epidemiology as a whole is not trustworthy and you're carrying that over to the discussion of linoleic acid or vegetable oils well you're looking at the data that specifically pertains to linoleic acid yes i don't think it's trustworthy okay that is that's i plan on we're discussing today so let's absolutely stick with that well i mean i don't i i don't want to drag you in simon but what do you think it's a fair question i think it's a reasonable question but i i also do understand tucker's position here and i tend to think that where this is headed is towards your disagreement as to what the randomized control trials are showing and not showing but i think you've made your point matt and i think it's probably best that we can just move forward otherwise we're going to end up spending 20 minutes on on similar sort of things going in circles and not really get anywhere okay yeah so then tucker i mean if if minnesota that's the one that you believe to be you know the best you're the one that they erroneously left out um in the uber analysis correct yes yeah so i i think it would be a good one to discuss then if it's kind of the top top tier for you um so i'm just wondering what do you think about the dropout rate or the attrition rate in this trial i think they had turnover yes they also had they also had on average the mean intervention period in this study was the longest of any of the big studies i think 68 of the people who've been in any one of these and studies were in this particular study so we're looking at the vast majority of data that we have in this uh topic is in this particular study so excluding it for an arbitrary and invalid reason as uh hooper did is you know definitely an invalid is an invalid adjustment i think okay so so i mean based on their own power calculations i've got here down at the right that this isn't this isn't something arbitrary this is their own calculation they require between you know 20 490 to 11 645 participants to obtain a five-year five-year significant you know difference um and they needed a follow-up of at least 3.6 years at the end of the trial they had 2000 or in this more recent analysis anyway they had 2 355 participants and they had 2.9 years average follow-up so do you like did they satisfy their own power calculations that's sort of relevant really right when you're doing an rct when you're doing an rct if you show harm as they did in this one and as they've done in the other two you don't get to discount it because you don't need a p-value no i'm not i'm not discounting it for that reason i want to be very clear i'm not this it's just it's one but it's a statistical test it's rather irrelevant to the outcomes yeah so but it's i think it's not really relevant because the results are actually compatible with there being a potentially a reduction in risk as well like it's still compatible with that being the case basically talent mortality tended to go up yeah sure so that's not a reduction of risk sure but it's still compatible it's still compelling you know like you you you talked about the dayton study the la veterans yup what was the final outcome of that study so can we talk about that after this one i'm more than happy to absolutely but can we just stay on topic for now we are on topic i've been letting you um sort of go along with this for quite some time um but i have i have what was the result you said that dayton showed benefit the la veteran study showed benefit what was the final result between the control group and the experimental group well depends on which outcome you're looking at for fatal atherosclerotic events there's a reduction and what was the total outcome so i i i see where you're going so i absolutely plan to address that but i don't want to get off of them let's get into it let's get into it now but we were just on we can't just jump around like that like let's systematically let's okay let's cover mce right now and then 100 the next thing we cover is la veterans absolutely totally fine with that but let's first go into this particular study as you said it's the best study you have i'm saying ellie veterans is the best one yeah sure but you i believe you said highest quality but regardless no okay okay if it's if it's the biggest anyway let's just go over that given that we're here and i will absolutely talk about la veterans right after okay is that fair sure okay so for one we talked about the power calculation didn't satisfy it we can argue around you know how important that is but if we look at the original publication here we see the the lion's share of the events actually occurred within those first couple years especially the first year as we see on this table on the left now if we can find the results to people who remained on the diet for at least two years we actually see a trend towards benefit in the highest risk group the older age group in the treatment group as far as acute heart attack sudden death silent uh heart attack i mean zero for for uh men uh six of uh uh in the treatment group zero to six for uh men in the control group for women we have one versus four um same with you know all deaths we see a trend now none of these are statistically significant but again if you're you're not concerned with that and if you think that the trends really matter when we when we specifically look at people who remained on the diet for at least two years and had good adherence and we look at the highest risk group we actually see benefit from the treatment right okay that's pulling them out together what are you doing here you're doing a post hoc analysis well first of all well this is the original total if the total finding well no i mean like look if you're taking people who were going to be dying or having a heart attack in the next few months and put them on an intervention nothing's necessarily going to save them in that case um this is this is from the this isn't like some analysis i've done myself this is from the original paper for the minnesota coronary survey which is where the the mce data came from right and they found a total harm we're not we're not talking about that we're talking about well that's right but that's the relative that's the relevant outcome to the debate on coronary heart disease absolutely and and even if we look at all those steps if you have an intervention if you have an intervention that you know like what happens in la veterans where you know they showed a bit of a benefit for coronary disease but more people died in the intervention group because they got cancer well personally i would rather die of a heart attack than die of cancer rather you know side effects are very significant in studies like this and you can't claim a subset of the data shows that the intervention is positive for instance because in hooper they show a tiny benefit for heart attack but no benefit at all for mortality well that's not good well no so but we're talking about the debate is specifically about heart disease and even if we look at all cause deaths here and we look at the higher age group again we see a trend towards benefit in the treatment group versus so even if i grant that that um that in this debate we should at all focus on all-cause death we still see a benefit in the treatment group compared to the control group do you agree with that in the higher age group for people who have actually been on the diet again you're doing a sub population analysis if we look at the total outcome we see harm so okay but we're not we're not talking about that here i'm talking about for those who were so do you agree that being if we were to see harm so if we were going to see harm from this intervention we would more likely see it with people who have been on the diet longer than those who who had it for wrong for a shorter period frankly it's irrelevant to me why is that so it's irrelevant so if i were to go and you're saying if you die in the short if you die in the short term but if you survive the long term it's a beneficial intervention no no who would take that who would take that pill well that's the argument that you're making oh no i'm saying so just yeah if i can just clarify you know answer that question um if we're talking i'd like to i'd like to share something for go ahead yeah go ahead can you see this yep yeah here we go so again the complete data mce does cross zero there are some people who get a benefit from it but ultimately you know they're showing harm the hazard ratio is over one but that's not what i was talking about i was asking i know but you're missing the point here right the point is not does the subgroup show benefit but that's so for starters we're talking about coronary heart disease which this is cardiovascular that you're showing i agree with that okay that that's fine but would you agree that if something was going to show harm you would likely see it in those like are these people immune to the problem or are so are you saying that people who remained on the diet longer somehow received benefit because they're immune to the harms in some way i mean they did better than the control well there's known genetic variation in how people handle uh omega-6 fats so it's plausible that the people who are you know just like there are smokers who die in fairly uh short order when they start smoking you know just because castro lived to 101 or whatever smoking cigars doesn't mean that cigars are a beneficial intervention so so you're generous here yeah just unclear so you're saying that those who remained on the diet longer were able to do so because they somehow conferred benefit from the linoleic acid based on some genetic predisposition because they did it it wasn't just they were less or they were less susceptible to the harm just like castro and his cigars but but no they did better than the controls it's not that they were less susceptible to the harm percentage-wise they did better than the controls who were also on the intervention for a length of time we have 14 percent of the women having uh in the control group 14 of the women having uh uh mi or sudden death 16 percent of the men or 17 of the men versus zero percent of the men in the treatment group uh who are on the diet for that period of time and uh and 33.8 you're still left with an intervention that overall showed harm no no you're not you're the controller that's what that's what the data shows that's what we're looking at here it's the top line that's not we're talking about like hazard ratio well over not well over one but over one but that's not indicates that the entire intervention had harm but that's not what i'm asking right you said that though i just want to make this really clear so you said that those who were on the diet for longer might have had some sort of genetic predisposition to prevent harm from linoleic acid is that much correct so it's a possible expert it's possible but it wasn't just that they didn't experience harm they did better they experienced benefit if anything that is what i'm trying to understand do you have an explanation for that no i don't but so what it's irrelevant well it's not irrelevance that you're saying because you don't know when you put somebody on an intervention you don't know whether if you put somebody on an inter if i tell you to start smoking tomorrow do we know if you're going to get lung cancer in 20 years or never get lung cancer and live to 100 and whatever no we have no idea so looking at a statistical outcome post hoc right i mean this is like the most rudimentary fallacy of uh study analysis when you're an individual looking at the intervention you know you decide not to smoke i presume because you know that on average it's harmful so you're saying that well you know what these people don't seem seem to be okay right the french have lower rates of heart disease from smoking than americans so it's probably okay to smoke okay so are you saying that the people in the in the treatment group here that were on it for a longer period of time were actually consuming less vegetable oil than the people in the control group or less linoleic acid than the people in the control group because they believed it to be harmful one way or another no the con i mean the data is quite clear that they were consuming more well yeah variation in outcomes variation in outcomes from an intervention is typical okay they all have variations that's why we have all these little error bars on here in confidence intervals not everybody has the same outcome yeah and but again like i i just don't understand what the explanation is here for them doing better like can you just yeah just please try to clarify that i think i've clarified it um now the problem that i have with some of the papers that you're looking at that or some of the arguments that you've made is that you're showing for instance you cite oz we agreed that we were going to discuss the effect of seed oil specifically linoleic acid right it's been demonstrated as these studies show that uh there are different effects of omega-6 fats oleic acid and short-chain and long-chain omega-3 fats right alpha linolenic acid and dha and epa why are you blending those two different types of studies to make this art to make your argument i mean what do you mean well if we go back to some of your slides you're saying oh look the oslo diet heart study showed benefit well the oslo diet heart study was not a study of linoleic acid okay the oslo diet heart study was a mixed intervention of linoleic acid and fats from fish yeah and so um so i'm understanding you you believe that the the epa dha come for benefit compared to the linoleic acid that's what the data seems to show okay okay i just want to be and and look i if that's the case like i agree that epa dha are also likely beneficial like i absolutely freaking okay okay yes for instance the sax 2017 american heart association presidential advisory did the same thing that you're doing they combined a bunch of studies that um showed benefit like oslo and their conclusion was something that was not supported by the study so you can't take the oslo diet heart study and conclude that based on that intervention that an intervention consisting solely of seed oils is beneficial it was it was polyunsaturated they were they were replacing saturated fat with polyunsaturated fat and epa dha are polyunsaturated fats that's what they were looking at but it's not linoleic acid yeah there is no dh there is no dha and epa in any seed sure and that's yeah i'm just saying that i don't think sax misrepresented that because they were specifically looking at a pufa for sfa replacement right which is accurate to what they did right now no actually they wrecked they turned around and they recommended increased consumption of omega-6 pufa based on a study that had a blended intervention that didn't support their recommendation can you show me that it's it's in the paper like they they've referred to rams then wrote an objection letter which i don't have in front of me and actually in sight so i'm not going to go into that okay and going back to ramsen for a sec two um the the last issue i just wanted to raise with mce before we move on to la veterans as if you still want to go there i think that was the agreement um was that there is the potential for confounding due to trans fats as well um if we look at the uh the predicted change in cholesterol versus the observed change the observed change was less than the predicted which could be explained by um inclusion of trans fats and uh and there were only three margins that were really widely available at the time and two of those three did contain uh trans fats as well so it's also a possible issue there which is one of the reasons nobody knows nobody knows the composition of those fats no that's what i'm saying you're misrepresenting it right no not additionally additionally we have a biomarker as ramsden has pointed out we have a biomarker that allows us to distinguish between the effects of trans fats and linoleic acid in these studies it's ldl trans fats cause ldl to go up linoleic acid causes it to go down all of these studies showed a decrease in cholesterol from linoleic acid consumption so if you're going to claim that the outcome was if the outcome was confounded by trans fat consumption which we can't test because we don't actually have any samples of any of the fats that were included at the time if it was such a significant effect we should see it in the ldl outcomes and we don't know no we don't we see a reduction which is the effect that we get from linoleic acid it cannot increase which is the effect that we get from trans fats can i share my screen i just want to show you the figure for their or the the table for their uh ldl reductions while you're doing that matt just to clarify i brought up the sax paper uh it does it does say at the end overall evidence supports the conclusion of polyunsaturated fat mainly from vegetable oils mainly omega-6 linoleic acid so that is in there somewhere yeah and i mean if they're saying if they're not yeah and if they're not saying specifically if they're saying mainly i think that is still compatible but that being said i i can see that yeah they did specify omega-6 bats i have no issue with that but um yeah do you mind if i just share my screen real quick uh here tucker oh yeah sorry yeah yeah no problem go ahead okay um so yeah so talking about trans fats um or in the the change in ldl i agree um that the ldl still went down or sorry total serum cholesterol still went down but it didn't go as it didn't decrease as much as we would expect based on keys equation and so that could be because trans fats were preventing it from dropping as low as it could have it's not the case that trans fat increase or trans fat consumption was so high that it completely blunted the ldl reducing effects but it could be the the case that trans fats were preventing the reduction that we otherwise would have seen um and you're speculating well sure and i and i i admit i am speculating i'm saying it could be a possibility i mean you know we could also observe that heat causes trans fat production in any cooking oil yeah but not too late all cooking oils when they're cooked start producing trans fats so any cooking oil intervention unless you use uncooked cooking oils which none of these studies did is going to have some amount of trans fats that's just fun that's a fundamental problem and do you think that that uh that like a oil with one percent trans fat would be just as bad as an oil with like 15 trans fat i don't think they had 15 percent trans fat and i haven't seen any i haven't seen any data that looks at trying to you know one of the problems with this whole field is that there is no data trying to decouple the effects of trans fats and linoleic acid well sure but but that's that's not really what i'm asking so if let's say it was 15 trans fat would you think that that is potentially worse than a one percent trans fat not necessarily not necessarily so you think that trans fats and linoleic acid are roughly equivalently battered the animal data that i've seen shows that linoleic acid is worse than trans fats and the fact that trans fats are saturated and not susceptible to oxidation makes them less harmful than linoleic acid so i would actually i would actually expect to see a better outcome from a study that had higher amounts of saturated and trans fats than completely linoleic acid okay so from the animal data so yeah so you suggest just to clarify that you think that replacing uh replacing linoleic acid if someone were to replace vegetable oils with something that is is lower linoleic acid but high in trans fats would probably be a net benefit possibly okay i've never seen a study that's done that but it's you know based on the animal data it would be consistent okay um and now i'm more than happy to jump into uh la veterans if you want um i know you mentioned a couple things about like all-cause mortality or or uh uh cancer i believe you mentioned cancer outcomes something along those lines right right yeah um so actually do you mind explaining a bit about that um just for those listening so that they can i can give you the screen as well if you want it and then we can jump into that so the death student on atherosclerotic causes numbered 71 in the control group and 85 in the experimental group right most of the difference occurred in the latter part of the study consideration of causes of death in this category suggests that this difference probably did not they say reflect the toxic effect of the experimental diet although as they noticed noted in other papers there is data suggesting that increased consumption of pufa can contribute to uh cancer outcomes you know so so cancer is specific okay yeah let's let's talk about the but that's starts getting very off topic well well no um i think it's it's relevant well if if you're suggesting that that uh you know as i mentioned like fatal atherosclerotic events and other cardiac outcomes they did better especially those with higher cholesterol baseline are you suggesting that that's because they were dying of other causes well i'm suggesting that any intervention that results in a higher all-cause mortality is not an intervention i would like to take regardless of what i die of wait where where is the all-cause death finding that you're speaking about can i ask one of you just to share your screen perhaps with this study oh yeah yeah um target do you do you want to as you can point out to me what uh yeah no i'm sorry dayton dayton did show a bit of a benefit yeah that's what i thought i just wanted i was really confused okay um so so yeah then then what would your objective dayton again was linoleic acid plus uh you know a blended intervention the blended interventions beg your pardon blended in what way like what do you mean by blended because like as far as i'm aware is just pretty much exclusively vegetable oils being infused into these various foods a blended intervention of linoleic acid and of omega-6 and omega-3 fats yeah and but omega-3s from where from uh vegetable oils yeah so the debates on vegetable oils i think the fact that they contain some ala like soybean oil for example doesn't mean that that it's it's all of a sudden irrelevant to what we're talking about unless you hold the position that soybean oil is healthy on the net uh i don't hold that position yeah so it shouldn't matter then if there's some ala in there right well the problem we have with ala and with omega-3 is that they both seem to show benefit but they show more benefit in a context of a lower linoleic acid diet which gets us back to the one diet we have here that showed a very significant improvement which was the um here let me yeah leon diehard i know you're talking about that um but right again just just to clarify though i i want to understand your position on healthy veterans right i want to clarify your position on la veterans first before we move on to leon diet um just because so if they showed a benefit with oils like soybean and other similar oils that yeah they contain some ala but they're high in linoleic acid would you agree that that is overall beneficial that's what the studies show yeah so you agree that the vegetable oils used in la veterans were beneficial then for for the you know the outcomes of interest uh slightly beneficial yes okay cool um and you know by slightly like can you quantify that i'm just wondering what you mean because the way the way i like you see my screen here's the here's the quantification right so we see again three high linoleic acid studies showing harm as i said in my introduction three blended studies showing slight benefit one of them didn't show any benefit and then we have two studies showing a uh more benefit from a higher intake of long-chain omega-3 fatty acids this is what the evidence shows us yes but with with la veterans i'm confused as to what outcome specifically that was because if we look at um i mean i can always share my screen too but if we look at say a aggregate of ischemic heart disease acute mi and acute cerebral infarction we have a what is it a you know 36 reduction in risk and statistically significant for fatal atherosclerotic cardiovascular disease uh 0.67 so yeah again 33 reduction in risk statistically significant for all-cause mortality uh again 35 reduction risk all um statistically significant these are substantial um so i'm just wondering like it's all what outcome was chosen for this figure here what outcome was chosen for this figure yeah it doesn't make sense based on the data that i have here so what's your point here matt well no so my point is like like i don't know what outcome that is but when i'm looking at um you know ischemic heart disease acute mi and cerebral infarction um as a combined endpoint a substantial reduction of risk fatal atherosclerotic cardiovascular disease substantial reduction risk all-cause mortality substantial reduction risk um to say that there's just a you know slight benefit non-statistically significant benefit based on this figure i'm just wondering what is that looking at because it's not actually concordant with what i'm seeing here and i can share my screen and go over this stuff too no problem from the from the la veterans paper itself so i'm looking at uh ramsden's meta-analysis from 2016 which is room you're familiar with yeah yeah yeah i think you sent that one uh yesterday but i'm just wondering what like i understand where it's from i recognize it i'm wondering what outcome is specifically on that figure there from the la veterans paper so the baseline and control diets in the la veteran study were rich in hydrogenated oils so we can assume had a good amount of trans fats no i nope i would absolutely not agree with that um and i'll explain why here in a second i'm curious to hear that because yeah pretty much by definition trans fats are produced in the process of hydrogenating oils yeah so i'll just share my screen i'll show you what i'm talking about um so this was from the usda uh data i sent oh it's loading give me a moment uh there you go um is it just in table form whoops uh table form essentially top left here now this is a this is kind of a extreme example of a 34 trans fat you know if we use like miracle margarine from back in the 60s it'd be like 15 or so but uh if you look at no hydrogenation versus fully hydrogenated um yeah it's like a percent it's not a huge amount by any means and even if we were to cut this uh the partially hydrogenated in half it's it's like or is a magnitude higher um as far as trans fat content if we look at the top right that is more of your standard uh in the red there more of the standard uh partially hydrogenated soybean oil around i think 14 15 trans fats as we see here um versus again the the really super high one but um but yeah we don't actually see a huge amount of trans fats in fully hydrogenated oils or shortenings but i do want to go back to although if you have a fully hydrogenated source a study like this you're not doing a linoleic acid trial well no no no that's not in the intervention that's in the um that's in the control the intervention group had literally had vegetable oils basically infused into foods to mimic um to mimic the standard fare and actually actually what's really cool about this is they did a pilot study first like no trials do this nowadays but they did a pilot study where they trialled these food replacements to see that if you know are they um uh do they taste good uh can people recognize them um like just making sure that they really did mimic and and that they were palatable to the people consuming them so they went above and beyond and it was specifically vegetable oils that were infused and uh whoops that's the earlier i have uh i have these later in my notes too um i want to pull it up right here um and if we look at um like i mean if we look at total where is it uh total you know all-cause mortality we see a benefit there and this is simply taken from the paper itself just put into a forest plot so it's easy to read um and uh for you know scheme of heart disease cute mi um yeah not statistically significant but uh if we add cerebral infarction so stroke to that substantial benefit 0.64 for fatal atherosclerotic cardiovascular disease um 0.33 for all cause mortality 0.65 oh sorry i meant 0.67 meaning at 33 reduction and 0.65 to 35 reduction so we do see a pretty consistent reduction risk of these cardiac events across the board right so dayton says we did not analyze our diets or tissue lipids for trans fatty acids uh presumably both diets contain some trans fatty acids from partially hydrogenated fats but such fats were only minor components of both diets yeah so it wasn't the choleste it wasn't the controlled diet well so okay so both i mean i know that in the controlled diet therefore as they say the cholesterol elevating and triglyceride elevating effects of uns unnatural fatty acids cannot have been an important important factor in this trial yeah i would agree with that cannot or can be cannot cannot yeah so i agree with that and i mean we have we have they measured adipose uh linoleic acid concentrations a pretty marked increase like three four times increase over the course of the trial there um a substantial drop in serum cholesterol levels body weight stayed about the same um i mean i again the reason i think that this is probably the best that we have but according to these results would you not take the position that for cardiovascular outcomes it was absolutely beneficial to have vegetable oils in the place of saturated fats it showed a small benefit a very small benefit a 35 35 reduction in risk of all-cause mortality and a 33 reduction in risk of fatal atherosclerotic cardiovascular disease is a small benefit well it also showed a large increase in cancer in the engine towards the end of it right oh that that's i mean you don't want to just ignore that okay so so i will get to that in a sec but do you agree for the cardiovascular outcomes there is benefit and substantial benefit at that i don't agree that you can just slice and dice a study like that without okay so some consideration over the right so let's go to cancer then if we look at the adherence level cancer outcomes stratified by levels of adherence to the diet the lion's share of the cancer death in the experimental group were in people who did not adhere 0 to 10 adherence they write this in the paper many of the cancer deaths in the experimental group were among those who did not adhere closely to the diet this reduces the possibility that the feeding of polyunsaturated oils was responsible for the excess carcinoma mortality observed in the experimental group however there was just the possibility yeah no i mean i couldn't rule it out i'm going to continue however there were significantly more low adherence in the entire experimental group than in the controls in both groups the number of cancer deaths amongst the various adherent strata are compatible with random distribution a high incidence of high adheres would be expected if some constituent of the experimental diet were contributing to cancer fatality so if if the the intervention was causing cancer outcomes we would see a higher rate in those who are actually following the diet and they measured diet adherence with tissue linoleic acid concentrations so we know whether or not they were adherent we can't say that the intervention caused the cancer outcomes here and it's very very very unlikely that that's the case so going with that being said well i don't know yeah i don't agree that it's you we can say that with the certainty that you're saying well i'm not saying they had enough uncertainty about it so that they wrote several subsequent papers looking at the effect of cancer on this intervention well and and we see cancer in those who didn't follow the diet i mean it's as simple as that we there is now i'm not saying it's impossible that there was a slight maybe maybe it's possible there's a slight increase but guess what it's possible that there was a slight decrease too based on on the higher adherence numbers here we don't know we can speculate that away but at the end of the day if i go back even for all-cause mortality despite having higher cancer deaths we still see a reduction in all-cause mortality and a substantial one at that 35 yes a slight increase which suggests that there is a benefit to omega-3 fat supplementation no they were predominantly linoleic acid and you know what that aside whether it's whether it's that's a crucial distinction it's it's not because it's not because in the american diet what we see is a massive intake of linoleic acid and a tiny intake i mean we're looking at a ratio of omega-6 to omega-3 of like 20 to 1 which is way outside any sort of evolutionary norm right okay so are you talking about uh can i interject here guys yeah just because this has come up uh quite a few times and i think it would be important to clarify here uh because it from from my end it seems like what tuck is saying is some of these trials are confounded by the inclusion of dha epa or ala i think it would be important to potentially establish if tucker has a view on how much of that would confound um and then perhaps that will help clarify some of this sure as in how much of that those fats would need to be in a diet that's an excellent question and unfortunately the answer there is not clear because the benefit of those fats seems to depend on the intake of linoleic acid which in the case of alpha-linolenic acid the short-chain omega-3 fatty acid the benefit seems to come from the long chain fatty acids as the chart that i showed indicates right so the benefit of short chain fatty acids depends on conversion to the long chain fatty acids and that is blocked in humans by high intake of linoleic acid so a given rate of alpha-linolenic at a given rate of the short-chain omega-3 fatty acids may be beneficial in some circumstances and less beneficial in other circumstances depending on how much conversion is done to the fats that the body actually needs okay um additionally long-chain fatty acids you know reduction of omega-6 fats in the diet increases the lipid pool of long-chain fatty acids through a mechanism that's not yet been determined so a lot of it depends you know getting outside the scope of this conversation other studies have shown you know in fact i think i included one of those um have shown that just lowering linoleic acid increases omega-3 fats and serum and they're i've never seen an explanation of why that happened so it depends a lot on the different components of the diet and yeah and and just to you know backtrack on the or to where we were before that too um the one thing that i think we need to clarify here is is i agree that you know linoleic acid in particular is a proxy for vegetable oil intake and it's a large component of vegetable oils we have to remember that this debate is centered on the vegetable oils or seed oils as we call them itself and so if this intervention la veterans is using things like soybean oil as a really major you know component uh there's safflower soybean cottonseed they're all listed on the left side here by my arrow a corn as well are you suggesting that because this intervention also increased ala by increasing consumption of these oils that those oils are healthy um as this study suggests i wouldn't suggest that any of them are healthy i mean we have to remember we're looking at a background of a population that has what seemed to be extremely high rates of cardiovascular disease based on historical and geographic distribution of the disease right so saying that a you know a small a relatively small benefit i mean what i want to know the question i want to know is why were those africans in lee 1964 showing a near zero rate of myocardial infarction and why were african americans showing 12 myocardial infarction right so if you're going to if you if you're going to tell me that going from 12 to 10 shows that these are healthy i'm going to disagree if you're going to show tell me that wait if you're going to tell me that going from 24 to 10 shows that they're healthy i'm going to disagree because i don't know what you're talking about with those percents can you just sorry clarify for me i'm just not following i'm going to thought yeah i'll go back to this um this paper right so these numbers are evidence of previous heart attack at autopsy right for the audience a heart attack leaves permanent damage in the heart and you can tell at autopsy whether or not a person has ever had a heart attack a myocardial infarction right so in and again this is an old paper in white americans 24 of the people had evidence of a heart attack in africans in africa 0.1 had evidence of a heart attack at autopsy right one out of 4 500 people had evidence what i want to know is what's that difference right why did african americans why did africans go from zero in africa to 13 in this study and why are why did cardiovascular disease go up so much in the united states and england as seed oils were introduced into the population right so you're looking at a background rate of high disease right populations that have low intakes of linoleic acid uniformly have low rates of cardiovascular disease uniformly i'm not aware of a single exception okay and and so with with like la veterans then would you say that in for the standard you know america person on a standard american diet if they were to and not that that's even necessarily i guess it's close to what they're having there but if if you can agree that a standard american diet is absolutely 100 i know you guys have very different attitudes about diet than me but i think we can all come together on that i think here we agree on that um but what i'm wondering is so in in a case where people are at high say baseline rate of cardiovascular disease would you take the position based on these results i've shared that if people were to replace saturated fat with vegetable oils they would actually benefit because that's what this show in general it's what did hooper 2020 say there's study that you started off with little or no benefit that that was not well that was a total pufa and it was it was combined amongst a bunch of trials i think this is the best trial so i would expect a little or no benefit but but i'm talking about specifically the la veterans trial here as being the one that's best designed to answer this question and even if we look at that it's best designed to answer this question why double binding you know all three of all three of those studies that showed harm from a high linoleic acid intake were replacing saturated fats with pufas and trans fats uh likely but there are other issues that we've already gone over now i mean i i think i've i've kind of made my point with la veteran so let's talk about this this data that you just presented here anyway uh you said that was lee right yeah i think right here um um yeah so if i mean do you have the paper there um if not i can share one of the tables i wanted to just mention really quick go ahead okay so yeah i think it's i think these results as far as low mi um i absolutely i want to say first off i absolutely agree that environment is a huge factor i think that we absolutely agree on um you know what factors amongst those environmental factors i think that's where our discriminant is but if we look at like for example uganda here look at the number of autopsies in the 20 to 29 30 to 39 40 49 group versus those in the upper end here we see that you know most of these people are dying at a very very young age they wouldn't even make it to the ages where we'd expect to see heart attack and at the same time that really incites potential for survivorship bias and so for those listening um who might not know what that is basically those who survive through especially if you look at say hunter gatherer populations and everything too those who survive through childhood uh you know don't fall victim to things like infant mortality or infectious disease they are likely the fittest of the population they're they're the ones who are less likely to develop a lot of these issues moving forward so you know oftentimes i see people argue that we should look at specifically um uh you know data that is adjusted for infant mortality so post infant mortality but i think that actually skews the data in a more favorable direction for that population um whereas if we look at like new orleans here we see very low death rates in the younger population and and more in the or at least total numbers here in the lower population and more in the higher population and yeah they're more susceptible to to am i of course that is not just the only issue there there are many other environmental factors that are at play whether that be dietary or otherwise i absolutely agree but i don't think that looking at these ecological associations like this that don't adjust for any confounding variables whatsoever really tell us that seed oils are all of a sudden the problem well i'm not claiming that this tells us anything about okay okay okay i just want to make sure yeah i'm claiming i mean it indicates that that that is a factor that was different at that time between specifically japan and the united states sure right i could agree with that yeah the point of this the point of this and the previous two studies that i showed in this was to indicate that cardiovascular disease is not an expected human disease in the prevalence that it exists nowadays okay yeah no i i think we fully agree on that anyway right i don't i don't think that's that's any point of uh contention now um you mentioned leon diet heart a bunch do you want to maybe talk about that at this point you said that's probably the strongest one that you got yeah you you talked about leon die at heart a lot the the trial well um you know yeah let's talk about that yeah go ahead uh you can i mean it was it was your reference go ahead uh you can talk about the main uh so leon diet heart is the one is the strongest heart disease prevention trial of any intervention that i'm aware of right and they unfortunately did it gets excluded from a lot of these analyses because they did a whole bunch of things they told them to eat less saturated fat but interestingly it's the only intervention done where they specifically decreased the linoleic acid intake in the diet and they specifically went against the ancel keys hypothesis that an increased polyunsaturated to saturated fat ratio would be um beneficial that study increased it right and they saw a massive decrease now this was a secondary prevention trial so they're looking at people who've already had a heart attack who are obviously most likely to have another heart attack and they saw i think it's a 70 reduction in mortality okay right yeah yeah so if if one's claiming that linoleic acid an increase of linoleic acid is going to be beneficial for cardiovascular disease one has to explain a why did cardiovascular disease in the united states go up as consumption of it went up in the early part of the 20th century but also why does the one trial that reduces it show the biggest benefit of any of these trials okay yeah and so here i'll talk about leon a little bit so i think you've already mentioned one thing is there were a lot of moving parts right i think we agreed there yeah it wasn't just one thing right yeah so uh just so that those those listening or watching can can get it basically i mean uh there's a difference in bread intake increased uh legumes went up fruits increased meat decreased uh butter and cream decreased margarine went up so we we see there's definitely multiple yeah go ahead yeah i'll also note that in if one looks at all the different foods that changed in most of these inner interventions there's typically a lot of things going on and this particularly the control which was a really poor control um they were apparently consuming some version of the um aha prudent cardiovascular diet although the aha is disputed exactly what version of it they were eating so which is also a low-fat low saturated fat diet pattern okay um and yeah so uh they also provide their nutrient intake data now i made this table uh just to make it a little prettier so people can follow along um and so just want to talk about a few of the moving parts here for one we see that in the control group saturated fat intake was 11.7 percent of calories and in the intervention 8.3 percent of calories um that in it in and of itself as we see on the top left here from the hooper 2020 paper crosses that threshold where we expect to see benefit or or a harm if you're increasing uh intake you know we go from about eight percent of calories up to you know 10 plus percent of calories i mean it's a striking increase in risk of cardiovascular disease so moving the opposite direction it's not super surprising that we see a benefit and this is based on the randomized trials as well um similarly as you mentioned with linoleic acid in the pink here or the dark or light purple whatever you want to call it uh you know going from 12.6 grams and the controlled 7.7 grams in the intervention about a 5 gram 4.9 gram difference so yeah they did reduce linoleic acid absolutely i mean you look at la veterans they increase from 10 grams up to like something crazy like 30 some odd grams um or 40 grams so it's a striking increase yet they saw benefit so i'm not sure how this you know unless that little bit of increase of ala somehow countered all of that and led to benefit uh but we see well i didn't see a benefit like this study saw a benefit yeah sure sure um but they they did see a benefit with a massive increase in linoleic acid intake so um which is confirmed with biomarker measures too uh if we look at ala this is an interesting one you go from 0.6 um grams up to 1.8 grams uh now if we look at this dose response analysis we see that going from about 0.6 grams to about 1.8 grams will lead to about a 25 to 30 percent reduction in risk of coronary heart disease mortality so this isn't events this is actually mortality um and so it is entirely possible that based on those findings that the benefits are actually driven by a reduction in saturated fat and an increase in ala despite reducing linoleic acid there's no way for us to yeah go ahead no continue yeah i'm just going to say it's absolutely possible that the benefits are driven by changes in saturated fat in ala but not linoleic acid and actually for all we know the benefits could have been even greater had linoleic acid increase uh and if we look at the relative changes except there are lots there are lots of studies that reduced saturated fat and increased linoleic acid and none of them showed this sort of benefit so it's unlikely you mean like mce sydney diehard rose yes yeah and so we've already talked about mce i don't know if we need to get into that more but we can talk about the other two as well um but uh but if we look at the relative changes as well the the relative change in saturated fat and the yellow here 1.56 in the linoleic acid 1.63 very similar and the relative change in ala is 2.8 i mean the the changes is there's no way to isolate these findings to linoleic acid or even to um to uh isolate it to linoleic acid and ala while ignoring the saturated fat binding i just don't see how that can really be done well i mean delorea girl's explicit rationale for this study was to adjust the ps ratio yeah right which key you know so we have a hypothesis we have the hypothesis that the increased ps ratio which came from ansel keys and was the justification for the aha recommending a decrease in saturated fat and an increase in polyunsaturated fat in the 1961 aha dietary guideline uh paper would decrease um cardiovascular disease rates and there were two major tests of that hypothesis ansel keys was one of the principal investigators on the mce so that was a direct test of his own hypothesis and we saw an increased harm and then delore gorill took that ratio in the other direction and showed a massive benefit so we have a we do have a test of a hypothesis here that's not yeah good you know it's this test of this switch from polyunsaturates from saturates to polyunsaturates and we have lots of papers you know we can go through hooper 2018 and hooper 2020 again and you know what do they find little to no effect maybe a benefit for myocardial infarction only but small benefits for heart disease for all cause mortality or heart disease okay and then we have this outlier which cooper excludes because it didn't modify dietary fat i think was the language that they used despite as you just noted they did modify diet diet dietary fat yeah so i mean i i don't know if that's if that's what they said then then fine um but yeah i would agree that they did modify dietary fact i i just fat i just don't think we can uh you know claim that it is at all due to the reduction in uh linoleic acid that we see these results why not because we see multiple we do see a reduction of saturated fat as per hooper we see um the increase in ala as i've suggested would show a benefit across and i think you agree that ala and that increases so did la veterans reduce saturated fat and increa increase ala yeah but it also increased linoleic acid so if linoleic acid was so if this benefit was due to and which one showed a bigger which one which studies showed a better result this one did but they're also different uh different study designs altogether this one is looking or sorry they have different participants all together this one is secondary prevention you're looking at higher risk patients um la veterans was primary prevention completely different sort of event rates that we're going to see um a completely different background and you're looking you look you're looking at a tougher pain patient population here these are the people who are most likely to have a bad outcome and they're also most likely to benefit from changing you true right yeah we have lots of secondary prevention studies none of them showed the results of this one sure but we also have um we have like a whole bunch of secondary analyses from uh la veterans too like we have uh we have if you look at specifically those who started at a younger age with higher baseline cholesterol levels sure it's not quite on the level of of beyond diet heart but we see a pretty substantial benefit there same with high age and high cholesterol um if you if you isolate it to those who are most likely to benefit that's where we see a bigger benefit um and so and and these patients as i mentioned were lower risk to begin with um so i don't think you can just compare the two i i think i think it's comparing apples to oranges really and then why is it comparing apples to oranges they're both fat modification studies yeah but one is looking at people who've had a heart or like recently had a heart attack and one is looking at people who never have those are completely different sets of participants and that's aside from the fact that leon diehard changed other things increasing like uh grain intake increasing um uh fruit intake uh legume intake there's all sorts of right and we have you know we have other studies that have done all those interventions and none of them have shown this sort of benefit okay yeah and i i don't i don't dispute that whatsoever i don't dispute the fact that none that none of the studies looking at very specific changes that occurred in leon diet heart study have shown the uh benefit that is seen in that study that used all of those interventions together i i absolutely agree with that um i don't like i i don't really know why that why we can't just you know agree that they change so many different things and they're i agree that there is no study that has gone about uh that has seen a similar reduction in risk by changing one or two individual factors that leon diet diet heart did change right so then we get back so then we have to get back to the mechanisms of heart disease right no yes yes can we just intervention can we just clarify just for the listeners on that leon diet heart study tucker is saying that there was a substantial decrease in linoleic acid omega-6 in the the group that saw benefit and matt your position is that that is confounded by other changes that resulted in a drop in saturated fat and an increase in uh omega-3s is that the current position where you both are yeah i would add that there were also other changes when i'm talking about the saturated fat and and the omega-3s i'm specifically talking about the changes with fat intake um because there were other things like as i mentioned uh bread there was a lower um intake of beat there was a higher intake of legumes and like all these other changes that also took place just to clarify right okay so are those plausibly going to affect heart disease risk based on what we know about the mechanisms yeah so the demonstrated the demonstrated mechanism for cardiovascular disease is a increase in oxidized ldl which has toxic effects throughout the body cytotoxic effects um i can if i can share this and we have multiple studies going back several day you know i mean thousands of studies going back several decades looking at that pathway right so so to evaluate what mechanism in a study like that is going to be effective one's got to look at the mechanisms right so at this point in 2022 as far as i'm aware the only explanation on the table for what causes cardiovascular disease to initiate and progress is what they call the oxidative stress model or the oxidation model where ldl becomes oxidized and [Music] initiates atherosclerosis and causes every step of it to progress right that's why this harvard olive oil paper that just came out that i think you included matt and i may have also i don't remember they are claiming the benefit of olive oil is because of it being less susceptible to oxidation and this takes us all the way back to 1989 and steinberg and wisdom's paper showing that what happens to ldl to cause it to become toxic is the oxidation of the omega-6 fats in it so if we look at leon dye and heart in light of that then the mech the effect is quite obvious right what's driving this is oxidized omega-6 fats exclusively because those are the bulk ninety percent of the polyunsaturated fats in ldl are omega-6 fats and i mean literally the definition of oxidized ldl is an ldl partial particle with oxidized omega-6 fats that's the test that's used for them right there are actually several different tests and they all look at that factor so to simply say oh it could be any of these things is to ignore the demonstrated mechanisms of how this progresses right and to argue that oh you know you know alpha linolenic acid we see in these models is much less atherogenic because it doesn't turn into the same toxins that omega-6 fats turn into so it's perfectly consistent that the three linoleic acid only studies that we see would cause the worst damage the one where we see it reduced causes the most beneficial effect and that fats that have a tendency to block the effects of omega-6 fat in the body like ala and dha and epa would all show more benefit than an la only intervention taking us back to um right taking us back to another paper that we both agree is important born at all the european atherosclerosis society their step one in the initiation of uh atherosclerotic cardiovascular disease is the oxidation of lipids right that's what kicks off the whole process so we can go back and forth on all of these antique studies but the simple fact is there's no other explanation that's on offer in the scientific literature for cardiovascular disease initiation and progression there's nothing else okay uh is it okay for me to comment yeah do you mind if i can share my screen too and just while you're doing that um just just uh curious where does that oxidation occur like in in the you know there are a couple of different hypotheses about where it occurs if you have one that you i go with the postprandial hypothesis where it starts in the gut and it is oxidized as it goes through the lymphatic system and into the blood okay okay so um and now you mentioned steinberg and then what was the second author wisdom wisdom okay so i just want to go over those really quick here um okay so this is from steinberg yeah steinberg 1989. can you see the screen by the way yeah okay um so they they suggest that you know these newer findings about the properties of oxidized ldl together with the information available from experimental pathology and cell biology of the macrophage make it possible to propose a hypothesis about the development of the fatty streak lesion that is based solely on the presence of elevated plasma lvl levels plus the oxidative modification of ldl within the artery wall i agree with that but the oxidation occurs within the artery wall according to this paper uh you have native ldl and that was no no no no that was a hypothesis okay that was not a fact as you presented it okay yeah according to their their hypothesis anyway or one of their hypotheses um and so they kind of describe and we'll get to another figure in a minute here but they kind of describe uh how native ldl you know enters the sub endothelial space becomes oxidatively modified engulfed by macrophages becomes a foam cell eventually goes on to form a fatty streak and so on um they also suggest that for several reasons it seems that the oxidation of ldl probably occurs not in the circulation but within the artery wall and first even if oxidized ldl were generated in the plasma it would be swept up within minutes by the liver second oxidation is inhibited by plasma and so probably requires a favorable conditions of a sequestered microenvironment now they go through an analogy i'm not going to bother with that but at the end they say the ldl molecule in the southern entire space may find itself from time to time trapped in a space between cells where antioxidant levels are low so it seems that once you know you have elevated plasma ldl you have more ldl moving into the seventh hill space and then we get retention there that leads to oxidation and i'm going to cover you know why that happens so going to wisdom which you also cited um you know they do mention uh increasing plasma cholesterol associated with higher risk of atherosclerosis and so on um they also mention that we have reason that significant degrees of oxidative modification of ldl do not take place in the plasma because of its high antioxidant content again similar to what i just mentioned but this is the real you know uh kind of a big quote on the left side here thus we now can understand how hyper lipoproteinemia could of itself be a sufficient basis for initiating the atherosclerotic process in all its manifestations so this is from the paper that you're citing and they do go on to suggest as far as the aux ldl kind of hypothesis that ultimately a clinical test of this hypothesis in man will be needed to demonstrate that inhibition of oxidation of ldl will inhibit the atherogenic process so just going back they themselves state from the reference that you've cited here we now can understand how hyper lipoprotein could of itself be a sufficient basis for initiating the atherosclerotic process in all its manifestations which is the genetic hypothesis for cardiovascular disease which lead these disproved way back in 1964. that paper was later on uh this this one it was 91. so were they just not only lee 1964 the geographic progression of cardiovascular dis disease ischemic cardiovascular disease if you're claiming that ldl per se causes cardiovascular disease then what you are claiming is that cardiovascular disease is a genetic disease because we all have ldl based on our genes and if that is the case you would not have wildly variant rates of cardiovascular disease in genetically similar populations in different environments okay so we knew back in the 1960s that the ldl causation hypothesis was false right what these guys added was showing specifically what in what you know not only these gentlemen but also brown and goldstein showed was specifically what element was required to turn a non-atherogenic ldl particle into an atherogenic ldl product particle it had to be modified it had to be modified by having the omega-6 fats in it oxidized now they went on beyond that to say okay if you have more ldl and it gets oxidized you know that that may be an additional factor on top of the oxidation but they're quite clear that this is a required step right and brown and goldstein's experiment showed that the initial step of atherosclerosis the conversion of a macrophage or what they posited as the initial step of atherosclerosis the conversion of a macrophage into a foam cell doesn't happen with native ldl um sure they they see they find that it doesn't engulf the the difference is that and i don't understand exactly what you're saying as far as it being a genetic disease are you saying that that genetics are what determine ldl concentration or can you just clarify what you mean when you say of course genetic ldl apob is a product of the apob gene right if apobi is april beef again for the audience i presume everybody knows this is the protein that carries fats and cholesterol around the body it goes through a couple of different stages and one of those stages is ldl um apob can become oxidized if it has omega-6 fats in it and you know but apob per se is not atherogenic right that's what they're that's what brown and goldstein's experiment demonstrated so why doesn't macrophages didn't engulf native ldl that's why that's right okay so i'll explain why the and actually the other thing i just want to clarify that uh based on the the studies cited uh in those previous couple that you that we just discussed um they they pretty clearly say how oxidation doesn't happen in the plasma at least not to any significant degree do you take issue with irrelevant where it happens the fact that they just said previously i mean look look look look we know that it happens in the plasma because it's detected in the plasma right when you go get an ox ldl test they are not sticking a needle in the endothelium in the uh inside of your artery they are taking a blood test so we know there is oxidized ldl we have interventions that can adjust the rate of ox ldl in your serum so yes i agree this paper from 1991 has a hypothesis about what is happening but all these decades later when you know you can go to your doctor and get an oxen ldl test we know that it's in the plasma there's no question that i mean that's not even i agree i agree that ox ldl is in the plasma but why does that mean that it was oxidized in the plasma why couldn't it have been oxidized in the southern elder space in the close uh space and then spilled out into the plasma why is that relevant no i'm just saying because if you're if you're claiming that it's oxidized and it's because i'm going to talk about um the the kind of mechanism of action here and and if you're saying that it's um oxidized in the plasma specifically i don't think you've actually suggested or shown data to support that why is that relevant it's relevant because okay i'll explain what my kind of position on this is and then we can talk about why it may be relevant okay so this is that boring paper that you cited now you cited figure three i believe or figure four this is the first figure um or actually second figure after the the chart that they give um this shows that as you have elevated ldl concentration which i know um you mentioned uh genetics i mean i think we both agree that you can modify ldl concentration by diet as well right i don't think we have any right agreement there yeah so as you have a higher ldl concentration it uh enters the seven ethereal space by transytosis um or you know and once it's there there's proteoglycan binding that happens so april b can bind to proteoglycans and that's where it aggregates once it's held there and retained in sub-individual space oxidation is almost inevitable it's just a matter of how long it takes for that to happen well we know it's not inevitable so you you have you have data showing that you can completely 100 prevent the oxidation of ldl once it's retained in the southern endothelial space well we have data that i mean multiple human studies that show that the composition of fats in ldl determines how susceptible it is to oxidation sure sure but that that's not what i'm asking i i'm saying you're saying well you're saying yes but that is what you're asking that is what you're asking because if you have ldl that has let's say an evolutionarily appropriate level of omega-6 fats that it is going to be much less susceptible to oxidation than something that is very high in omega-6 fats and it's been demonstrated repeatedly that that is controlled by dietary intake of omega-6 fats the susceptibility of ldl to oxidation well no so just uh if you don't mind um i'm going to go through a couple slides really quick than happy to hear your response i just want to be able to get off a couple points here um really quick that okay fire away yeah okay so basically my my uh approach this would be you have the elevated concentration enters uh there's retention of the ldl um oxidation whether it happens sooner or later to to what degree sure we can we can debate argue that but uh seems like pretty well inevitable and and uh we might have disagreements there um and this is where they talk about oxidative susceptibility right all ldl particles their atherogenicity to variable degrees which can be influenced by the proteome lipodome uh proteoglycan binding uh and you know eventually oxidize oxidative susceptibility now if we look at the specific reference for that they actually don't hold this view for omega-6 fats even though it may be shown to be uh to increase susceptibility to in vitro oxidation um they seem to ultimately protect against coronary heart disease now we've talked about the trials obviously there's some disagreements there and that's that's fine i just wanted to point that out but i think the real key point to this comes from this claim on the left side genetic alteration of either the proteoglycan binding domain of april b or april b binding domain of arterial wall proteoglycans greatly reduces atherogenesis so if you affect the ability of the ldl particle to bind to proteoglycans you can greatly reduce atherogenesis right there and this is the study that they're citing so what they did was they took mice um with uh different types of uh of ldl so some were defective in ldl receptor binding some had defects in proteoglycan binding some were normal some had defects in both and they actually found that those that had normal ldl receptor binding but defects in proteoglycan binding had significantly less development um and a much lower trajectory as far as atherosclerosis development so they they made it pretty clear that after atherosclerosis is initiated by sub endothelial retention of atherogenic lipoproteins but they actually went further they verified that the differences in atherogenicity were due to the affinities for proteoglycans and not due to oxidation so they actually took the different types of ldl um the whether it be the uh the defective ldl or the control ldl um they performed copper stimulated oxidation and there were just no differences in the ability to oxidize so it seems that the atherosclerosis development here was driven by proteobycan binding not by um oxidation that's not their first when they go through the steps of atherosclerosis the first step that they list is the oxidation of the fats in the ldl no they don't it's right here the ldl particles influx no the the oxidation happens once it's in the sub endothelial space it's laid out right here the figure that you're taking is looking at i believe extracellular and intracellular space that's already in here um that's not in the plasma right so you have high concentration of ldl in the plasma it uh transcytosis enters the sub-endothelial space is retained and then um it goes on to be oxidized which then initiates the um you know the macrophage engulfing it becoming a foam cell and so on matt can you can you summarize your point and the relevance and then give her a chance to to rebut here i'm just conscious you've had the mic for a bit yeah yeah totally so basically what i'm saying is that if you know tucker's saying that oxidation of ldl is is the initiating factor and what i'm i don't disagree that it's involved in the process my point is that elevated ldl cholesterol levels or more specifically apob as we both mentioned leads to um ldl particles or b ending up in the sub-endothelial space being retained in that sub-endothelial space where oxidation is pretty much guaranteed to happen and this study demonstrates that if you can reduce that proteoglycan binding if you can reduce the ability of the ldl particles to be retained in the sub-endothelial space then you can greatly reduce the development of atherosclerosis aside from any changes in oxidation that there were no differences in oxidative uh capacity of the ldl here so i i just i think it's pretty clear and one way to reduce the amount of able be in the sub-endothelial space is to reduce ldl or apob concentration which is something that can be done by eating polyunsaturated fats so that would be kind of a quick summary i guess of my my points right i think you're misunderstanding the process of oxidation right it doesn't have it is a requirement that you have omega-6 fats for the ldl to become oxidized and to be have the toxic effects that it's been shown repeatedly to have it doesn't happen right it chemically can't happen there are two okay when you're looking at an ox at an ldl and an oxidized ldl the process of it converting from an ldl to an oxidized ldl is the conversion of linoleic acid and arachidonic acid the two primary fats in the body omega-6 fats in the body to 4-hydroxynonenol hne and aldehyde mda right the definition of an oxidized ldl the definition is an ldl with mda that's the test that we use right the two those two aldehydes define an ldl so if you're claiming that it can happen through some other pathway it's never been demonstrated anywhere none of these papers make that claim that i have ever seen right if you go and you look at the tests that they do the e06 antibody test or the 4e6 antibody test that you cited in one of your harvard papers they're both looking at mda on and mda and mda modified proteins on the ldl there is no other way that it can happen or at least not known to science at this point okay and d but do you agree that some level of um linoleic acid is guaranteed to be there right like you're not going to have something yes absolutely it's a standard part of the immune system to have linoleic acid in ldl okay so then my question would be why is this not borne out in the la veterans trial again where you go from 10 up to like 40 uh for at least adipose tissue levels or or uh actually intake was around that similar because there's a there's a plateau effect right it immediately goes up it goes up at very low levels and then additional additional levels of intake make no difference so so what what where would that threshold happen what percent of calories if you had to kind of given a rough that's that you know ten percent is it five percent you know if one would if one were to look at leon diet heart they reduced it down to three point eight percent okay so so and most most americans in their diet are we were at like what five percent back in the 50s and now we're at anywhere from seven to twenty percent of our diet is linoleic acid okay so so yeah you know there are clear threshold effects i mean the dha paper on omega-6 fats doesn't mention it but they mention that in their paper on omega-3 fats that there are clear threshold effects of these fats in a variety of different disease situations okay so yes you're not you know if you max out ldl composition of linoleic acid and given the structure of an ldl particle you know okay so a triglyceride has an sn1 position in an sn2 position the sn1 position is always the saturated fat the sn2 position is always an unsaturated fat it can be oleic acid or it can be linoleic acid or you know any any other polyunsaturated fat most likely it's kind of irrelevant but that fat is what makes that fat in that position in the triglycerides and in the 5 like leon would be the the kind of the one that indicates yeah that indicates to me that they're getting it down below the threshold and they're you know through the combination of omega-6 fats that don't convert you know hne and mda are only well hne is only produced from omega-6 fats mda can be produced from omega-3 fats or omega-6 fats in the ldl which is primarily omega-6 fats most of the mda is coming from the omega-6 fats because that's what's what's there okay and and so if leon went from three point eight i think you said three point eight percent two three point eight down sorry down to three from from like 5.3 or something or 5.5 or i don't believe that they did an analysis of the or no sorry not from baseline but from uh in the in the control group anyway uh in the control group i think it was around five point or something right five and a half we'll see just through just to clarify you're talking about the percentage reduction in calories from linoleic acid percentage of calories from linoleic acid yeah yeah and so so that threshold then is is going to be somewhere between at the very least somewhere between we'll say roughly five and a half to 3.8 right is that fair that would be consistent with the results okay so i just want to show the um the hooper uh analysis that you cited if we look at their subgroup analyses actually when you start below five percent of calories and you increase omega-6 intake from there you see again uh now it wasn't quite statistically significant but a 0.77 there reduction risk of chd events and for cardiovascular mortality it's a 30 reduction and it was statistically significant cherry-picking it's the one it's the one analysis that crosses the threshold that you just cited right it's it's the only study included here according to your threshold that is capable of answering i can't even see what study you're talking about you've pulled a table out of a matt can you can you use that paper can you bring that study back up from the beginning and just summarize your position here a little more clearly yeah yeah so so this was the paper that i i talked about earlier that tucker had cited as well and you know i was trying to get an idea of what the threshold was because i knew that this paper broke it down by you know baseline amounts or comparative comparison group amounts um or no sorry it was baseline and then increasing from there and so if we break it down by baseline and we cross the threshold that that uh tucker is talking about then we do see um an increased risk there uh or sorry a benefit crossing that exactly what page is this table on uh page oh it's from i got it from the uh the website so it wasn't like by pages you know what i mean the the when you go to the table yeah it's a table is it should be saying there if you go sub group by baseline it should come right up if you control f that sub group by baseline yeah there are so many analyses i don't know if they they numbered them like that that would have been useful my bad subgroup by baseline omega-6 yeah yeah that would be the one while tucker's doing that matt can you just explain uh why this threshold is important in case listeners are a little bit lost because we've been bouncing around a lot oh yeah sure um so i'm actually not super convinced that there is a threshold at this specific value i'm sure there is a threshold which we no longer see benefit of course i i think it goes in the opposite direction that tucker does i think increasing is beneficial versus versus harmful um but the reason i wanted to know that is that if if there are certain situations where we see the opposite effect or we don't see an effect according to what either tucker or i might expect to see i want to know why that is and and and tucker was suggesting that it's because at a certain point of linoleic acid intake you no longer benefit or sorry you no longer experience harm so i was trying to find out like where is that threshold and it just happens to be that that when you do cross that threshold according to again this meta that he had cited we actually see a benefit there um and this is more specific to uh this is going to be more specific to omega-6 intake versus like a multi-factorial intervention i like the olympia heart study yeah sorry i am joking i can see if i can grab it on the i can see if i can grab it on the pdf um because i prepared this a while back and i might be able to direct you towards a specific page are you getting to a point here right we have multiple studies which i've linked showing that oxidation of ldl is controlled by the fatty acid composition and that this is controlled in humans by diet right so what's your point here well my point is that when it comes to these mechanisms there can be multiple things at play i've i've highlighted there are always multiple things at play when we're talking human biology yeah exactly and so you know and i've yeah we've discussed some of these mechanisms i thought i thought it was you know fair for each of us to present our cases there but ultimately what matters in my view is does eating more seed oil actually lead to higher risk of disease right that's ultimately what we need to look at we need to look at the outcomes and you know there's there's a reason for that when we look at the translation rates of mechanistic reasons yeah go ahead but that's not strictly what we're arguing about here right the proposition is is it reasonable to think that it's going to be beneficial and my in my response is that it's not because it doesn't show when you do a strict high c to a high linoleic acid intervention in humans you show a tendency towards harm not a benefit right and even the papers even the papers that you're citing like hooper 2020 or hooper 2018 say that at best there's little to no impact on it right does that make it reasonable to think that having a high consumption of this stuff when even your own sources say that it's going to have little to no effect well to be clear 2018 was your reference not mine um but yours was huge was 20 20. yeah yeah which says the same thing yeah but when we look at the threshold that you've cited we see a benefit um no i didn't cite a threshold i speculated that there is a threshold and that that would explain a paper and then you came up with some table that we can't seem to find in this well no i i can i can send you over the pdf it's the exact same one it's it's hooper 2018 the the omega-6 paper like i um what page i've got right in front of me yeah so it's page i was just added i think it was 89. um let me just go back up to 89. yeah i'm seeing literally exactly that like exactly what i have on on screen on page at the bottom of page 89 right now i'm not you know it's going to take quite a while for me to go through and figure out exactly what it is you're looking at at this point i think we're i think we should move along but you know i think okay yeah can i just clarify the the positions here and then we can you can just we can agree to disagree and move on if that's the case but matt what you're you're saying is that in in leon diet heart study there was a significant reduction in was it cardiovascular uh events or death yeah yeah and that's right and we were talking about that at that earlier and in that in the intervention group there was a uh two or three percent reduction in linoleic acid right and tucker has taken takes the position that the benefit in that trial is a result of that reduction in the omega-6s you said earlier you think it's confounded potentially by saturated fat reduction and an increase in omega-3s we got to that point now um you you are bringing up a cochrane review that is suggesting if you take omega-6s uh linoleic acid from zero percent of calories to six percent calories less than less than five percent to about two sorry less than five percent to above five percent that there's a reduction in what uh that was cardiovascular mortality and then for chd um it was suggested a benefit it wasn't quite statistically significant and what study were they i mean this is a meta analysis what study were they citing for that change um that one actually funny enough i think was the la veterans um going back to that so le veterans actually increased from about four percent of calories up to like something ridiculous like 20 for linoleic acid or 18 or something wrong complex yeah i i can show you if you wanna look i have the table or the figures here and the tables for intake in that trial um it's easier for you for me to do that yes and that's a confounded study with it's not because the debate is on seed oils these seed oils that they use soybean cottonseed corn sunflower they are the seed oils that we're talking about if you're saying that those seed oils are confounded by ala that they produce a beneficial effect then you would have to well corn oil has virtually no seed oil it depends on which seed oil you're looking at we agreed to talk about high linoleic acid seed oils what do you mean right yeah on my on my presentation i showed you know grouped high linoleic acid linolenic acid plus ala linoleic acid plus dha showing that the harm is specific to the high linoleic acid interventions right yeah which unfortunately happens to be what most americans are eating so yes do i think that if americans reduce their linoleic acid intake and increase their alpha linolenic acid intake like in leon die at heart would their health improve absolutely and i'll note that leone used a canola if i remember correctly or rapeseed oil which is what they call canola over in europe right but most americans don't eat that way most americans get very little ala which is often taken out of these fats because it oxidizes so fast so so the actual diet that we're looking at is closer to the three in the united states what people are eating is closer to the three harmful ones but what we're talking about again we aren't talking about isolated linoleic acid here we're talking about vegetable oils that are yeah typically very high in linoleic acid and some of them might come with a bit of ala so if we look on the very far right part of this figure that's the the increase in ala and yes there was an increase and relative to the benefits we see with ala that might be a significant increase i'm not going to argue that but look at linoleic acid it's like four times as high we see palmitic acid decreased stearic acid cut in half as well um and so like essentially what you what you'd be saying is that that this intervention showing benefit by like predominantly using soybean oil as far as i'm aware actually but you know soybean corn um cottonseed and uh safflower oil they saw a striking benefit across the board for cardiovascular outcomes and all-cause mortality yeah they saw a fairly small effect not a striking benefit so uh what is it i mean you think a 35 reduction in risk and all cause mortality a uh 33 reduction risk of fatal atherosclerotic cardiovascular small let's let's go back to the meta-analysis which meta oh that the one that you've decided yeah and we can talk about those trials i i know we've already talked about minnesota i think we're going to agree to disagree on a few things there and i do want to talk about sydney diet heart and rose as well but what are you saying and i just want to like a simple yes or no it's fine too but like you're saying a 33 reduction risk of fatal ascvd or 35 reduction risk of all cause mortality is a small so we have three studies in this group that's not my question we have dart la plus ala which showed a hr question yeah it is it is the question because you can't take one study right and i mean your little chart showed meta analyses at the top right you can't take one single rct and prove an effect which is what you're trying to do here yeah let's go through these trials one by one one at a time so so matt uh you're presenting the la veteran's trial give tucker a chance to say what he thinks wrong with it and then let tucker talk about the other trials and you can you can bring up any issues that you have with them yeah can i just ask the one clarification question i wanted to ask um sure that's cool and that was just i i just want to know in your view is a 35 reduction risk and all-cause mortality and a 33 reduction risk of fatal afcv considered a small reduction that's just what i want to know regardless of everything else outside of that i just want to know the answer to that question in cardiovascular disease yes it's a small reduction it's a large relative risk it is a small reduction in a lot small absolute risk of death okay right as i said what i'm really interested in is why the half of americans or whatever the number is nowadays die of cardiovascular disease when apparently the number in a traditional society is far far lower right so yes that's a small reduction because most of those people are still going to die of cardiovascular disease but they didn't hear like that's that's the thing that i don't like i don't most most people in the united states sorry correct me if i'm wrong here but isn't cardiovascular disease still the leading cause of death in the united states it is yeah absolutely and if you reduce that by 33 say [Music] is that going where is it going to be it's going to go from what number one to number three that's still a huge death rate oh for sure i'm not saying that it isn't um i'm just saying that as a single intervention like remember a single intervention of just replacing saturated fat with vegetable oils separate fat rich foods and vegetable oils leads to that much of a reduction ignoring all of the other changes no no no no no no no it doesn't because based on your study hooper 2020 it has little to no effect yes that inc and we can i mean we can talk about based on your meta analysis the data from the top of the pyramid that you presented it has little to no effect hooper 2018 looking specifically at omega-6 fats also says that it has little to no effect so matt you seem to do a thing here where you're cherry-picking data out of your pyramid to try and make a point and when you see one number in a study you glom onto it and say oh look this proves my case and yet you ignore your other references and you ignore the hierarchy of evidence that you provided so that you can try and make try and score a point right well no your paper your paper hooper 2020 shows that replacing saturated fat with polyunsaturated fat has little to no effect on low quality evidence if i remember correctly so turning around and claiming oh this is you know this is going to do it all we need to do is increase seed oils and we'll reduce heart disease by 30 percent there's no evidence for that there's just no evidence based on what you've presented well saying there's no evidence as far as i just showed evidence of that so like that's meta-analysis meta-analysis you can't prove it based on one study we agree with we agree about that that was your pyramid that you showed i agree just just quickly and this was your openings this was your opening paper yeah guys um let's just try and keep this respectful but yeah but um with regards as we move on from la veteran study i just i just want to clarify is tucker what what are the the major problems that you see with that study in terms of the design or the intervention i mean it's not you know it it's not a terrible well i mean as matt but there are some problems with it but they don't seem to be super critical in the outcome of the paper i mean it did show a small benefit right a small benefit so does that is that consistent across all like studies not released one of them another one showed no benefit another one showed a teeny benefit right so the problem i have is ignoring the other ignoring the in ignoring the evidence that contradicts the case oh well no i i don't want to ignore i definitely want to chat about all the other um especially like the ones that you mentioned and minnesota again like i said i think we both made our case but happy to talk about some of the others uh like the city diet heart and roads um if you wanted to as well well let's let's let's go to hooper 2020 what are there what's that top line summary yeah so um yeah so they mentioned for what was it cvd mortality acm was that where they said little no benefit their top line intake all cause mortality critical comments critical importance reducing saturated fat intake probably makes little or no difference to all-cause mortality yeah reducing saturated fat intake probably makes little or no difference to cardiovascular mortality right finally we get down on the third line reducing saturated fat intake probably reduces cardiovascular events and then the effect of reducing saturated fat intake on risk of myocardial infarction is unclear as the evidence is of very low quality okay right so when we take your meta analysis that you opened up your argument with where's the evidence yeah so far down in this paper that's geez 290 pages do we have to go before we find some evidence yes just just to interject here matt when you when you reply i think it's important let's just try and stay in line with the debate which is about coronary heart disease where we can yeah that should be the first i agree yeah i was just going to mention that like yeah i'll cause mortality i know i mentioned foreign veterans too i mean that is a separate outcome but um but as tucker said yeah we didn't see a benefit there um overall now for like cbd mortality the issue is uh power you know very few trials are actually powered to detect a difference in risk there now if the reduction in risk is not statistically significant they're going to say yeah probably no benefit um that study are we talking about here for for hooper right as you just read they said little to no benefit for cbd mortality i believe correct you just read that right yeah and so yeah when it comes to mortality i don't think these trials are really adequately powered to even determine that i think events is a far more sensitive um endpoint to look at and you know for that case we do see a benefit and sure it is smaller than the benefit we see in the own diet heart um power power is a statistical game right yeah it's abs look okay here's what power means power means your effect is so small that you need a huge population to detect it and get your p-value up above p05 right so if you're talk if you're telling me about oh it's not powered to detect something what you're saying in english is this effect is so tiny that we're only going to see this tiny little effect when we look at a massive population so you're conceding the point you're considering the fact that they have little to no effect no because that's what power means statistically well no because when you look at specifically mortality now cardiovascular mortality is going to happen less than cardiovascular events i think we can agree on that right not everybody who has an event is gonna die from it right right do we agree on that yeah so not everybody who has a cardiovascular event is going to die because of that event now would you hold the position that reducing overall number of events is beneficial it really depends what the trade-off is i mean i don't know if you know all alcohol is equal let's say all else held equal loss is never held equal in biology hypothetical hypothetical would i want to do a given you know the totality of the evidence on seed oils if i thought i was still going to die on the same day no i don't think that's useful okay but just right so i have two hearted so i have one heart attack that kills me instead of two small heart attacks that's what you're saying basically no i'm not because but that's but that's exactly but that's exactly it right you're saying can you can you rephrase your question matt and i think i think it's a fair one to get a straight answer yeah so so my question is basically that i mean well before i ask the question we both agree that you know cardiovascular mortality is going to happen less often than events right not everybody who has a heart attack will die that much i think tucker and i already agreed upon correct sure pretty much indisputable yeah yeah yeah okay i just i just want to make sure so it's absolutely clear for people listening so then my question would be provided all else is held equal let's just give that that hypothetical for a second here provided all health all else held equal is reducing the number of cardiovascular events a good thing not necessarily man all that so no there's no other factor look if i'm going to have two heart attacks right let's say i'm let's say we're talking about three heart attacks here right one is going to be a little one that i may not even notice right the autopsy data makes it clear the lots people have heart attacks and don't even know about them and then i'm gonna have there are two other ones which is gonna kill me do i care no so okay so you would take the position that a reduction in advance is not necessarily anything to really care about i mean there's one reason why we care about heart attacks or atherosclerosis and that's because it kills you no that is absolutely false i would 100 disagree i have family members who survived heart attacks it is not fun it is not again no it's not ripple you i i got to watch my dad go through it no it's not fun yeah so so i i think saying that the reason we care about heart attacks is just because of death is is not you know it doesn't follow there i mean i certainly care about family members or myself having a heart attack it's it's clearly important right and if they have a heart attack that they don't notice how much do you care about that sure but are you saying that that when we look at these reduction in events we're going off we're going off down on another tangent here let's get back to hooper right no little or no effect that's that's not but when we look at events there was a reduction and are you suggesting that the reduction in events was specific to silent mi actually what they specifically what they said let me go back here because i was trying to look for something else how are you guys going for time by the way i'm okay for a bit i mean yeah i'm fine for a little bit so what did so what do they say for your paper hooper 2020 myocardial infarctions specifically the effect of reducing saturated fat intake on risk of myocardial infarction is unclear as the evidence is of very low quality yeah i can so okay let's put that aside now right you're making an argument that is not supported by your reference well no the cardiovascular events i have it on the screen right here total cardiovascular events i'm looking at the summary of the paper myocardial page three yeah sorry i'm talking about cbd events total cbd events not am i here we can talk about that blind cardiovascular events it's the next line up reducing saturated fat intake probably reduces cardiovascular events yeah and they found it yeah and they found a 22 percent reduction moderate evidence yeah sure that i don't i don't just does the paper define what a cardiovascular event yeah so so it's like mi um it includes uh angina i believe um i believe it includes uh cerebral infarction as well it's a combined endpoint but like so so again my my question or i mean it sounds like like you right so we're looking at small rrs here 0.96 0.94 0.83 0.90 right for which which table i uh and i have it on page three it's right at the top oh no but but i'm looking at cardiovascular events i'm still i'm still talking about that specific um well here then let me let me show the screen page three from hooper this top line summary of this paper right all of these items are critical oh i i yeah i know why we were having uh uh why we're crossing um why we weren't uh seeing the same thing you're looking at reduction of saturated fat i was looking specifically at replacing saturated fat with pufa i'm looking at your paper this is what you opened yes and in that yeah and what i showed was specifically the results for replacing saturated fat with pufa right this is a this isn't a debate about saturated fat it's about which is still no little or no which is still little or no effect i mean but again we can get we can talk about whether or not that the reduction in events is um is important um right so to put this back in the big picture here we are looking at a population that has cardiovascular disease as the number one cause of death and the biggest change in this population over the last 100 years has been a massive increase in the consumption of omega-6 polyunsaturated fats and my question to you is given that we've already done the intervention that you're recommending and observing that there are some huge confounders like smoking in there at what point of influence smoking in which smoking and smoking smoking's a big confounder for cardiovascular disease over the years which specifically agree on that yeah but which specifics paper you said that i cited i'm talking generally here big okay okay sorry i thought you were okay big picture so you're recommending an intervention to lower cardiovascular disease risk you know i would like to get it back to where it is in its cmna where it's undetectable or in the messiah or in the even i'll take the navajo indians from the paper you sent over where they have a fraction of the risk of the framingham cohort right at what point does this massive increase in vegetable oil consumption in the united states start to work wait so we've done the intervention no so we have to look at how it's consumed though i don't believe in consuming vegetable oils in the form of ultra processed foods that are hyper palatable is uh is the same as consuming it in a like veggie stir fry or something i mean there are huge differences there i don't think it is an applesauce okay i agree with that the the most dangerous form of i actually just inter i mean this is i didn't provide you with this reference because we're going off topic here um martin van martin groot felt i think his name is he is the world expert on damage an omega-6 fat and you know what the worst thing to do to it is you know what makes it the most toxic stir-fry worst thing i mean i haven't seen the study so i can't really comment on it um one way or the other but but like happy to see okay so we're so it sounds like though we're making progress here i mean it almost sounds like you're conceding that they're harmful and that how we take them that they're harmful in the context of a highly processed food you just said no they're not harmful in the context of a vegetable stir-fry that's probably not i want to be very clear that's not what i'm saying okay so what are you saying i'm saying that they in the package that is and say typically ultra processed foods uh for the most part you have so much going on you have yeah you have a lot of fat content in the form of oftentimes vegetable oils you have um a lot of refined carbohydrates added sugars salts you have all sorts of things going on to make it hyperpalatable and easy to overeat um that in and of itself can lead to an increased risk of disease i am not saying that the vegetable oils themselves are the culprit there not at all okay fine but let's go back let's go back to looking at what's happened to the american diet over the last hundred odd years the single biggest change has been the increase in vegetable oil okay yeah sure i mean so at what point do we start to see these benefits well look i mean uh on a if you want to look at ecological associations can i share my screen for a sec sure let's clarify matt because i think tug is making an interesting point okay yeah tucker's saying that there has been a significant increase in linoleic acid consumption but we haven't seen the the incidence of cardiovascular disease drop as you may otherwise suspect you're saying well that's because a lot of that linoleic acid is in the form of ultra processed food yes i'm saying that it is not the linoleic acid vegetable oils specifically causing that problem um they could very well be an innocent bystander there um for all we know it could be even worse if they were throwing say butter in there or something uh we can't we can't based on these ecological associations of seeing a trend in one way uh correlating with another trend with zero adjustment for confounders or even consideration for confounders i don't think you can draw any sort of convolutional i started off by saying that there's a huge confounder in there right smoking is a huge confounder yeah i'll i'll concede that a big part of the bump in cardiovascular disease in the middle of the last century was probably due to smoking and that the decline was probably due to to the decline in smoking i mean that's what research for the um american heart association another paper i already didn't include here indicates right i mean nobody's nobody's taking the other side that smoking's really bad for cardiovascular disease risk i presume i want to pose a question to both of you here because i think we're getting to a bit more of an instructive point of this conversation how do you disentangle the influence of an increase in the omega c omega-6 linoleic acid and the fact that say 60 of the us calories come from ultra processed foods how do you go about disentangling that to identify if the omega-6s are inherently harmful yeah so i mean on my end i would look to trials that aim to keep the rest of the diet very similar and look at specifically what happens with uh vegetable oil you know consumption in place of obviously it has to be in place of something ideally in place of saturated fat given that that's kind of what most people would be replacing anyway that's where i like the la veteran's trial as i mentioned similarly if you were to look at prospective cohorts then you're considering okay what are dietary confounders can we adjust for overall diet quality can we adjust for smoking status physical activity and so on it's not perfect but do your best to isolate that as the soul variable that would be my approach which relies on the epidemiol epidemiological data that we've already discussed the problems answer your question not looking specifically at cardiovascular disease but i mean obesity is very closely related to it right we agree that it's predisposing i think we can say and you know there are animal models where they've varied the macronutrients you know carbohydrates fat and protein to try and see which try and disentangle which um experimentally what's driving obesity and in that case it's clearly the fat but they used the same kind of fat all the way through the diet so they didn't try and disentangle the different fats so the short answer to your uh a very astute question is the research hasn't been done there and as i said i'm not a fan of the epidemiology i mean they're you know johnny and edis the famed epidemiologist wrote a great paper where he said every every food causes cancer and he went through and looked at the epidemiological research showing that like you can pick anything and there's a paper out there showing that it causes cancer just to throw out a disease so it's impossible to disentangle that using epidemiological evidence because the data errors are so huge on the inside right maybe you can find a difference between something like a macronutrient where it's a huge percentage and your error in measurement isn't going to be that important but if we're talking about a food that's a small or ought to be evolutionarily a small percentage of the diet and you have these massive measurement errors i mean you know the papers that um matt sent over to support his argument which you know he hasn't brought up yet but i will looking at the correlation between ffq data and linoleic acid it's weak correlation at best in those epidemiological studies i mean that's one of the worst data points that they have just because that's not true that's absolutely true those are what that's what your papers say right and then we can get into in issues like you know nhanes 80 of the debt 50 to 80 of the data they collect is physiologically implausible because a human can't survive on that intake that they get from the ffu and then we look at things like you know where do where will where do people get a lot of their seed oils from eating out does the nurse's health study include that data no you know they look at oh you know they so okay you ate um salad dressing right go look at the back of a bottle of salad dressing it may say vegetable oil includes corn oil or canola oil or soybean oil and they're throwing whichever oil they have that's cheapest on the market in there which is fine for them but terrible if you're an epidemiologist trying to figure out you know if you're at three percent of the diet or five percent of the diet because you have no idea what's coming in the front end and then to turn around and take those small little you know those massive measurement errors trying to figure out these small little percentages of diet and then extrapolate that through to disease outcomes you know grossly it's possible but if you're looking at minute stuff we won't even get into the fact that the adjustments are all baloney i mean oh yeah yeah one of the papers matt sent over this was a beaut it was harvard school of public health i think it was wang 2016 maybe anyway i'm aware yeah i know i know which one you're talking about just let me know okay well just you know for the benefit of the audience so they're looking at oxy oxidized ldl and disease risk or in cardiovascular risk right and they're saying this was a butte harvard school of public health so first first pass they find that oxidative stress is associated with cardiovascular disease and then they adjust it by ldl and they say look it's not related to cardiovascular disease the problem is a month after that paper came out another paper came out showing that the test they were using can't distinguish between oxidized ldl and ldl and in fact in one study i saw the correlation of that test for ox ldl and ldl was 0.999 so harvard adjusted by adjusted ox to ldl by itself and then published a paper that they've never corrected saying oh look it doesn't have any risk well that's completely bogus completely bogus they literally were doing statistical tricks based on a test that granted you know that they just flat didn't understand and they didn't understand wasn't doing the thing that they were doing it for and that is something that you see all the time in epidemiological research they may be really good at building ffqs they may be really good at um you know doing the process of producing an epidemiological paper walter willett he's you know the he's the guy and epidemiological nutritional just i'm getting to a point here okay right so he did he came out with you know some antioxidant papers showing oh we think these papers are going to show benefit turn around when you do the actual study they show harm we have lots of papers looking at for instance vitamin e intake vitamin e intake it's an antioxidant we're talking about an oxidative process here should show benefit well it turns out it doesn't it shows harm why does it show harm because when you look at the mechanisms vitamin e actually accelerates lipid peroxidation oops right it's it's frustrating because there's this huge body of research much of which is totally misleading and when you rely on that you're just looking at you know it's i mean hey it's wall street garbage in garbage out i mean if i got a data set and it failed a simple realities check you know if i got like a trade file from a counterparty and i started figuring out that it was just bogus i don't go through and adjust it i throw it out i get a good file i think you've said you've said quite a bit here so i just want to be conscious of giving matt sorry um yeah let's give that a chance to reply there's a lot there i think the ox ldl stuff we've largely covered i mean we could get into that but i'll leave that to the side for a second here um regarding the the you know ecological associations essentially that you're pointing out and for those listening but those are kind of the correlations between intake and seed oils over time and maybe cardiovascular uh risk can i interject the worst sort of observational evidence by the way so so that but that's what you've been citing right that's what you've been saying because it's a critical point that's what we live in what so it's the worst type of observational research but it's critical to this argument yes because every one of us goes out every day into the real world and eats the food environment that is around us we live in the ecology we don't live in a mathematical model in an epidemiological paper so then if we look at um uh association or if we look at uh cigarette consumption and life expectancy across populations we see a correlation a pretty pretty consistent correlation uh between higher cigarette consumption and higher life expectancy should we consider that critical data here well fine sort of makes my point well no no it doesn't it this is this is the ecological data now if we perform epidemiology which is how we know cigarettes cause cancer by the way um we isolate the effects of the cigarettes as best we can and we see you know very consistent findings there across various populations across levels of exposure and so on um so this is the type of data that you're suggesting is important simply because it's referring to population level intake but we're making inferences based on individual level intake right i want to know if i eat vegetable oils or if my family eats vegetable oils or friends or whoever will they be increasing the risk of disease i don't want to know if on a population basis if they're consuming vegetable oils in whatever possible combinations and and forms that they can possibly do does that correlate at a population level with higher rates of disease it's not really useful to the individual level of like does it actually affect your risk and if if it was i mean then here you go cigarettes increase lifespan apparently and we know that that's not the case uh based on largely epidemiology now i've gone over the um the concordance again between epidemiology and rcts we see a 69 concordance based on the shrink shackle paper and you were mentioning the trials like with vitamin e and all that was that referring to the young and car analysis that you cited i know they like that just i mean there's a whole body of evidence looking at vitamin e as an intervention to try and prevent ox ldl oxidation or ldl oxidation it doesn't work and in fact it's been found to be worse okay you know just i don't remember the paper right off the top my head i don't think it's one of the ones that i included it included yeah and you know i um i don't have many qualms with that i just wanted to to understand uh specifically where it was coming from um if it was that paper or something else um other than that i can't actually there was a bit of a a lot of info that came out there um i'm not sure exactly if there's anything else that i didn't address uh but i think i've made my main kind of point there so um i think as far as trials that we didn't talk about like you know the one that that i definitely wanted to talk about a lot was early veterans we got to that uh we talked about minnesota um maybe i don't know if you want to talk about like sydney and rose as the two other ones that you suggest were um specific to linoleic acid is that well yeah they they both show harm i mean rose shows a lot of harm right and yes they're small studies and i know all the issues with it but you know yeah so again you can't you can't keep you know i mean this is one of the big problems that i have with folks like mauzafarian or even hooper for that matter is that they are very good at tossing studies for in some cases invalid reasons right so so yeah like i mean with with sydney i think it is valid simply because i mean it's clear that they use the miracle brand margarine um in their intervention group and we see here that the deputy church the composition of which we have no idea um no well the the person who worked for unilever who was the company that produced the margin says that it had about 50 15 trans fats um and how does he know that how when were they measuring it they surely weren't measuring it then well i have no idea honestly how they know that one way or another but i'm sure that that if anybody were to know it would be him i mean based on the fact that like honestly not good yeah that's just not compelling and at any rate we're back to the same issue we had with mce and you know at some level trans fats cause and i think they even knew it right so they specifically designed the intervention because they wanted to lower their whole point of their intervention was to lower cholesterol and they were aware that trans fats wouldn't have that effect and they did lower cholesterol yeah you know we get back to the same argument we had about mce and at any rate you know we're always going to look at these studies being confounded by trans fats well right because the process of cooking seed oils produces trans fats not the same amounts if we're if like let's say for a second it's 15 trans fat in there you're not producing that by cooking and unhydraulic show me okay but show me the number what's the what's the percentage of trans fats from a reliable source in that paper in that paper again we don't know we know the brand we know the time we know that we don't know yeah i agree okay so it is possible we can at the very least say it is absolutely possible or we can say it's absolutely not possible okay so we can say that the biomarker that we're tracking the whole point of the intervention shows that the effect we were having the effect that we wanted to have which was lowering cholesterol and that nevertheless more of them died yeah but i mean you could lower cholesterol and and continue smoking or they could have adopted a smoking habit and lowered their cholesterol through dietary means and we wouldn't say that they you know benefited there could be other mechanisms that play i'm not married to the idea okay the trans fats simply you know one other increase one other point back to the mce data one of the things that ramsden points out is that the lower the greater the cholesterol lowering the worse the mortality right okay which would suggest that the people with the least trans fats by evidenced by their lower cholesterol are having worse effects or they were having more of the margarine that also was high in pufas the lower like i mean they had the lowest they had the lowest cholesterol so cholesterol level is a marker of trans fat intake no well it's a marker of various fats i'm sure do you agree that there are like it goes up on certain types of saturated fats it goes up on trans fats it goes down on linoleic acid that's why linoleic acid is being prescribed in all of these trials because it has the effect of lowering cholesterol yeah and so if which is supposed to reduce heart disease rates except when you do it on a strict linoleic acid intervention it doesn't well so if we look at it is absolutely possible that linoleic acid intake in a trial that also included trans fats leads to a net reduction in um ldl absolutely possible or serum they measured you know typically serum total serum cholesterol in these studies based on the time or at that time um but we know that if you have a high linoleic acid margin that is also you know relative to typical trans fat content high in trans fats although much lower than the overall linoleic acid or pufa content you can still see a net reduction in cholesterol but still have net harm i am not at all married to the idea that trans fats um increase risk solely based on ldl uh lowering or sorry uh raising uh there could absolutely be other things that play there right well they're they're yes they have effects negative effects on mitochondria and you know so does so does a little anic acid type so it's all demonstrated the fact that they lowered um cholesterol does not negate the possibility that um that the trans fats were able to produce harm okay so let's take a look at your if i can find this um so this i thought was a fascinating little paper because they're showing olive oil is beneficial they're claiming it's because of the oxidative effect oddly and here's what they say we estimate that substituting 10 grams a day of other fats including margarine butter mayonnaise and dairy fat with olive oil was associated with reductions in the risk of total and cause specific mortality well i'll observe that two of those margarine and mayonnaise are seed oil foods right so this paper seems to be saying that certain seed oil foods which are not cooked mayonnaise generally is not cooked are harmful right now this paper has so many problems with it not least of all is that they seem to have adjusted their data set to eliminate polyunsaturated fats which i find and i've looked and looked and looked to try and figure out what exactly they're doing and why they're doing it sorry can you explain that what do you mean by their adjusting for polyunsaturated fats yeah i know um third the models were adjusted for modified are we looking at this now uh yeah can you guys see my screen can say it the models were adjusted for modified a-h-e-i which is their healthy eating index their adjusted healthy eating index without polyunsaturated fatty acids and alcohol components why on earth would they be doing that and they've adjusted these models several times in the past to try and come up with something that was actually healthy because the first couple versions didn't work so we're looking at a study where they're claiming that replacing olive oil of certain seed oil fats with olive oil is a benefit to mortality and will note that this is all a model it's not actually intervention right it's the sort of thing that we were discussing before right but they're showing that lowering the seed oil consumption for certain elements increa lowers mortality and they're claiming this is the benefit of olive oil which by the way i completely agree with them with agree with agree with them on but it's a bit of a mystery you know just as kind of you know this was the one of the charts that i led with up here less susceptible to oxidation why is olive oil less susceptible to oxidation because it has less omega-6 fats right those are the fats that are oxidized um and what does that lead to anti-inflammatory and anti-atherogenic and they list you know i mean this is a classic case of a paper that's got something for everybody matt can be happy that they show butter is bad and i'm happy that they show mayo and margarine is bad this is 22 2022 paper right so even this seems to show that there are cases where seed oil consumption is harmful although observing this paper the fact that they remove polyunsaturated fatty acids i don't even know what to make of this paper how can they how are they including mayo if they've removed polyunsaturated fatty acids so um yeah i think that's explainable so for starters the alcohol you know how they left that out that's because it was already adjusted previously so it was adjusted in a previous model right yes yeah okay just making sure that's clear um now for polyunsaturated fat it wouldn't make sense to adjust for pufa when a part of what you're comparing is is you know includes pufa right you don't want to adjust by itself it doesn't make sense so well they're saying without the components they're taking these components out yeah so they're adjusting for the ultimate healthy eating index but not including pufa in that score because you want to look at the effect of pufa given that the oils that they're substituting right you don't want to adjust by the intervention that would be an over adjustment or not the intervention you know i mean the the exposure that we're looking at it would be an over adjustment to adjust for pufa right no no but they're not adjusting it they're removing it yeah they're exactly that's my point they're removing it from the model yeah they're removing they're removing something that they have to have in here to show an effect of olive oil no no no no they're they're they're they're and they're obviously and matt matt they're obviously not removing it because they're including margarine and mayonnaise right in the model that's what that's well no just so they're they're we estimated that substituting 10 grams a day of other fats including margarine puffa butter not puffa mayonnaise puffa and dairy fat not buffa with olive oil was associated with reductions in the risk of total and cost-specific mortality yeah so can i offer a suggestion here because it seems like you're both looking at this from different angles can you provide an example of why you're saying that would be over adjustment using some other example perhaps yeah so what would be a good example um i mean i okay i mean i think we disagree on this anyway but um ldl is the typical one so let's just let me just use like saturated fat for example and just just bear with me tucker i know we disagree on the topic anyway but it's just easy for me to use this as an example um so if you want to look at the association between say saturated fat intake and cardiovascular disease right and if you suspect that saturated fat increased risk of cardiovascular disease by raising your ldl cholesterol levels you don't want to adjust for ldl levels because you would be removing kind of the factor that is responsible for the outcome or at least one of the the main factors that is responsible for the outcome right if a leads to b and b leads to c but you remove b you're not going to see an association between a and c that's not exactly what they're doing now i know i know it's not i'm just i just wanted to explain over adjustment for a second and i'll bring it back to this um here so in this study when they're looking at replacing fatty or substituting fatty acid sources and when some of those fatty acid sources are rich in polyunsaturated fats and all of them to a degree obviously contain some polyunsaturated fats um you don't want to adjust by polyunsaturated fat intake because you if you if you're feeding something like vegetable oils that are high in polyunsaturated fats but you make sure you're only comparing that amongst people who are eating similar amounts of polyunsaturated fats well then that's useless information and so they removed it from the model so that they don't over adjust in that sense yeah that sort of sounds like they're trying to show that uh saturated fat causes cardiovascular disease and they're removing dairy from it sorry um did they do that here with the dairy i don't see that no they didn't do it with dairy they did it with polyunsaturated fatty acids right which are the components that lead to oxidation in this model anyway we're definitely going oh i i think there's a huge misunderstanding here so it's not the same right it's you are look the way them removing that from the model um or from the um healthy eating index score allows us to see the effect of changes in polyunsaturated fat intake it doesn't remove that effect if they were including it in the model and adjusting for that variable only comparing people consuming similar amounts of polyunsaturated fats then you would be removing the impact of vegetable oils so the way that they did it is actually appropriate for determining the effect the vegetable oils may have and funny enough in this in the plot when you replace olive oil with vegetable oils if anything there is a trend towards benefit for cardiovascular mortality um which one would expect to see um at any rate sorry this is a bit of a tangent here um although you know again if we're going back to the mechanisms of how these things are causing harm what this is suggesting is that reducing polyunsaturated fats here i mean they kind of backend it where they say that the you know olive oil also exerts anti-inflammatory and anti-atherogenic effects yes it does because it's low in omega-6 fats and it may help reduce oxidative stress again the definition of oxidative stress as used in the literature is the presence of oxidized omega-6 fats right okay but according to their results um if anything vegetable oils may have even been beneficial over olive oil unless they're in mayonnaise or margarine in which case in which case um i believe this uh if you scroll up to the um i mean if you if you care to the uh timeline here this is going back to i believe the 80s um and uh and again trans fats they've been compounding the margarine there and mayonnaise is not just vegetable oil it's also filled with eggs egg yolks and that as well so again our debate is centered on seed oils that was the debate proposition that was the topic at hand it is not these other uh foods that might contain vegetable oils one way or another yeah guys let's come back to the actual debate at hand and what was that what was it in this study here what what was found for vegetable oils seed oils yeah so um basically they did a subsidy i mean this study was centered on olive oil let's be clear but they did a substitution which is a low linoleic acid that's not sweetened saturated right and they found that when you replace olive oil this was per 100 grams per day uh based on their model i have you replace olive oil with margarine butter or mayonnaise uh you have high risk of cardiovascular mortality uh same with if you replace with dairy fat but if you replace olive oil with vegetable oils there weren't any statistically significant differences and if anything it kind of leaned towards their being benefit with the vegetable oils over the olive oil for again the cardiovascular outcome which is what we're uh talking about yeah it's an odd study but like do you have any you know criticisms for that otherwise as far as the you know the [ __ ] yeah i have i have a criticism i mean i went through this with another study from this group where they hid data that they i gather didn't like um by grouping odd things together right like grouping margarine butter mayonnaise and dairy fat now there are lots of studies on dairy fat and mortality showing that it has a benefit so it's a little odd here that they are grouping it together with margarine and mayonnaise given their composition with then butter and dairy fat even in studies that show dairy fat has a benefit you know butter specifically seems to show a bit of harm and we you know that's another long topic but yeah and what they did at the end of that what they did in another study you know back to the correlation between obesity and heart disease is they showed that potatoes fried in seed oils were far and away the most obesity genetic food that they looked at and then potatoes not fried in seed oils boiled or with butter had minimal obesogenic impact and so they grouped them in such a way to minimize the effect to minimize the to make it less obvious that what they were seeing was an effect of seed oils not an effect of um carbohydrates as they claimed because obviously when you compare a boiled potato to a french fried potato the thing that doesn't include increases carbohydrates so they seem to be yeah i was just going to say that just like this all seems a little bit tangential to the point right like the question was specifically about the olive oil for vegetable oil replacement not the issues that you might have with the dairy fat replacement or the you know the other outcomes we're looking specifically at vegetable oils right being the topic of debate here so um like do you know why so why did you include this paper then um didn't you i'm pretty sure you included this one i mean we both did we both did actually yeah i included it simply to show that like i think both olive oil and and vegetable oils are healthy i think both are great and this shows that both you know overall for total mortality very similar for cardiovascular mortality which is the topic of debate possibly even beneficial for vegetable oils um so i included it really for that reason because i figured you do as far as i'm aware anyway believe olive oil is a pretty healthy oil so um i'm just wondering like do you have any qualms with that finding specifically not necessarily like i understand there's different types of dairy fat different sources of different effects like i agree with all of that uh but specific to the olive oil for vegetable oil replacement well i think you know i think that the claim decline mechanism here supports my argument that you know if they're trying to born out in the outcomes well potential mechanisms for olive oil intake and mortality less susceptible to oxidation right well and that's only one thing that's getting oxidized in any of these fats i think guys guys i i think the contention here is around why there was not a significant difference between olive oil and vegetable oils in terms of risk that's where it seems like you guys are disagreeing here yeah i want to agree that is that is what i'm trying to um you know hammer home here right and yeah i'm just making the point that if you're you know as with all of these studies that i have referenced previously if you want to reduce oxidative stress you do it by reducing omega-6 fats in the diet practically speaking they're showing they're saying here that that has a benefit sure and i i can grant that all to all fine that's fine but why do we not well then the flip side of that argument is you know fats that are more susceptible to oxidation would have pro-inflammatory and pro-atherogenic properties but if you scroll up to that list of all the mechanisms there are a whole bunch of things at play it is totally possible that one of them could be slightly worse in the case of pufa and all the other ones improve the mechanisms right central administration yeah yeah right so let's just say i'll grant that let's say that the vegetable oils increase oxidation but they actually improve guess what lipid profile they improve even more than than olive oil um or although i don't know if there's really statistically significant differences but anyways that generally trends that direction for some of these other um mechanisms maybe mechanisms that aren't listed here there could be further benefit to the vegetable oils over olive oil and they balance out or potentially even end up being better for the vegetable oils so my question would just come back to that why do we not see a difference and if anything why does it trend towards benefit for vegetable oils if it is the case that vegetable oils are wreaking havoc in these ways that you present i think that's a fair question i feel like that's kind of where we need to get to in terms of uh resolving this paper i don't know that this paper gives an answer to that question at least it doesn't to me um you know we go back to the question that if you want to resolve that if you want to figure that out you have to do an rct in people okay so it's been done so this is just a this is just a model so but for this specific study then you don't actually have a specific issue with that finding just to be clear other than the fact that you think you think there might be better data that's fine but for that finding you don't actually have an issue no i'm just planning i'm pointing it out that there you know if you think that omega-6 fats are beneficial then you wouldn't be saying that olive oil is better because it doesn't have the qualities that are associ uniquely associated with omega-6 fats in the diet um i didn't say olive oil is better well this no you didn't this paper does where does it say olive oil is better than vegetable oils better in general right i mean if you're if you're thinking it says it right in that paper that we were looking at right where potential mechanisms for olive oil intakes and mortality less susceptible to oxidation there's nowhere there that it says there's nowhere in this paper where they say that olive oil would be better than vegetable oil no they didn't pass their disease you're right they don't say it you have to be familiar with the mechanisms behind the terms that they use right sure and so when they say oxidative stress the markers that are used for oxidative stress are oxidized omega-6 fats right there's i mean i've already cited a whole bunch of papers showing that what is oxidized in ldl is the omega-6 fats and that if you replace them like they would suggest you do with uh mufas like oleic acid in olive oil that you make them less susceptible to oxidation so no they don't explicitly say it you have to understand the mechanisms behind what they're saying you have to read between you have to understand the terms that they're using and the implications of those terms i mean i think i think a good point here or it'd be a good point um if you don't mind if we just talk a little bit about why i don't place as much stock in these mechanisms um so if you don't mind if you're okay with that drop my screen and you just have to unshare okay yeah so um as i mentioned earlier i mean i talked about the concordance between nutritional epidemiology and randomized controlled trials human outcomes um here i want to just review some of the research on mechanisms and how they translate to outcomes so for starters especially given that animal models are typically used if we look at the translation of cardiovascular animal models to humans wait a minute simon i haven't my main argument has been using human data not animal model translation this is a huge tangent no it's not because i'm going to get to in vitro as well it's just i have a few slides in a row i if you just indulge for like a second here i'm not going to matt you can you can go through it but just be careful not to misrepresent tiger's position yeah and and so i mean just the we can even skip over the animal part that's fine um but if we look at for example highly promising cancer uh biology discoveries so you know mechanisms or you know this could be a mutual research it could be animal research too sure um stuff that is thought to be groundbreaking you know these are our mechanisms that are thought to really change the game over the course of 15 years we see a translation rate of about 19 yeah okay matt again you were talking about something that is totally outside the scope of our discussion here we started off pretty much talking about human data and i've tried to stick with the human data and not to get off on tangents on animal models sweet no this isn't necessarily relate to any data that tiger has presented yeah so when we're talking about the mechanisms that play like oxidation and stuff like these are oftentimes done by um you know if you're testing say lag time to oxidation these aren't done in the plasma these are done in like petri dishes essentially you exposure to copper yeah yeah and the model i say this paper i studied even 1993 was looking at humans not okay and that yeah but but were they were they oxidizing sorry which paper in particular or even 1993 where they were feeding olive oil or some seed oil i can't remember which one and then looking at the effect that that had on the oxidizability of ldl okay yeah and so and that is the case by the way where there are animal models and the human data is been shown to be consistent with the animal models of the oxidizability of ldl that was done back in the 1980s in describing the mechanisms now this is important here what i'm showing because this is translating the mechanisms to effective treatment to actual outcomes and this can include human trials looking at mechanisms this can include animal trials looking at mechanisms this can also include in vitro trials it's not it's not split up but what i'm pointing out is that there are you know mechanisms that could be at play that don't bear out when we look at the um the outcome data similarly if we look at clinical trial success rates from phase one which are human studies by the way phase one human trials through phase three and through approval we see a six point nine percent clinical success rate or a success rate in translation that is abysmal especially compared you add a you had a or you shift a decimal point matt this is totally a tangent how this is comparing human data um in phase one trials uh this is because we've already been discussing human endpoint rct i understand i understand but what you were just referring to were a bunch of mechanisms as a reason to believe something is the case that have already been demonstrated in human rcts yes but you're but you're suggesting no but there's contradictory there's contradictory information amongst the rcts and you know we've discussed some specifics of individual ones sure we do i mean take statins there are contradictory results in those rcts too that doesn't mean that we don't prescribe them matt so just to clarify you're saying because you were looking at the outcomes in randomized controlled trials and that there are you're agreeing to disagree at certain parts there and there are some some conflicts in the data that you want to bring it back to why the mechanism uh may not be something i want i want to basically highlight why suggesting that mechanisms are at play does not trump the outcome data now we can we can disagree on what the outcome data shows that is fine but we i don't believe we can disagree about the outcome data based on mechanisms that is the point that i'm trying to make we have abysmal translation rates between mechanisms to outcomes and we have far higher translation rates from nutritional epidemiology to outcomes in rcts so if one were to consider mechanisms as a reason to claim that an outcome whether it be in a randomized trial or otherwise is not true or is somehow debunked by this mechanism you'd have to accept that nutritional epidemiology is really solid because we have a much higher according to this 10 times higher translation rate that is the point i'm trying to make i still don't see the relevance but whatever um what i was trying to explain to you is the mechanisms in the paper that you shared not really supporting your argument but and i'm just saying that the outcomes do right that that's what i'm saying i'm saying the outcomes do regardless of what the they're not even in that paper do they because they say that replacing foods which are rich in vegetables in seed oils benefits mortality in your paper yeah well well i'm not i'm not talking about foods that contain vegetable oils this debate is about but you can't distinguish them all foods look nobody well nobody drinks soybean oil yes soybean oil is a component in food you cannot separate seed oils from foods that contain them that's an absurd claim to make well nobody's like but that's like saying you live in a house with um smokers and you wouldn't be at higher risk because you don't smoke or something like it's that there's you're still being you're being exposed to other factors poor analogy sorry i mean it's right but you can't distinguish you can't distinguish the you can't distinguish the effect of drinking soybean oil and the effect of consuming soybean oil and mayonnaise can you bring this study back up matt yeah uh which one in particular well it seems like you're still talking about that olive oil paper to me yeah i'll bring it up i just know we're nearly at uh three and a half hours so i kind of i think that we haven't okay i think we have another i think we have another 10 minutes of going back and forth and then let's get into the closing statements okay so this is it anyway simon if you wanted to if you wanted any specific clarification one way or the other and what's the point that you're trying to make here matt so basically what i was saying was that um why i you know clarified that the fact that they mentioned certain mechanisms here which they don't actually use to say that um that olive oil is superior to vegetable oils in any way they just suggest that they are some of the benefits potentially to olive oil consumption that's not entirely yeah they suggest that the benefits are due to those mechanisms sure sure right and then as in this chart they show specifically that replacing olive oil for margarine and for mayonnaise shows a benefit and then this is my big problem with this paper and then they segregate out for other vegetable oils so of the three categories of vegetable oils that they show in this paper two of them have harm that is reduced by switching out to olive oil okay yeah and i don't take issue with that because those aren't looking specifically at vegetable oils my point is that when you look special what's margarine made of it's made of uh vegetable oils and other compounds i mean it could be especially given the time of the follow-up now i can't recall exactly when this trial started but i believe it was around the 80s um and margarine has been made with vegetable oils for at least the last hundred years they stopped using beef pillow in the age but i'm talking about trans fats no i'm talking about trans fats but the thing is you don't think trans fats are necessarily a problem no but i'll look i'll concede but also they started reducing trans fats in foods quite a long time ago and they i do believe they address that in this paper but again trans fats are not going to be in mayonnaise and eggs have been shown repeatedly not to be harmful so i mean that's a separate no that's a separate topic but for sure that's a separate topic maybe another time but uh but i think we can agree at the very least that they are the margin of mayonnaise are not isolated vegetable oils can we at least agree on that no we can't agree with that because they are because everything isolated vegetables every food that you nobody eats vegetable oils alone you eat them in something else you eat them in french fries or in salad dressing or whatever else so you can't say that there's some distinction between mayonnaise and salad dressing right i'm still consuming vegetable oils okay i agree that we're still consuming vegetable oils my question is do we agree that if we're just to take a bottle if i just a bottle of canola oil sitting here on my my desk and would you say that that is that that isolated vegetable oil is exactly the same in its contents to a bottle of mayonnaise or a container of margarine that's exactly not come on yeah so that's my point they aren't the same now margaret but they are the same at the end of the day when you eat them as part of your diet right and you're not really guys let me let me interject here because we are getting to the end and i think you're kind of talking past each other a little bit i think the interesting question here for you to discuss is why there was not a significant difference between olive oil and vegetable oil when vegetables were isolated uh with regards to total mortality dvd mortality cancer mortality etc i think that's the interesting point that you guys are currently disputing yeah and that's that was my point like so we can we can talk about you know is mayonnaise going to be beneficial versus the other thing but the question i'd like to ask is if you have a salad that you normally put olive oil on and you swapped it out for canola oil would you be at greater harm like let's just start with that very simple would you be fatty acid composition's almost the same so if you replaced it with um soybean oil we'll say or safflower oil i i would expect so yes you would expect the results to be the same no i would expect it to be worse to be worse okay well why isn't that borne out in the evidence here it's epidemiology man so i mean but then again i mean we can't really rely on mechanisms if that's the case based on the well you can rely i mean that's why rct's trump epidemiology yeah and i think i've i mean i think yes i do find it really interesting when even the harvard school of public health shows a study that suggests that there is harm from vegetable oils like they do in this study um you know it's interesting because that's not their position and when they're claiming the mechanisms are the same mechanisms that i'm claiming i find that very interesting because those mechanisms are well demonstrated and they're well demonstrated to be pertinent in humans right over the last 40 years of research and you know when your uh born 2020 study is citing the same line of evidence starting with the same 1979 paper from gol from uh or the 1989 paper you know there's there's a clear path here of harm from vegetable oils in the literature including in the papers that you're citing and when you're opening paper hooper 2020 is showing no benefits then you know i think that's really interesting and i don't know maybe we should that's starting to sound like a summary simon so maybe we should transition into that yeah i think we should transition into that i think uh it would be good to get some closing statements from you both and then i might pose a few questions to both of you from a very impartial point of view uh and we can close it off would you mind doing that before we do the closing statements yeah sure uh one question i did have uh and it comes from the olive oil paper that we were just looking at there i know we we went into mechanisms and tucker you've mainly focused on oxidation matt you didn't really bring a lot of different mechanistic data but i'm i'm interested with regards to seed oils uh and different mechanistic pathways other than oxidation do have either of you come across data that looks at well how do seed oils affect inflammation how do they affect the microbiota how do they affect endothelial function lipids has come up how do they affect insulin sensitivity how do they affect blood pressure because i'm kind of thinking about zooming out here what what is the the net effect of these seed oils on physiology other than the mechanisms that both of you have sort of brought to the table so far how many more hours do you have so yeah say like this is gonna be um yeah and i'll just i'll just clarify on the point where you said i didn't really um talk about mechanism i mean the one thing that i did bring up was uh able b or ldl concentration or proxy for april b i think that's in my view the primary mechanism there may be others at play uh at the end of the day i think what matters is ultimately the outcomes um which is why that's where i place my focus but um as far as some of these other outcomes like for you know inflammation overall i mean there's a meta-analysis i think i sent this one to tucker as well um there wasn't really association with any inflammatory markers now in the dose response those with or might even a subgroup analysis or anyways the highest intake of canola oil suggested a slight increase in crp which is one of the markers of inflammation however if you go into the specific studies on the high end there there was still a reduction it's an artifact of the way that they converted the findings from i believe it was median intakes to means to averages so they had some people were really high inflammation in the control group at baseline so they reduced their inflammation more than the other improvement like again that's why i'm like tucker's like oh man how many hours do you have because i feel like we could talk about that forever too um but i see a bit of a reduction there i believe there's some data on some other you know um mechanisms as well but i just i don't think it's worth uh spending however many um hours at this point uh talking more and more about yes i'll just kind of leave it at that i think it's you know take a couple minutes yourself and then we can just move on yeah so ramsden moved on from heart disease to migraines and he did a targeted intervention lowering specifically linoleic acid and also one lowering lowering linoleic acid and increasing omega-3 fats and showed clinically significant benefits in an inflammatory condition so we have clear evidence through that one paper that um reducing omega-6 fats can reduce inflammation and you know it's a really fascinating paper for a variety of reasons but the main one was that the dietary intervention was more successful than the drugs used to treat migraines so you know ox ldl is well described in the literature source of inflammation um yeah just hordes of papers on that particular topic especially as regards cardiovascular disease the standard mechanic the standard protocol for inducing insulin resistance in humans is to inject them with soybean oil so yeah as like you know there's there's a lot going on and a lot of that ties back into this stuff i mean if you give monkeys an antibody against ox ldl and it lowers their insulin resistance so you know there's there's a lot going on beyond just this i've let me ask this question and i think we've both pretty successfully managed to stay on topic on this debate though is there i'm going to ask this question to both of you so for tucker are there any uh mechanisms whereby you see seed oils being beneficial and matt are there any mechanisms where you see seed oils as being harmful or deleterious there are isolated conditions where they do seem to be beneficial for instance heart failure um heart failures associated with mitochondrial breakdown and the meth the mechanism for that is the omega-6 fats and the mitochondria are getting oxidized replacing them with more omega-6 fats repairs replace allows the body to replace the damaged fats with undamaged fats so i mean there are some specific cases where you know they do show benefit i think the totality of the evidence is not really there for anything other than specific cases where they do show you know specific edge cases like that where they're showing benefits okay and yeah for for me i think one of the ones that you know can't really go without mentioning is is when we look at any kind of oil really they're incredibly calorically dense right it's like pure fat it can be super caloric dense chlorophyll does um when you know if replacing more say fiber rich foods lower calorie density with vegetable oils it is entirely possible it can contribute to something like weight gain you know in the cases of some people i have some athlete you know friends in that who um have a difficult time eating enough calories especially on a more plant-based diet where it's actually beneficial in that sense but for someone who's maybe trying to lose weight uh you know keep keep uh excess weight off then it can be detrimental in that fact in that sense as well so i think that would be one of the kind of simpler ones to just mention okay my final question before your closing statements is to both of you and that is whether or not anything today changed your position at all in any way shape or form or at least gave you something to go away and think about a little bit further i mean i would say like my overall position hasn't shifted here i mean i don't think tucker says either um but at the same time i think yo he raised some points that maybe i hadn't considered they were um regardless of how important i think these mechanisms are when we have outcome-based data i think they're interesting they're interesting to learn a little bit more about uh potentially uh learn a bit more about some of the pathways there um but uh yeah ultimately at the end of the day i don't think it shifted the needle for me as far as as the effect that vegetable oil consumption would have on our overall health or specific to cardiac outcomes as the topic of the debate kind of states um yeah matt did an excellent job of assembling a body of evidence to support his position um almost none of it was new to me so it didn't move the needle for me okay and i do have one other one and that is uh if you could be in charge of designing a study setting up a study that would bring greater clarity to this topic what what would that trial uh look like well uh you'd have to do it in animals for starters um and i would want to see there are a bunch of different papers looking at isolated or semi-isolated fatty acids in the diet and the effect on various factors there are clear signals there for harms from omega-6 fats but it's really you know often it's just one or two fats and they're not looking at a spectrum of fats i mentioned before a study where they looked at three different macronutrients and they vary the ratios of all the macronutrients to determine the effects on obesity i'd like to see somebody do something like that with different fatty acids you know the natural spectrum of fatty acids that would be included in a natural diet and see see if we can tease out what the effects are because there are clearly confounders i mean as we've discussed here um omega-3 fats are a confounder for omega-6 fats and vice versa and we need to in order to have good dietary recommendations we need to understand what all those interactions what those interactions are because you know the dietary guidelines are lacking when it comes to the omega fats and their effects on health it's not accurately reflected in the recommendations i think okay and honestly i would like to see i mean this will never be done nowadays but i'd like to see a similar trial to the la veterans trial but with three groups i would love to see one where um one group got olive oil in the same way that the intervention group got the vegetable oil so you'd have the you know control group with the saturated fat rich foods you'd have one with olive oil and then you have one with the vegetable oils that are higher and polyunsaturated fats lower in monounsaturated fats i think that would it would just be really interesting because at least across the epidemiology we don't really see much of a difference between olive and vegetable oil so uh teasing that out in a long term rose corn oil did that yeah and and i mean but it was just a really small trial overall there right there weren't any statistically significant findings if anything olive oil were trended towards harm well for for the outcomes that we're talking about of course yeah um and you know even olive oil ended up on the side of potentially a little bit of harm relative but i i think it's largely due to just being underpowered let's not get back into it sounds like we're both on the same page there that there needs that there's a lot lacking in the existing body of evidence and it would be nice to see some more of the interactions of these different fats to help us get a better idea of what's going on here yeah matt you can answer that sorry i didn't i don't know what the question was no i was just observing that we seem to be on the same page that it would be good if we had more evidence looking at the interactions of these different fats um yeah specific to outcomes in my opinion but yes right i agree seeing the differences would be good i'm gonna be gradient and ask one more question you said last question twice now i quite like i quite like uh getting you guys out of the the debate mindset i think it's refreshing to finish this way um but last question is there a chance that your position is wrong i think there is always a chance that your position essentially anything could be wrong now for me to change my view i would need to be shown that my you know the the data i presented or my position which uh you know we can say relate to the evidence hierarchy is systematically false and that the data presented otherwise is free from the same shortcomings or similar or even worse shortcomings than the data i presented if that were the case i have no problem changing my view and um yeah so that's kind of how i would answer that yeah every time i read a paper i get butterflies in my stomach thinking that it might prove me wrong going through matt's series of papers was and painful exercise i thought you saw them already um well you know i'd seen most of them um but not all of them and there were a couple of interesting ones in there um so this was a useful exercise that way um nothing moved you know nothing to move the needle but yeah it's you know if if you're not willing to be proven wrong you're not a serious person and i got into this whole thing through basically being proven wrong doing something that i thought you know i mean i hated dietary diet stuff growing up and i watched my mother yo-yo on weight watchers and i thought the whole thing was bogus until i saw huge health benefits myself and had to do 180 on a couple of things i mean it's just sort of totally off topic but i'm super gluten sensitive and when i initially heard the proposal that wheat could be bad for you i thought it was bogus and you know it wasn't bogus for me it like totally transformed my life so yeah it's a painful process to go through sometimes but um you know you're not serious if you're not willing to be open to the evidence okay great and i think it is worth reiterating to everyone that you too may come at things differently but you do agree on the fact that ultra processed foods are probably not a great addition to the diet and they do make up 60 of the average person's calories so you agree on more than than what you disagree on when we look at it like that i might hand over to tucker first to do his closing statement given matt went first at the start um and then matt you can jump in sounds good yeah so as i think i got across in this strict the problem with the modern american diet in my opinion is that it's a high linoleic low omega-3 diet and the studies that we've looked at today that look at that intervention tend to show harm and they don't tend to show benefit and to the extent that you can lower omega-6 and raise omega-3 that's when they start to show benefit um seed oils specifically tend to be on that end of the spectrum the high linoleic low omega low omega-3 and you know the meta-analyses that we went over you know there's very little evidence to show benefit from these interventions and what little bit of evidence there is tends to show harm from these interventions so after several decades of research i think it's unreasonable to claim that there's going to be a big benefit from upping your consumption of linoleic acid in pursuit of the prevention of heart disease but definitely do stop smoking another thing we agree on um yeah so i guess i'll take it away um i think ultimately when it there's always going to be conflicting evidence of course on either side and we can find the same for smoking too but the thing is when it comes to the trials on vegetable oil consumption as a whole so not isolating to you know specifically one type of fatty acid over the other but actual oils soybean corn cotton seed etc we see that there is benefit whether it's generally in that direction of benefit whether it is a significant benefit or in the trial that as i've outlined i think is best suited to answer the question uh being benefit um i think it's pretty convincing data and sure we can look at some of the studies like cindy diehard and mce as uh showing us harm from specifically linoleic acid but they were also the ones that use a lot more margarine and of course in my view that could very well be trans fat compound um obviously we disagree there and people i'm sure have heard that by now but basically when we look at the evidence from dietary intake in rct's dietary intake in cohorts biomarkers of intake mendelian randomization as i outlined at the beginning we do see a consistent direction of effect whether or not these are all statistically significant and so i think that um the data as it sits right now is pretty compelling uh that consumption overall is beneficial okay great well thank you very much tucker thank you matt there's not going to be any comments from my end here and in terms of my view on this topic uh i think the fairest approach within this episode is to let the listener uh come to their own view without my influence i would however like to close this out do you want to say something taco did you needle move at all did my needle move at all and you don't have to say which direction i'm just curious all i'll say is that i appreciate the complexity within this topic and there is a lot of nuance and i think both of you did a good job of covering that but i'll leave it there because i don't want to sway people i just i just want to comment really quick that i think simon you did a really good job as a mod i mean i don't know if tucker has any like i think yeah no you did you did an excellent job and you were an excellent opponent matt thank you very much to both of you no problem thanks tucker thank you guys uh finally i do want to just ask those who leave comments on youtube as this will be on youtube please uh i know that the whole diet was can get a little heated please refrain from any personal attacks feel free to comment on the claims or the studies but let's try and keep things respectful despite any differences that we may have and with that gents i think we did it thank you so much for your time today thank you thank you for joining me for this episode and your interest in science-based conversation i hope you enjoyed it and found the information covered interesting and instructive if you did and you'd like to show your support for the show please subscribe to our youtube channel where you can stay up to date with new episodes and watch them in video format yes the full length videos please also consider subscribing to the show on the spotify and or apple podcast app wherever you enjoy listening to podcasts you can also leave a review on apple or spotify again a great way to support the show and make our content more discoverable for others to enjoy and learn from if you have any comments about the episodes suggestions for future episodes including guests you'd like to see on the show or questions that you'd like to have answered please leave those in the comments section on youtube i myself and my team will take note of these comments when planning future episodes finally the best way to support the show and receive discounts on products we love is by checking out our sponsors at theproof.com forward slash friends enjoy your week stay well and i look forward to catching you in the next episode

Channel: The Proof with Simon Hill

Views: 61,742

Keywords: simon hill, science, nutrition, evidence, facts, diet, how to, vegan, plant based, healthy living, wellness, podcast, conversation, the proof, the proof podcast, plant proof, seed oil, healthy, debate, optimal diet, doctor

Id: QGNNsiINehI

Length: 217min 39sec (13059 seconds)

Published: Thu May 12 2022