Stay Off My Operating Table

Evan Cohen: Are US Dietary Guidelines to Blame for the Obesity Crisis? 162

• Dr. Philip Ovadia • Episode 162

Share episode

Tired of conflicting nutrition information? Join Dr. Philip Ovadia and guest Evan Cohen as they discuss groundbreaking research that challenges everything we thought we knew about nutrition advice.

This video provides clear, evidence-based insights that could transform your health. Listen for the revelation at 31:15 that exposes why popular diet advice might be making you sick.

GUEST BIO:
Evan Cohen is a Principal and the Chairman of The Brattle Group, a 600-person international economic consulting firm.  He is a litigation consultant with expertise in tax, valuation, and damages, who has provided expert testimony in various federal and state courts on issues relating to finance and economics.  He has also led a nutrition research group at Brattle since 2011, where he has co-authored several peer-reviewed articles on nutritional epidemiology, and regulatory and nutritional economics, including sin taxes, such as those on soda, and nutrition and poverty.  Most recently, he has published an article using data from the Nurses' Health Study, one of the largest and most influential longitudinal epidemiological data sets, examining the association between compliance with U.S. Government nutrition advice since the 1970s and obesity in The Journal of Nutrition (see link below). 

Most importantly, he was a elementary, middle, high and Hebrew school classmate of Dr. Philip Ovadia. 

LINKS:
Website: Evan Cohen
Study: Compliance with U.S. Government Nutrition Advice and Concurrent Obesity Trends Using Nurses' Health Study Data, 1980–2011 - ScienceDirect
Social Media Handles:
Linked In: https://www.linkedin.com/in/evan-cohen-3196605/
Twitter: @EvanKCohen

Send us a text

Joburg Meats
Keto/Carnivore-friendly meat snacks. Tasty+Clean. 4 ingredients. Use code “iFixHearts” to save 15%.

Chances are, you wouldn't be listening to this podcast if you didn't need to change your life and get healthier.

So take action right now. Book a call with Dr. Ovadia's team. 

One small step in the right direction is all it takes to get started. 

How to connect with Stay Off My Operating Table:

Twitter:

Learn more:

Theme Song : Rage Against
Written & Performed by Logan Gritton & Colin Gailey
(c) 2016 Mercury Retro Recordings

Any use of this intellectual property for text and data mining or computational analysis including as training material for artificial intelligence systems is strictly prohibited without express written consent from Dr. Philip Ovadia.

Speaker 1:

Welcome back to the Stay Off my Operating Table podcast. You can't tell it, but Phil has been on vacation for the last couple of weeks, but by the time you hear this, that will have been months ago and he'll be ready for another one. Phil, I'm glad you're back. I've missed these weekly conversations and we've got somebody today that this man occupies a category that we have never, ever, ever had a guest on occupy, um, and if that isn't a tasty setup, I don't know what is introducing.

Speaker 2:

Sure thing, cause this man actually occupies probably a couple of categories that, uh, we've never had on before.

Speaker 2:

Um One is the type of work that he does that really has nothing to do with medicine or, you know, health or nutrition directly, but he's kind of found a way to marry his day job as it is with some nutritional interest. But I think the more interesting category that he fills that you're referring to is I've actually known this guest way longer than any other guest we've ever had on this, and this is actually a childhood friend of mine and it's been quite the, I guess, sort of circle of life that we have reconnected over this nutritional interest. And here he is on my podcast, which, when we were growing up back in the day on Long Island, I don't think anyone would have quite predicted that we'd be podcasting together. Of course back then there were no podcasts so no one would have even known what that is. But here we are.

Speaker 2:

So a real honor to introduce Evan Cohen, and I'm going to let Evan give his sort of professional background and then we'll kind of get into how he got interested in nutrition and how it started to intersect, and then we'll get into a real interesting scientific piece that Evan was the lead author on, which is really what prompted bringing him on to the podcast today. So, evan, it's all yours.

Speaker 3:

Thanks, Phil Jack. Thanks, Good to be here. It's all yours. Thanks, Phil Jack. Thanks Good to be here. Yeah, and I think people would be surprised also that we would be giving any health advice, given that most of our bonding was around Entenmann's Donuts, Pop'ems and staying up all night playing poker. So now we're going to tell you about how to stay healthy.

Speaker 3:

Don't do that, Don't do that. Yeah, so I'm trained as an economist, a former investment banker. I have an MBA in finance and financial engineering from analytical skills in settings high stakes business litigation. So, you know, my firm is involved. The Brattle Group is involved in a variety of, you know, high stakes litigation where we're providing expert testimony in front of courts and tribunals all over the world.

Speaker 3:

And so, all that said, none of that is really why I'm here. The data skills that I acquired throughout all of that suddenly became applicable in a new area of life I started caring about after 2005, when I was diagnosed with celiac disease out of nowhere, and started really thinking about what I knew, or thought I knew, about nutrition and what I was eating, and you know that was 2005. By 2011, I was sort of hooked on nutrition research and understanding what we were being told and I was able to negotiate at a relatively junior level back then access to some of my colleagues to start nutrition research and start a nutrition research group. So you know that's the beginning of the background about how I got started on nutrition research and I've been doing it, you know pretty well since 2011 now.

Speaker 1:

Wow. Okay, at some point we're going to circle around and find out about picking up cans on the beach at the end of the day, but let's talk more about, let's at least start with this study that you're the lead author on. You shared with us the title of it is.

Speaker 1:

I'm scrolling back out to the top because I went down to the bottom and read the conclusions yeah, compliance with US government nutrition advice and concurrent obesity trends using nurses' health study data 1980 to 2011. I mean, the title alone just makes you, yeah, it's in the Journal of Nutrition.

Speaker 3:

Yeah.

Speaker 1:

Tell us about it.

Speaker 3:

Well, the context is interesting. You know a title like that that's descriptive. You know, nutrition research has certain ways that things are described, and this is an epidemiological journal. I'm used to writing in very different environments. We actually had to hire a former editor here to help us get this into a format, because none of us are nutrition scientists, we're economists, we're econometricians that worked on this over the years.

Speaker 1:

Oh, this is fascinating.

Speaker 3:

None of you are no, none of you are. No, you know, one of my co-authors is Dr Dennis Beer, who is an MD and former editor of a lot of you know of nutrition journals and very well published in the area. He's definitely about you know truth and using data for scientific advice, which you know seems obvious. But it makes him a bit of a revolutionary but and he was a consultant here he helped edit, but my firm and my colleagues are the ones that are all economists and that did the analytics here. Oh, that's cool.

Speaker 3:

And you know, from the beginning of when I was interested in these nutrition research, I knew I was bringing more analytical power to these types of data sets than had ever been brought. And we are able to do things with this data that nutrition scientists just don't have in their arsenal. And we didn't deploy a lot of it with this study. But the type of standards that are used for big data sets inside of economics research really need to cross disciplines. My colleagues are sometimes surprised that some of the advances that have been made in big data research just have not been brought to bear in nutrition science in the same way.

Speaker 1:

I know I'm not answering your question, so I'm going to go back about the advances in big data handling. That happens inside economics and finance. I guess modeling hasn't been applied to nutrition research.

Speaker 3:

No, and the interesting nurses health data is the gold standard inside of nutritional epidemiology for longitudinal data studies. They've tracked, you know, hundreds of thousands of you know female nurses and male health professionals inside of these data sets and collected a ton of information that's now available. The thing is you have hundreds of thousands of data points and when you talk about all of this statistics and analytics that go into this kind of big data research, I'm sure you've heard the term p-hacking. P is the p-value and it says how significant a variable is. P has a direct relationship to the number of variables in a data set. Having several hundred thousand data points or any piece of analytics makes it a different set of standards you need to apply that are kind of well known and established inside of the economics profession and you know you have to be really careful using this kind of data because the p-values themselves don't tell you all that much. When you have this much information, p-values themselves don't tell you all that much when you have this much information.

Speaker 2:

Yeah, I think that's, you know, a great point that maybe gets lost a lot in, you know, in nutritional research. So you know, one of the concerns, one of the problems, one of the criticisms about much of our nutritional research is it is based on epidemiology, it's based on these large data sets and it at least seems that oftentimes nutritional researchers will go into these data sets looking for a particular outcome, to prove a certain point. And because the data sets are so large, you can basically massage the data in such a way to come to whatever conclusion you want ultimately and it will look significant because the numbers are so large.

Speaker 2:

So maybe talk about that a little bit and maybe you can contrast that with the way you might be approaching a large financial data set where you don't want to fudge the numbers.

Speaker 3:

Ultimately, yeah, there's no fudging and I'm not accusing anyone of fudging. There's just a certain amount of care that needs to be taken in working with data sets this large. We've done a bunch of work on the side with this, with regression work. We've done some bootstrapping, where you try to strip the data down from the hundreds of thousands to get a distribution that's of a similar shape and see what happens to the p-values.

Speaker 3:

Obviously, a lot of the statistical significance of many of those studies is based on just the sheer size of the data set and there are techniques that are better deployed inside of the economics profession and better known to economists that I don't see in the readings I've done used inside of nutrition research. But that is what makes us, I believe, unique in working with this data set, because I'm not funded. You know I have a day job and you know over the years I've carved this out. My company is essentially through use of employees and our data sets and our knowledge is funding it, but they have no interest like a university might, or you know any of the funders A food company or a pharmaceutical company.

Speaker 3:

Exactly so I'm allowed to explore this data with. You know there are rules inside of getting access, so the nurse's health study had me submit proposals and they approved it and inside of that I had access to the full data set and I could do a ton of work. I also had access to what all previous researchers for the thousands of other papers that had been published. You kind of have to have your work up on their server and it's you know. Once you're in, you're in and you have access to that. So we were able to see things that were, you know, basic and important, like how to translate the food surveys in any given year into nutritional information for any given person, for any given survey of how many calories did they eat and what are their micronutrients, and down to the types of fats and molecules that they were eating. That's basic, but you could also see what other researchers did in terms of their analytics did in terms of their analytics.

Speaker 1:

Okay. And that begs the question what did you see other researchers did in terms of analytics?

Speaker 3:

Yeah, and I'm not published in that area, so I'm not going to talk a ton about what was done, except to say there's a lot more care that needs to be taken in working with big data. And, phil, like you were saying, for financial data sets or data sets inside of litigation here where I work and where my firm works is inside of high-stakes business litigation you might have a data set that's not just this big but multiples of this terabytes worth of data and you have to be really careful with how you work on it worth of data and you have to be really careful with how you work on it, that you're using an established peer-reviewed framework, that you're using the data in such a way that other economists or other experts in that area can replicate your results and that you're using all of you're using the right experts to do this analysis. You know, and even in a peer review process, it's hard to verify everything that's been done. You look at some regression results, but the data is. You know, unless you're inside of the nurse's health study, you don't have access to that.

Speaker 3:

I don't think any of my peer reviewers had access to the underlying data. They had to take my output, as you know as given and I don't know what happens with other peer review processes, but you would really have to take a look at the econometrics or the regression analysis that's done for many of these other studies to see if it could be replicated. So in a litigation context. The other side is doing that. They want to win the litigation here. In a peer review context. That's happening, but in a different way because the journal is trying to keep their scientific standards. I just don't know what journal standards are when they approach publication. If they're doing that, that same kind of rigorous analytical checking.

Speaker 2:

So you know, when you started doing the research for this article, I guess what was the original question that you guys had in mind? And maybe also talk about what was unique about the nurses' health study data that kind of attracted you to that data set as opposed to maybe some of the others that you could have used.

Speaker 3:

Yeah, maybe I'm going to go on a little bit of a tangent here and I'm going to get to your answer, but this kind of starts with some you know, readings I'm doing in like 2007, 2008. Like, the first time I approached any of this, I was reading. I got access to this cookbook called by Sally Fallon, sally Fallon Burrell, and the first you know, 75 pages of it. I'm reading and I'm like either this person is a complete crackpot or everything I know about diet and nutrition and health is is wrong, and over the years I you know she's at least a 90, 95% right in what what she was saying. But it kind of prompted me to do other reading and people. That leads you to people like Gary Taubes and good calories, bad calories and a lot of other articles. So I started, you know, in 2005 with the mindset that I should be eating carbohydrates and low fat milk. And you know, a bowl of Lucky Charms is with. Low fat milk is much healthier than a, than a fatty steak. That's how I went into it Once I got celiac disease and I was restricted from eating, you know, gluten, which is wheat, barley and rye, or at least the protein from it.

Speaker 3:

Um, I started eating actually a lot of other carbohydrates. You know I'd see something gluten-free and be like, oh, it must be good for me, it's gluten-free. And you can see like, if you look at pictures of me 15, 20 years ago, I'm different than I am now. I'm thinner now Like I was 20 pounds heavier, I was doughy, I was inflamed. I used to wear wristbands to work, because both of my, you know, they thought I might have carpal tunnel, but it ends up I was just inflamed all over. You know so how my personal health and dietary journey went and how that lined up with what I was doing. I wanted to see well, are we finding in the data that people are really eating a lot more, you know, fat and a lot more saturated fat, and that's what's causing all of this obesity and many of the other health things that I became interested in.

Speaker 3:

So in 2011, we went looking for data sets and immediately was like, okay, a bunch of economists rolling around for data set, this is exactly what they should be doing. So I found the first data set I found was NHANES, which is a government sponsored survey that was done every you know periodically through the 70s and 80s and is done a little bit more regularly Now. It only had about 5,000 people they weren't followed longitudinally in any given year but we put together what I believe at the time to be the first continuous series and that was an achievement. It took us two years, not of constant work, but I had interns and I had junior people, and I was working with that. Trying to understand each one of the surveys required separate skills of correcting the data and understanding who did they overweight this year and why. And then, what are we telling?

Speaker 3:

Through all of this, we were able to, you know, put together a trends paper and in 2015, with the work we'd done there, we put together these trends that showed that people were eating a ton more carbohydrates. They were, you know, on average much more obese. And, importantly, we looked at the distribution. Whereas in the early 70s you had this distribution of BMI where people had much lower BMI, but also there wasn't a ton of obesity or severe obesity what happened in the ensuing years and I have a similar graph in the nurse's health paper because it shows, you know, with all the differences between the data sets, it shows basically the same thing that early on you had a certain distribution for BMI, but through whatever happened in the last 30 years, between 1980 and 2010, the distribution changed completely and people weren't just more obese on average, severe obesity really increased by hundreds of percent, you know. And obesity, so it was this distribution changing. That was also just remarkable.

Speaker 3:

And then I did a few other things with other data sets and other journals, but eventually I was looking for other data sets that I could do better and more analysis on.

Speaker 3:

I looked at Framingham. I think it had, you know, there just weren't enough people in the Framingham Heart Study to do what I wanted to do and show these trends. And eventually, you know, take a look at some of the other associations because I agree with you, working with an epidemiological data set, the best I can do as a researcher is point out associations. But eventually got to the nurse's health study and I started, you know, figuring out how I could apply and we, we put in a proposal, we went back and forth and they, they approved us to do research in obesity, you know, and I gave them a general. I kind of wanted to do what I had been doing of like here are some trends, here are trends of obesity and let's see if we could find the same thing with the nurse's health study, or, you know, a completely different group of people with a lot of different data and variables, all longitudinal.

Speaker 1:

Real quick. I let this pass For our non-statisticians and our non-medical people. Define longitudinal. What is a longitudinal study?

Speaker 3:

and they track them. They track them until they drop out. You either drop out because you stop responding, which not that many people do, or you die. And so there's still Nurses. Health 1 is still ongoing. They're still getting data out of it.

Speaker 1:

Okay, so longitudinal study is one that's tracking a set of people over a long period of time.

Speaker 3:

As opposed to NHANES, which is not longitudinal, and each year they do it, they get a new sample.

Speaker 1:

Okay, sorry, just wanted to make sure that was clear for our people.

Speaker 2:

And the other thing we should throw in there to make clear about a longitudinal study is you're not really intervening in any way in these patients, in these study participants' life. You're just tracking what they're doing and then what measures of health or outcomes they're achieving, and that's what Nurses Health did.

Speaker 3:

They had these nurses in for checkups and they weren't just filling out surveys and they weren't just doing about nutrition. They were tracking their blood and they were, you know, as time went on and blood work became more sophisticated, they were able to track more and more. They were tracking their diseases and their medical history and they were putting this all together with, like health and nutrition information, dietary exercise, smoking. At the end of the day, I think there's a lot more they could have tracked and that would have made a much more robust data set for analysis, where a lot of things could have been controlled. For, for example, you can't control for income in this data set. They just don't track it. I've seen a lot of studies that can't control for income in this data set. They just don't track it. I've seen a lot of studies that say they control for income. They don't. You know, in one year they asked about income and it was early. Early in the nurses health one study they asked how much money does your husband make? Essentially because everybody in nurses health one was a married female nurse. Um, that's it. But otherwise, you're a nurse, you're working as a nurse in the US. There's not a ton of income range.

Speaker 3:

What we found with the Anne Haines that was really interesting was a lot of what happens in nutrition happens at two times the poverty line and below. What I found absolutely striking and after we published the NHANES study, we edited a journal of the Economist's Voice and we solicited articles on nutrition and poverty, because the number one indicator of your health outcome in the United States, according to the NHANES, was how much money you made, like your income, and specifically NHANES stopped tracking. Once you're at $75,000 per year you were considered rich, but two times the poverty line and below is where all the action happened in NHANES and many other studies. Poverty in America is the number one determinant of your health outcomes and specifically your nutrition, nutrition access. It's just massively important and I think, even understudied, if that's possible.

Speaker 1:

All right.

Speaker 3:

Yeah, I'm saying, nurses help.

Speaker 2:

Take us home. What did we find from this study so?

Speaker 3:

what we eventually were able to put together, and was we really? You know, we submitted a paper that was similar to the 2015 paper that track trends and trends in what people are eating and trends in obesity or BMI, and the journal was basically like you could do better than this, and we went and we rewrote it, and what we did was we put together a variable that was a proxy for complying with US nutritional advice. So, before the nutrition guidelines came out in 1980, the Senate put together some hearings that were really led by George McGovern, the 1972 presidential candidate and a pretty influential senator at the time, started a set of hearings that ultimately led to the dietary guidelines. And so, from 1977, you could see what the government was saying people should eat in terms of carbohydrates and fat restrictions and salt and cholesterol. There were only so many things that were relatively consistent, and those were the amount of fat was relatively consistent from the 1977 until my study access ended in about 2011,.

Speaker 3:

Saturated fat and cholesterol. So these were things that were more or less the same throughout the period. Keep your fat total fat consumption under 30% of your total calories, keep your saturated fat under 10% of your total calories and keep your cholesterol under 300 milligrams. They vary here and there. There are other things like salt or sugar that are in there. They're just not consistent and they're much more difficult to track. So what we did was we took compliance with those three variables as a proxy for your overall compliance and tried to see if compliance with government advice was associated with better outcomes with obesity.

Speaker 1:

What were the outcomes for following the government's guidelines?

Speaker 3:

You know people were incredibly compliant. You know people say that they didn't follow. You had four percent joint compliance for those three variables in 1980 when the guidelines first came out. It went to almost 50% in the late nineties. Late nineties is kind of the peak in this kind of nutritional data work for uh, maximum carbs consumption, the lowest fat consumption and also the highest compliance as it would fall out with um, the the U S nutritional guidelines or guidance. In the late 90s almost 50% of people were complying with that and a lot of that was driven by their compliance with keeping their fat consumption under 30% of their total calories per day.

Speaker 1:

Now saturated fat, which is oh no, you said saturated fat was 10%.

Speaker 3:

Yeah, saturated fat is 10% and total fat is 30%. You know they've moved off this a little bit. Now Cholesterol ends up not being super interesting because the compliance rates were through the roof, you know, and they just kept going up through the period. People really take this cholesterol advice seriously and have largely complied with it, and Phil could tell you about the outcomes on heart disease that you know how much that's prevented heart disease.

Speaker 3:

But what we did in this study was we then looked at obesity because that's what we were, you know.

Speaker 3:

That's what we proposed to study and we looked at the association between compliance and obesity.

Speaker 3:

States people were, like I said before, the average BMI moved up several BMI points and when we're talking about the distribution, a lot more people were severely or morbidly obese and obese as a percentage of the total population.

Speaker 3:

So the overall time trend dominated everything and what we found was there was a very small benefit to compliance that was associated with slightly lower BMI outcomes, but inside of that was all the other behavior that's tied with compliance and that we didn't even attempt to suss out, which is physical activity and smoking and a ton of other things that are probably associated with people that are complying with government nutrition advice are also probably doing a number of other things. You know, once you say I'm not following the advice, you're much more likely to not be exercising and to be smoking, and so that in that small amount of statistical significance is buried a ton of other things. And so what we conclude in this paper is that there's little, if any, effect to complying with the US government nutrition advice on fat consumption, saturated fat and cholesterol as something that would be associated with lower BMI or really lower BMI increases than the general population.

Speaker 2:

So just to summarize, basically, everyone got more obese over time, but there really was no difference in you know how much your obesity increased whether or not you followed these guidelines or didn't follow these guidelines.

Speaker 3:

That's exactly yeah, and we're talking about just association. There's no correlation, there's no causation. That that was tested here. That's, that's a totally. That's another set of of research that could or should be done off of this. But yeah, just complying with the dietary guidelines didn't protect you against obesity, and if it did, it was so minuscule that it's a statistical blip.

Speaker 1:

So let me just quote from the very first line of the abstract blip. So let me just quote from the very first line of the abstract. Beginning in 1977, the US government began formally issuing dietary advice, a main objective of which was to reduce and prevent the prevalence of obesity in the American population. Yeah, and that's straight from this. Advice was designed. A main objective was to reduce and prevent obesity, and the outcome of half the cohort complying was identical to the outcome of the cohort not complying. Am I getting?

Speaker 3:

that right. That's not identical, but you know we're talking about the time.

Speaker 1:

You're talking as an economist, I'm talking as a dude.

Speaker 3:

Well then, as a dude, you got it totally right, okay, okay, yeah, and. I, you know it's a lot of what happens in nutrition is we told you to do this and you didn't do it. That's why you had a bad outcome here. That's not. That's not what we're seeing. You know, people, for the most part, even if you didn't comply technically, most people just most people move down, they, they, they switched their fat consumption and their saturated fat to carbohydrates, right, and that peaked in the late 90s.

Speaker 2:

but the health effects of that were felt years later, you know if you spend 20 years eating a certain way, you're going to be affected by that for the rest of your life. So, yeah, yeah, and I think it's a great point you know that one of the most common excuses a high percentage, you know, did comply and doing so had no effect is really, I think, the other important part of this, because you know we could certainly say, ok, these were nurses, so they're probably more likely to comply with the dietary guidelines. Right, because they should be, at least theoretically. This doesn't necessarily translate to practice, but theoretically they should be interested in health and you know they're being educated on this advice because they have to pass this advice on to their patients to help them from a nutritional standpoint. And so you'd think they would be more likely to comply. And you know their compliance probably is higher than the general population, and yet from that compliance we see that there's no effect.

Speaker 3:

I will say that we did compare demographically the nurses health demographic with the Anne Haynes cohort and there's a slight difference. I think it's related to income. You know that, but I don't know that that what you said is true. Like you would think, nurses would be more likely to comply, but I don't know that that to be true and when we look at the overall trends across these two studies that were done in completely different ways, they're showing the same things at the same time, especially when you control for the demographics. So it's sort of this is something that happened to our whole population, and you know.

Speaker 3:

My question is if you're a policymaker or a researcher and you have the best of intentions, you have to look at what's happening right and what's happened over the last 40, 45 years, since you know we've been in the giving nutrition advice business.

Speaker 3:

You're presiding over a public health crisis. That's happened in slow motion, but it's happening right now. Something should be done differently and you know I can talk about it from the research perspective. You know, if I went back to the beginning of my career and I became a nutritional epidemiologist, what would I have done differently? Using what I know today, using the standards from economics and big data and using, you know, a scientific approach where I'm not accusing them of not using a scientific approach but you have to look at the outcomes and say you know if you're depending on p-values in these papers to give dietary advice that confirm the same things over and over again that we know people are more or less doing and aren't working, you have to think what are other ways to approach this type of research where you can have a science-based and evidence-based set of outcomes that you could be giving a broader set of advice set of outcomes that you could be giving a broader set of advice.

Speaker 1:

Could we back up just a little bit? Um, p hacking just because I'm kind of nerdy about stuff like this, I've, I've, I'm aware of what that word means, um, and half of our audience totally gets it and the other half is just kind of dudes who are slightly less nerdy than me. So explain what happens with p hacking. Explain what's happening when the results of the study are accused of p hacking yeah, and I I will.

Speaker 3:

I'm not accusing anyone of p hacking, but the way it works.

Speaker 3:

We know this happens yeah, and the way it works is that you're generally taking advantage of statistical tools. The p value of any particular value variable is, you know, you're taking a sample in a study, um, and as your sample size gets so big that is basically the population the the error term on any of these variables, because it's related to the number of observations. The more observations you have, all else equal, the less your error is going to be. And so by using a big data set and not doing something like bootstrapping, where you're taking hundreds of thousands of data points and getting it down to like 10,000, which is still relatively big, um, by using the full data set and not doing any other kind of corrections, you're going to get, um, these P values that say this variable is like significant to the 99.9 percentile right. And in this kind of statistical work that means that the variable, just from a p-value perspective, is unapproachable, like that, 999 times out of000,. This variable would basically be different than zero.

Speaker 3:

You know the number you're finding. It may not be that exact number, but that there is something you've observed that you're now able to report on, and that's why we get a lot of. You know the advice that goes back and forth like the egg is good, the egg is bad, just eat the whites. Like you get these, these observational studies that are really, you know the researchers should be pointing out these are associations. And then you get the mainstream media that takes that and says, oh, we got this study that came out that said there's this statistical significance and don't eat eggs anymore out that said, there's this statistical significance, and don't eat eggs anymore.

Speaker 2:

Now, along those lines, are you aware of other data, other studies that have looked at the dietary guidelines and shown that they have had benefit regarding obesity?

Speaker 3:

No, you know a lot, I haven't. No, I haven't seen that, and you know it's what studies are getting funded and and who's doing it. It's just, I think, for a lot of researchers that particular study Isn't that interesting.

Speaker 2:

Well, yeah, I mean, you know, it does strike me that that's not a question that a lot of people have chosen to look at. Because you would think, you know, here we are 40 plus years into the dietary guidelines and you know they had their objectives, one of which was, you know, combating at that time what was still a relatively low but increasing incidence of obesity. And you know, here we are 40 plus years later. It's pretty obvious that obesity hasn't gotten better and has gotten, you know, immeasurably worse. You know multiple Xs worse.

Speaker 2:

And you would think that some researcher would want to say, hey, you know, are people following the guidelines and does it help with obesity? And I would also hope, now that we have this data, you know that there would be some of these stakeholders will say, paying attention to something like this. Now, I know that your story, your article study, was not featured on any evening news programs that I'm aware of and, to be honest, the only reason that I'm aware that your study happened was because we were friends and we had been conversing along the line, as you were doing some of it. But it you know, I don't hear my colleagues in the hallways of the hospitals talking about this study that basically shows that the US dietary guidelines are useless.

Speaker 3:

Yeah, and you know I'm not. I'm not amplifying it, it's not anything I'm marketing. It's something I was interested in and I wanted to put out there. You know this is. The disadvantage of being an unfunded group is that this is not my day job, so I'm super interested in it. I love working with the Nurses Health study data. It's interesting. With the nurses health study data, it's it's. It's interesting. It's the richest type of study we have. For you know, whatever shortcomings it has in terms of, you know, collecting collecting, uh, the type of data that were collected it's still the most robust data set you have to work with and the disease outcomes are fascinating.

Speaker 3:

Like I'd love to publish on on heart disease. I just need a good cardiac doctor to publish with. You might know a guy. Yeah, I don't know. You know, yeah, I mean, that's the way things work. I'm not Walter Willett, I'm not at the Harvard school of public health. I'm just not amplifying this.

Speaker 3:

And as interested as I am, I'm just not amplifying this and as interested as I am, you know, for me it doesn't change much in my personal life. I've totally changed the way I've eaten and thought about food over the last, you know, 20 years and you know, I don't know that what I do is scalable, because I do feel probably a lot of what you do, of eating a lot of meat and sourcing my food at local farms and having access to a rich variety. I just, you know, industrialized agriculture is designed to feed giant populations and until we figure out a better way to feed people with better methods you methods we're going to have a big, big problem. I have a really good example. I have a friend from college who he's 50. We're all 50. Phil turns 50 next month and he's winning his division in races.

Speaker 3:

This guy's in incredible shape. He hasn't missed a day running in 20-something years and he's a vegan right, and his diet is really thoughtful. He works for a food company. We couldn't be more different in how we approach food, except that the one thing we agree on is industrialized agriculture. Like we both stay away from it, and I don't know that there's an exact right answer. I know what works for me, but for him it's something completely different and that's the problem I'd love to see solved. And it just gets our food system, our pharmaceutical and health care system. Our policymaking isn't designed. It's designed to feed a lot of people and Americans will just they eat what's affordable and it's led to these type of health outcomes. So I'd love to see that system change. I just don't see it happening in our lifetime, not to be depressing.

Speaker 2:

Let's actually talk about your personal story a little bit and honestly, let's talk about our personal stories because I think there are some very interesting things that we can pull out of it. So you know we grow up together very good friends, eating, you know, probably largely the same food and spending a lot of time together. You know I would say our activity levels were probably pretty well matched during childhood and you know I end up significantly obese. You never really had a weight problem. But you end up with celiac disease and, lo and behold, you know after a number of years that we really weren't in much contact. We reconnect because we had both stumbled upon the same solution to our different problems me for obesity, you for celiac disease and we both end up eating a low carbohydrate. You know carnivore type diet, independent of each other.

Speaker 2:

And you know the way that Evan and I sort of reconnected was I was at a low carb conference and I tweeted out something or social media or something about you know being at the low carb conference with I think it was Adele Height or maybe it was Nina, I don't remember exactly and all of a sudden I get a text message from my, you know, kind of long lost friend saying you're with Adele Height. You know I wrote a paper with her.

Speaker 3:

I think it was right and Adele was a co author of my 2015 paper.

Speaker 2:

Yeah, and so that's how we sort of re-engaged and reconnected and came to found that. You know, like I said, we had two very different problems that came from growing up the same way and we found this solution. So talk about you know how you got to low carb from your health problems?

Speaker 3:

Yeah, you know how you got to low carb from your health problems. Yeah, and you know I, I, I was not obese, but I was. I weighed like 25 pounds more than I do right now at my height and I, I was just. I was uncomfortable, I was inflamed. It was more than celiac, like I was, I had, I had acid reflux and I, the type of foods I could eat was getting smaller and smaller. You know the type of foods I could eat and still be comfortable. And that's why I got curious about the food. The celiac really sparked it.

Speaker 3:

When I was diagnosed with celiac disease in 2005, I'd never heard of celiac disease. I think I'd seen the word gluten a few times in the Whole Food store, which I didn't shop in all that often, and I thought it was in one of those fringe sections with all the supplements. And then you learn, and I took my doctor's advice of eat a lot of fiber and stay away from celiac grains. And you know a bunch of things. I don't. I stay away from celiac grains, but I don't follow most of that advice anymore because it's population level advice that comes from studies rather than saying what works for me and the books I read, the authors that I came to my own research, like we arrived at the same place, because we, we stopped following what, what we were told, and started saying what? What are people who are smarter than we are? What do they say? What are they looking at? How have we made these types of policy conclusions and societal investments? Conclusions and societal investments? And I stopped being interested in what doctors told me to do and what policymakers told me to do and started, you know, experimenting of what foods make me feel good. You know what foods make me feel energized, and you know I'm.

Speaker 3:

I don't take any pharmaceuticals anymore. I took reflux medication for 15 years. I don't wear the, the. I have zero inflammation markers, um, and you know it's working. Like it feels good. I feel full, you know, with a breakfast of, you know, eggs and cheese and and meats. Then I feel I feel good throughout the day. A lot of times I don't even eat lunch, you know. So I, I stumbled on it through trial and error and through a ton of research and a ton of problems. I, I, I have GI doctors that know me by name because of my journey over the last 20 years and I just I don't see them as often, you know, I think they miss me, I guess.

Speaker 1:

All right, phil, as somebody who's a cardiac surgeon, dealing with people who you know, I see this tweet from you now and then where you say something like 50% of the people I operate on have normal cholesterol. What's what is? I know what your reaction is, but what is the reaction you would expect from your colleagues? I don't want to put you on this on the spot, but let's make a difference. What's the reaction you would?

Speaker 2:

expect.

Speaker 2:

What I uh expect of my colleagues and I'm increasingly vocal about telling them this uh, you know, whether it's in person or kind of, you know, broadcasting on social media is we start to change our ways.

Speaker 2:

Because, you know, I look at a paper like this and say, okay, if the US dietary guidelines were such a failure in, you know, meeting this objective regarding obesity, you know what else have they been a failure in?

Speaker 2:

And you know, to me, heart disease is another obvious thing, and you know, yeah, it would be great to have similar data, but the honest truth is is I don't need a high level analysis like this. I can just look around every day at the patients I'm operating on and a lot of them are following the guidelines. I ask them what they eat now, because you know that's an interest of mine and I, you know a high percentage are eating the low fat diet and they're, you know, using the margarine instead of the butter, and they're drinking the skim milk, and they're, you know, they are following the guidelines, and yet here they are talking to a cardiac surgeon to deal with their heart disease. So I hope my colleagues wake up to this and start saying maybe we've been giving lousy advice instead of falling back on presumption that people aren't following the advice, and I think this paper is a good. You know some good data we can use to support that ultimately.

Speaker 1:

Oh yeah.

Speaker 3:

Can I ask a follow up to that? Because I once I broke my orbital bone and I was going skiing and I said to the doctor, can I go skiing? And he's like, no, you definitely should not go skiing this weekend. And I'm like, look, doc, if I go skiing and have a good time, I'm not writing you a thank you note. But if you tell me I go skiing and I, you know, I damage myself further, I'm probably going to sue you and he you know he's he told me that he would give the same advice to his family member, advice to his family member. But I think you have a basic like problem here, which is your colleagues, you know, would be going against nutrition advice and the state of the research and the literature and they're totally exposed. Why would they ever do?

Speaker 2:

that, yeah, no, it's a very valid point and ultimately the reason why we would do that is because we're doing the right thing by our patient.

Speaker 2:

And you know, I know that is challenging for my colleagues really don't worry about it anymore and actually the way that I look at it is, I am quite confident that if it came to be, I could defend in a court of law that what I am doing is supported by scientific research and is, you know, within the standards.

Speaker 2:

Right, and you know, and unfortunately there have been a few physicians that have had to, you know, prove this in a court of law. You know we have people like Tim Noakes and Gary Fetke who have gone through it and they came out successful. But you know, ultimately, again, you know, our obligation as physicians is not to stay within these guidelines that have been set by politicians, by even our medical societies, but our obligation is to our patients and if we see, over and over, the advice that has been given is failing our patients, we need to start doing differently. And that's the way that I look at it and I understand that that can challenge, can be a challenge on many levels for physicians, but we have to do better by our patients ultimately positions, but we have to do better by our patience. Ultimately and you know the way that I look at it is, we are failing our patience, we are failing society because you know, this is continuing to get worse in front of our eyes and we need to do something different different.

Speaker 1:

Wow, evan, what month was this journal published? This is a 2024 edition, but I don't see. Yeah, I think it's April. Okay, so this report has only been out about three months. We will make sure it is linked in the show notes because we've got lots of listeners who will say I want to see that for myself and we'll get a little better distribution of it than what's happened otherwise.

Speaker 2:

Thanks, Thanks.

Speaker 1:

Phil, I did not. I mean, I I reading the the report prior to evan coming on, I was, uh, I was kind of blown away, but this is an even bigger deal than I realized. Um, this is really really significant. Um, evan, thank you so much for agreeing to do this. I think this is. I think this may prove to be a pivotal moment. We've got there. We've got overwhelming data, studied by somebody who knows how to work with gigantic data sets and who wasn't attempting to prove something.

Speaker 3:

Following the numbers.

Speaker 1:

Kind of seems to speak for itself Any last words.

Speaker 2:

No, I mean, you know it just honestly you know. Something else I'm thankful for, I guess you know along this whole journey, is that you know there are longtime friends, longtime, you know they're longtime friends, longtime acquaintances that have, you know, unknowing, initially unknowing you know, un, low-carb, nutritional approaches are having on people that I've known for a long time. So it's great to talk to everyone, great to have them on the podcast like this Again, not something I ever imagined would happen you know, and it's pretty cool.

Speaker 1:

You know, and it's pretty cool, evan. Keep us, just keep us posted as you continue to play around with this data. We want to. We want to know what's going on, and it's obvious that it's this kind of research that we all wish would be happening Actual, scientific approach, disinterested, just looking for what's the data say.

Speaker 3:

All right, can do. Thanks for having me on, guys. Good to talk to you for this hour. Happy to keep you up to speed with what I'm doing.

Speaker 1:

You know I have Phil's phone number so I could text him. All right, um, evan, if people want to want to learn more about you, I trust that there's ways to do that as well, yeah. I'm not sure that shows up at the show notes. Great.

Speaker 3:

Thanks.

Speaker 1:

All right For Dr Philip Ovadia and Evan Cohen. This has been the Stay Off my Operating Table podcast. We will talk to you next time.

People on this episode

Head Shot

Dr. Philip Ovadia

Host
Head Shot

Jack Heald

Host