Meat Controversy and Why Nutrition Research Is Hard - Michael Hull

Guest

On this episode of humanOS Radio, Dan talks to Michael Hull. Michael has an MSc in human nutrition from McGill University, and works as a full-time senior researcher at Examine.com, the largest database of nutrition and supplement research on the internet. Mike writes and updates the Supplement Guides, maintains the company’s vast database of supplement studies, and blogs about various topics in the realm of health (yep, sounds like our kind of guy).

You might remember that back in October, a series of studies were published that addressed the impact of red and processed meat consumption on a number of health outcomes. Importantly, these papers did not present any new evidence on the subject. Instead, they summarized the findings of existing RCTs and observational studies, using the GRADE certainty of evidence score to evaluate research quality. The group that performed these reviews concluded that adults should continue consuming red and processed meat at current levels of intake – an obvious contradiction of most established guidelines.

This, naturally, elicited a lot of turbulence online, from all across the diet spectrum.

So, are they right?

Mike wrote an excellent piece for Examine.com sorting out these studies, and was kind enough to come on the show to discuss the papers and their implications.

As you’ll see, this is arguably less a discussion of the health effects of meat per se, and more about the aforementioned difficulties in performing nutrition studies and how we go about interpreting research.

To learn more, check out the interview below!

Example of laying out lifestyle evidence in a helpful way: World Cancer Research Fund – Interactive Risk Matrix

(This is only half of the chart. Click on the link to see the entire matrix. Really cool graphic!)

Listen Here

Youtube

Support

Have you considered becoming a Pro member of humanOS.me? It costs just $9.99 per month, and when you go Pro, you get access to all our courses, tools, recipes, and workouts. Pro members also support our work on blogs and podcasts, so thanks!

Leave a Review

If you think other people would benefit from listening to this show, you can help us spread the word by leaving a review at iTunes. Positive reviews really help raise the profile of our show!

Transcript

Mike Hull - 00:01: Does red meat intake matter? Well, compared to what else? Let's look at your genetics. Let's look at your overall dietary pattern. Let's look at your exercise and your vitamin and mineral intake and your overall calorie intake. I don't want people to so myopically focus on single food items. They're missing the host of other factors that can greatly affect their quality of life.

Dan Pardi - 00:22: It has become very popular to think about and approach nutrition almost like medicine. In fact, traditionally, food was medicine, quite literally, and as we learn more about the functional properties of certain compounds in food that we eat, like flavonoids - that's just one example - it has become easy to think of dietary exposures as being similar to a drug.

00:44: In theory, we should be able to understand both by similar scientific evaluation methods, but when we look at attempts by researchers to elucidate the effects of diet on health outcomes, it becomes obvious that food is different, and nutrition science has some unique challenges.

01:01: You probably already know that when scientists want to test a pharmaceutical intervention for safety and efficacy, the gold standard is a randomized controlled clinical trial. Participants are randomly assigned to either a substance being tested or a placebo or usual standard of care and are not told which they are taking. In this way, we have a greater certainty that any significant difference in outcomes is due to the experimental treatment, because the groups are otherwise similar to one another.

01:31: This is much more difficult to achieve in studies of nutrition, for many obvious reasons. For one thing, it is surprisingly difficult to accurately measure what people are eating, especially outside of the lab setting. Blinding is often impossible for studies of food, because people can see and taste what they are eating. Compliance is frequently woefully low for dietary interventions compared to drug trials, often rendering the results suspect in meaning.

01:58: The most rigorous trials in metabolic wards are extremely expensive and are often limited in duration and sample size. I could go on. Due to ethical, financial, and practical limitations, nutrition science is very complicated and has historically relied upon combination of observational research, particularly prospective cohort studies, randomized controlled clinical trials, and mechanistic studies.

02:25: Nutrition is messy, and it's difficult to assess the quality of information we have at our disposal. That's why I am pleased to have Michael Hull on the podcast today. Mike has a master's degree in human nutrition from McGill University. He is a full-time senior researcher at examine.com, where he blogs about various health topics.

02:46: Examine is the largest database of nutrition and supplement research on the Internet. They provide fully referenced evidence summaries on a vast array of health-related questions. It's definitely worth checking out if you haven't visited that site already.

03:02: You may have heard a lot about a series of studies that came out recently suggesting that adults should continue consuming red or processed meat at current levels of intake, contradicting other guidelines. Mike wrote a great piece for examined.com assessing the content of these studies. As you'll see, this is perhaps less a discussion of the health effects of meat and more about the aforementioned difficulties in nutrition science and dietary recommendations. So, without further ado, Mike, welcome to the show.

Mike Hull - 03:33: Thank you. Thanks for having me on.

Dan Pardi - 03:35: Tell us a little bit more about your background and your work at Examine.

Mike Hull - 03:39: Yeah, so I did my undergraduate degree at George Mason University, where I studied exercise science, and then I did my minor in nutrition. During that period, I became very fascinated with nutrition. I began working at Examine as an intern, and I recently graduated from McGill University, where I did my master's thesis in human nutrition.

03:58: This is about ... I think I'm coming up on my fifth year with Examine, and I mainly work on their supplement guides, which are these easy-to-read guides to give you some very simple step-by-step directions on which supplements to take and what not to waste money on for specific health goals, like if you wanted to support a healthy testosterone level, if you want to improve your sleep quality, if you want to look at things that are good for muscle building, et cetera. So that is where I spend most of my time, and then I'll do an occasional blog post every now and again for the main website, which my most recent one was this red meat article.

Dan Pardi - 04:31: From a high level, what distinguishes Examine from other places that might have an opinion about supplements or food? Give us an overview about your process. That's what's so impressive about what you guys do.

Mike Hull - 04:44: Yeah, so we are really keen on evidence summaries and communication. So at the core of our website is our ethos that we're not here to tell you what to do. We simply want to give you high-quality evidence or summaries of the evidence to tell you what it says or what it doesn't say so that you can then go take that information, consider the other factors going on in your life, and then make a decision about what you would like to do for your circumstances.

05:11: I think that's one thing that kind of sets us apart, and not unique to Examine, but it's something that we strive to communicate whenever we write articles or we are doing evidence summaries on supplements. We really don't want to tell you what to do. We just want to give you good information to empower you to make your own decisions.

Dan Pardi - 05:28: It really is an excellent resource, so thank you for your work there. Let's move on to the controversy of the day. Give us a little bit of a background on this controversy. So what studies were published, and what set them apart? Where was this even published and when?

Mike Hull - 05:44: Yeah, so this was published a couple of months ago, pretty recently, beginning of October, I believe. It was published in the Annals of Internal Medicine, and they published five reviews from the Nutritional Recommendations International Consortium, which goes by the NutriRECS for short.

05:59: So they're looking at the impact of red and processed meat consumption on a whole host of health outcomes. They looked at cancer mortality, cancer risk, diabetes, heart disease, et cetera. So this really went in and tried to summarize all of the best available human trial evidence that we have on these given topics.

Dan Pardi - 06:20: One method that was utilized in this evaluation by this research group was something called GRADE. Tell us about what GRADE is and what it does.

Mike Hull - 06:30: Yeah, so GRADE stands for the Grading of Recommendations, Assessment, Development and Evaluation, and it's a tool that is used for evaluating research quality and the strength of the recommendations that might stem from that research quality.

06:46: The part of the GRADE system that got tossed around a lot in articles was the GRADE Certainty of Evidence Score, which has four possible levels. It's got high, moderate, low, or very low. So this score is designed to tell you the likelihood that the effect that was seen in the study, be they positive, neutral, or negative, will be substantially different from the real world.

07:06: If their rating is high, that means that this research provides a good indication that the effect is likely and that the likelihood of the effect will be substantially different in the real world is very low. On the flip side of that, if the grade is very low, then the researchers are saying, "We think that the effect that we're seeing in this research study is going to be substantially different from what we're seeing in the real world."

Dan Pardi - 07:29: So is GRADE an appropriate tool for our nutrition research, or would it make more sense to use a different rating system for studies dealing with diet and lifestyle? Are there alternatives?

Mike Hull - 07:40: Yeah, this was a major sticking point in this whole debate between both the NutriRECS team that put this out and other researchers that were saying, "Hey, because of the limitations inherent to nutrition research, such as not being able to run a randomized controlled trial for red meat for 60 years to see if anything comes of it, that we should be applying different standards to assessing that evidence."

08:03: Dr. Gordon Guyatt is one of the lead authors in the five studies that were published. His argument is, "Why should you have one set of rules for judging nutrition and another set of rules judging some areas, such as drug interventions?"

08:16: One of the critiques of the GRADE guidelines is, "Hey, they were originally developed for drug interventions, and we don't think that is wholly appropriate for nutrition interventions, which come with their own set of limitations."

08:29: There have been alternatives proposed in order to assess nutritional evidence specifically. My take on those is that they're really ... They're saying, "Hey, what if we concede the point that we're not going to be able to make these very long term RCTs that really get at the questions that we're trying to answer, because they're just infeasible or sometimes unethical? So if we're conceding that point, let's develop a framework that we can assess nutrition studies, given that concession."

08:59: It's really just a debate between is GRADE this great tool for assessing nutrition research, or should we come up with some other developments that are geared toward helping us create guidelines, if we can see that very high-quality research is very difficult or unlikely within the nutrition space?

Dan Pardi 09:15: The review of randomized trials only includes one trial, the Women's Health Initiative, but it wasn't a trial that directly tested red meat restriction, and the dietary modification that they were able to achieve in this group was extremely small. We see that, of course, in many different nutrition studies. Thus, the magnitude of effect was also very small. Then PREDIMED and Leone, which are two famous dietary trials that are often used to promote a Mediterranean style of eating, those two other large trials were excluded. Why do you suppose that is, that only that one trial, the Women's Health Initiative, was included and then PREDIMED and Leone were excluded?

Mike Hull - 09:56: Yeah, so I believe that Leone was excluded because I believe the author stated that it was implausibly large effect sizes. They were saying, essentially, that they're suspect of the large effect size as seen, given the design of the study. I'm not sure what the reason they gave was for the Mediterranean diet study, but even if you did include those studies, I don't think it would have hugely swayed the outcomes that they came out. It may have bumped up GRADE a little bit.

10:23: Right now, for that meta analysis, they rated the three major outcomes, all-cause mortality, cardiovascular mortality, and cancer mortality, as low or very low. It might've bumped it up from very low to low. The confidence intervals might have been shifted a little bit. But I think, fundamentally, it wouldn't have shifted the results in such a huge manner that would have changed the outcomes that much.

10:47: I do understand the criticism from the other side of this. They're saying, "Well, why not even analyze it to see what happens, even if we can see that maybe these two studies just weren't the best well-designed to assess for red meat outcome?" I understand that, too, and it would have been nice to see the whole analysis just so we have it, but it is what it is.

Dan Pardi - 11:04: Would any nutrition research fall outside of very low or low according to the GRADE scale, simply due to the limitations of nutritional research?

Mike Hull - 11:15: I mean, if we're talking about dietary interventions, I think it is possible. I'm not intimately familiar with GRADE enough to be able to concretely answer that, because I know that there are ... It's its own whole separate beast to contend with. So, yeah, I would say it's possible, at least if we're talking about dietary interventions. It's more likely when you're talking about supplement interventions, because those more closely mimic a drug intervention, for example, where you can actually achieve proper blinding. It is possible, just depending on what area of nutrition you're looking at. But dietary interventions, maybe not so much.

Dan Pardi - 11:48: Yeah. So anything that had a large enough effect size would potentially show better on the GRADE scale. But if you're looking at differences in terms of single types of foods within a context of a whole diet, it just might not have that resolution, perhaps.

Mike Hull - 12:02: Yeah. For example, you're never going to be able to achieve proper blinding in a study like that.

Dan Pardi - 12:07: Yeah.

Mike Hull - 12:07: So that's always going to be an inherent limitation to that.

Dan Pardi - 12:10: I mean, this seems like it's less about meat consumption, per se, and more about research design and how we infer cause and effect from research. So, in theory, randomized controlled trials would be ideal, but nutrition science still relies heavily on observational research. So why is that, and what challenges make nutrition unique, in that respect?

Mike Hull - 12:29: Well, one of them is just the cost of running clinical trials, even if they are only single-blinded. For example, you can't blind the participants to what diet intervention they're receiving, but you could blind the people analyzing the data, for example. So you can have a single-blind trial, but those trials are very expensive. They're not expensive compared to a drug trial, which might be $100 million, but we do run the gamut from 1.6 million to all the way up to $12 million for trials that are only including maybe 20 people.

12:55: So, for example, there was two metabolic ward trials. One of them had 20 people, one of them had 17 people, and the cost of this was 1.6 million and 1.3 million, respectively, for a one-month and a two-month trial. So if you go with outpatient studies, it gets a little less expensive, but you're still talking just one study for $8 million with 600 participants, $12 million for 200 participants.

13:19: Particularly in the nutrition space, where money's a little tight compared to the pharmaceutical space, there's a couple of big challenges that are going to be facing anybody that wants to design a very well-conducted clinical trial trying to look at some of these more nuanced questions that we have.

Dan Pardi - 13:34: In a conversation with Christopher Gardner in his office a few years ago, he was telling me about a study that he did, where he was aiming to look at the effect of garlic on ...

Mike Hull - 13:43: Oh, yeah.

Dan Pardi - 13:44: ... health outcomes and ...

Mike Hull - 13:46: Yeah, I know that one.

Dan Pardi - 13:47: Yeah, and how he and his staff would get up every morning at four AM for months to make sandwiches that included garlic in them. Think about how difficult that is, as a researcher, to make food every day for an extended period of time for all the amount of people in this study. That is a huge amount of effort.

Mike Hull - 14:08: I mean, it really is. That's why these inpatient, where the food is provided, studies, the costs just balloon, not only because of the food, but because of the manpower that it takes to prepare that food and then deliver it to your ... It's a lot of logistics and a lot of money.

Dan Pardi - 14:21: One of the biggest problems, just generally, in observation research is measurement error. Particularly in nutrition, we rely heavily on self-report.

Mike Hull - 14:28: Yeah.

Dan Pardi - 14:29: We might, let's say, do a dietary assessment at one point in time and then track outcomes over 15 years. How accurate was that original assessment? How likely is that person's diet consistent over that next period of time? It's just difficult. What can scientists do, perhaps, in the future to improve accuracy of measurement and thus improve the quality of the data acquired in nutritional epidemiology? I ask this because I think we could be entering into a new phase, where, because of technology, we could do some novel and interesting things to take nutritional research to the next level.

Mike Hull - 15:01: I agree with that. I really hope that some of these methods that we're talking about now actually get proper validated studies in the next five, ten years so we can start employing some more objective measures, other than presenting somebody with a [inaudible 00:15:14] to recount what they've eaten in the past year. I don't know if anybody has ever had the displeasure of filling one of those out, but they're incredibly tedious, very boring, but necessarily so. Right? Because they're trying to get at specific data points.

15:26: But, yeah, there are a couple methods that are hopefully on the horizon of becoming validated methods and can actually use in some of these observational studies. I know one popular one was adding photographs with cell phones of the food you're eating, as opposed to asking somebody to specifically log every single food that they eat, which can get tricky and has a little bit more error involved, depending on who the person is. There's apps that you can use to more easily log eating episodes in the moment or accurately.

15:55: I know that, on the statistical side, there's analyses that are trying to be developed to help us get at questions about, for example, which foods are the most effective at capturing overall eating patterns so the number of questions can be decreased, which would also decrease the time burden. You only have to fill out a 50-question questionnaire, as opposed to 200, which I don't think anybody would complain about, researchers and participants alike.

16:18: Then probably the most fascinating to me is the area of biomarkers, using biomarkers to help confirm eating patterns. The easiest example of this is the keto diet. You can measure somebody's ketone levels in their blood as a proxy to confirm what they might be reporting from their food logs. So if their food logs are saying they're eating 20 grams of carbs, but they effectively have no ketones in their blood, then you can call the veracity of that questionnaire into question. Maybe reporting on this particular diet is not the most accurate. So I think those are some of the four most promising areas to keep an eye out for in the next coming years.

Dan Pardi - 16:54: Another thought would be to use something like Blue Apron or meal delivery service to provide the meals. Whenever you can do that, you have a greater chance of assessing what you plan to assess.

Mike Hull - 17:06: Yeah.

Dan Pardi - 17:07: So, in reference to another clinical trial that Christopher Gardner did, the A TO Z study ...

Mike Hull - 17:13: Yeah.

Dan Pardi - 17:13: ... Dean Ornish was upset because the Ornish diet didn't look good relative to the other diets. Ornish's pushback was the adherence to the Ornish diet was nowhere close to what he actually recommends. His point is, "You didn't assess my diet. You assigned them to the diet, and this is what they did." Gardner's take on that was also an important one, too. "Yes, we provided education. We assigned them to the diet, and this is what it looked like in real life."

Mike Hull - 17:42: Yeah. I understand both sides, too. Right? They're asking two different questions. They're asking what happens when you actually follow the diet versus what happens when you just provide support education about the diet and set people free into the world? The inpatient, outpatient, and they're both important questions to ask and interesting questions to get that. I see why both of those help us get to the answer of what might work best.

Dan Pardi - 18:03: There are many reasons why doing nutritional research is difficult. We might be entering into a new phase, where some of the available technology, food delivery, assessment of biomarkers, photo capturing of meals, which could provide more accurate information about what actually occurre. Even then, however, if you care mostly about outcomes, the effect of diet on outcomes can oftentimes take many decades.

18:28: Let's say you're looking at biomarkers. You put somebody onto a ketogenic diet, and the biomarkers shift. How reflective is that of pathology or simply those biomarkers are conditional for a new dietary style? It's an aberration from what was the standard, but it might not necessarily be reflective of a pathological condition. That's just one example, but, of course, it can apply in a lot of ways, though. Again, another confound to interpretation.

Mike Hull - 18:53: Yeah, yeah. You're kind of stuck on the predictive power of these biomarkers, and it's messy.

Dan Pardi - 18:59: So then we circle back to GRADE, and the point of GRADE is to apply this very high, rigorous standard. We don't see that the evidence is very strong. That is an important point to make, even though the people that are doing nutritional research are doing the best they can with the tools they have.

Mike Hull - 19:14: Yeah, yeah. It's a tricky needle to read, because, on the one hand, I understand the argument of holding all the research up to the same standard and it being okay to say, "Hey, the human evidence that we have for this specific topic is not super great. It might be able to get a little bit better, but we're never going to be able to achieve that top tier of research quality, just given the limitations inherent to nutrition research."

19:39: This is kind of where it becomes very tricky. I think this is where a lot of the argument online came about other recommendations that stemmed from this group, is that it is okay to say, "The evidence isn't great," but then how do you go about communicating that to the general public in a way that is being both truthful to the results, but it's also actionable? Right?

19:59: So public health messaging is a game of telephone. So where a study might initially say, "Hey, we want you to reduce your red and processed meat intake," by the time it filters down through all the various outlets and mechanisms through which people obtain their information, that might have gotten shifted into, "Don't ever eat red meat." Right? So part of the issue is just simply getting out an accurate message, period, and the second part of that is how do you communicate uncertainty in research results?

20:30: A great example of this that I was able to find was on the World Cancer Research Fund's website, where they have something called the Interactive Cancer Risk Matrix, and it's a bubble chart. So each topic has its own bubble. So dairy products, foods containing fiber, processed meat, refined grain, and [inaudible 00:20:50]. From top to bottom, it goes, "These things have convincing evidence that they will decrease your cancer risk," and it becomes a little bit less concrete as you scroll down. Then you flip over to, "Oh, these might increase your risk. There's a limited suggestion of increased risk." You just keep scrolling down. "These might be a probable increased risk." Scroll down further, it's a convincing increase in risk.

21:10: So I think this is a pretty good example of how do you communicate uncertainty within the evidence, to the point where you're not dumbing it down enough where people think that you're saying, "Don't eat red meat," and then a study comes out like this, and they feel that they have not been given accurate information, or they feel that the whole field is unreliable. Right? They just start to ignore messages that we're trying to send to them, because they feel like they've been burned in the past.

21:33: I think eggs are a great example of this. One day, they're healthy. One day, they're not. But, yeah, eggs, coffee, and red meat seem to be the trifecta of flip-flopping nutrition research ...

Dan Pardi - 21:43: Right.

Mike Hull - 21:43: ... when I think the root of the problem is just communicating that uncertainty, and how do you do that? That is very challenging, because people tend to like more concrete, actionable items.

Dan Pardi - 21:53: I love the saying, "Have a strong opinion held loosely" ...

Mike Hull - 21:56: Yeah.

Dan Pardi - 21:56: ... which, to me, means take the time to formulate an opinion based off of what we know, but be open to new information that might upgrade your level of understanding closer to what might be more true. That means that you have to basically make decisions and deal in uncertainty. It is one of the reasons why I like to not make dietary choices based simply off of individual elements of a diet, but rather what's heuristic for a pattern that I tend to like, and I personally fall around a paleo-Mediterranean style. That might change in time.

Mike Hull - 22:31: Yeah, I would agree with that overall sentiment, and one hopeful thing that I've seen from dietary guidelines in recent years is the shift away from these individual food prescriptions or individual micro, macronutrient prescriptions to starting to talk about more of the overall dietary pattern, which can better reflect how people eat, period.

22:49: I think Canada's health guidelines were the most recent example of this, where they made a pretty substantial shift in the way they're just communicating to people what constitutes an overall healthy eating pattern and being a little bit more general guidelines, as opposed to specifically don't eat more than 500 grams of red meat in a week, which is not how people eat.

Dan Pardi - 23:08: Right. Give somebody a heuristic or a proverb, kind of like what Michael Pollan did in Omnivore's Dilemma. "Eat food, not too much, mostly plants." That's something that is short enough that encapsulates a broader philosophy that then lets me make better decisions in the face of real-life guidance towards your overall dietary pattern, versus 500 grams of red meat. I honestly don't personally know what that would look like, in terms of how much that is ...

Mike Hull - 23:35: Who does, except for research and researchers?

Dan Pardi - 23:38: Exactly. So, given that you've spent so much time on this, what is your strategy towards making better food choices in your life?

Mike Hull - 23:46: So mine is honestly ... It's very similar to yours. I'm looking at the overall dietary pattern. Mine has been a gradual shift, and it's changed as I've aged as well. When I was younger, it was all about how much muscle can I put on? How fit can I be? Then I ended up doing a nine-month stint in a hospital, where I worked with colorectal cancer patients, and it was a multimodal intervention, where we were giving them pre-surgery exercise, diet advice, psychological help needed.

24:13: That really fundamentally changed the way that I both approach eating and exercising. So now it's more for quality of life, longevity than it is for that optimal muscle-building diet. It has very much shifted away from that, into a pattern that will provide me a high quality of life for the longest period possible.

24:34: But that has become very important to me. I want to age well, because I really saw the full range of health possibilities when I was working with these older colorectal cancer patients. I had patients coming in in their eighties and crushing all of these exercise tests that we're throwing at them, and I had people in their forties coming in who weren't even able to administer some of the tests, because they just weren't able to do that.

24:56: So that has definitely shifted my perspective in the way that I approach eating. I don't particularly follow a known template. Mine's pretty omnivorous, and I like beans. That's pretty much beans and milk. It's pretty simple. It's pretty straightforward. Don't get too crazy.

Dan Pardi - 25:10: I see a book on the horizon, Bean and Milk Diet.

Mike Hull - 25:13: It'll sell fives and tens of copies.

Dan Pardi - 25:16: This almost seems like a non-sequitur here, but, also, when you're assessing an individual food type, particularly when we're talking about meat, what about if you do an 11 to 7 eating window or you do some sort of fasting? Does the outcome change? What if a lot of the problems with meat consumption, if they exist where how the meat was prepared, instead of it being grilled?

Mike Hull - 25:38: What else question.

Dan Pardi - 25:38: Yeah.

Mike Hull - 25:39: Yeah.

Dan Pardi - 25:40: Grilled, we know that that adds a significant burden of advanced glycation end products. If you're interested to learn more about those, go listen to my show. It's for the audience with Pankaj Kopahi from the Buck Institute. What if, instead of grilling the meat, you're stewing it? Does, again, the outcome change if you're now preparing it differently?

26:00: So whenever we do see this type of food as necessarily bad, there are those other factors that could play a huge role that's either entirely causing an issue that does exist or might be at least a significant portion of it. So it's really challenging to make sense of it all, but you have to keep that stuff in mind.

Mike Hull - 26:19: Yeah. That's basically the point that we made at the end of this article, is that it's easy to get caught up in a singular food vortex, essentially a black hole, but people can miss the forest for the trees sometimes. Right? Does red meat intake matter? Well, compared to what else? Let's look at your genetics. Let's look at your overall dietary pattern. Let's look at your exercise and your vitamin and mineral intake and your overall calorie intake. I don't want people to so myopically focus on single food items. They're missing a host of other factors that can greatly affect their quality of life.

Dan Pardi - 26:52: We will definitely link to your thorough article on examine.com. I'd recommend taking a look at what Michael has put together. That's a lot of work. How long did that take you?

Mike Hull - 27:02: I don't even remember. I think I blacked out and wrote the article. We actually had to co-write this article. My colleague, Zad Chow, who is a big stats nerd, also came in and wrote about some of the uncertainties of evidence and how do you interpret a confidence interval so that people, if they want do, they can go in and actually dive in and read these studies and have an understanding of what they're actually communicating.

Dan Pardi - 27:22: Having a stats nerd as a wingman, it makes you extra formidable.

Mike Hull - 27:25: It makes it a whole lot easier.

Dan Pardi - 27:27: Totally. Michael, thank you. This helps to add important context to the space of nutrition research, where we might go next, and then how to make better decisions in your life today without driving yourself mad. So valuable time, and thank you for coming on to humanOS Radio.

Mike Hull - 27:44: Thanks for having me.