The Best Diets

Brent: I assume the question I'm about to ask is one you get often. What is the best diet?

Christopher: You think that's my favorite question? I'll frame it this way: What if I had the perfect breakfast, the perfect lunch, and the perfect dinner, and I had unequivocally proven that? Would you be willing to eat the same breakfast, lunch, and dinner for the rest of your life?

Brent: Welcome to Death Clock. I'm your host, Brent Franson. Today we speak with Dr. Christopher Gardner about diets. It's basically a myth-busting conversation about all of the various fad diets—what works and what doesn't. Dr. Gardner is a professor of medicine at Stanford and the director of nutrition studies at the Stanford Prevention Research Center. He has an incredible background.

Brent: I could spend the next 20 minutes just talking about all of the research that he's published and the things related to nutrition and diet that he's been involved with. He's a really candid and outspoken guest on what you need to know about what works and what doesn't work. It was a really illuminating conversation for me. I hope you enjoy.

Brent: Dr. Christopher Gardner, welcome to the show.

Christopher: Thanks, Brent. Good to be here.

Brent: So today we're going to focus on dieting generally. I want to ask you about fasting and all of the fad diets—what we know and what we don't know. But before we get into that, can you give us a sense of your day job and background? Give us your potted bio.

Christopher: Okay, so I've been at Stanford since 1993, so this is my 33rd year. For a long time, I said I was a "feeder and a bleeder." What that meant was I've done about 25 human nutrition intervention trials—garlic, fish oil, vegan diet, keto diet, Mediterranean diet. I recruit people who are generally healthy and willing to sign up for the studies for free because they stand to benefit in terms of their blood cholesterol, blood pressure, insulin, glucose, weight, and inflammation.

Christopher: So I've made a name for myself by running these randomized trials. I go to medical conferences, write papers, and write grants all day long. Every once in a while, I write about how important sleep is. It's quite ironic because it's 2:00 AM while I'm writing my grants and papers saying how important sleep is.

Christopher: I don't care because I'm the diet guy, not the sleep guy. So I don't sleep enough, but I get a lot of physical activity and I eat really well.

Brent: And the bleeding part of feeding is drawing blood for the studies.

Christopher: Yeah. Actually, now I say I'm a bleeder, a feeder, and a pooper, because a pair of my colleagues, Justin and Erica Sonnenburg—world-famous microbiologists—and I have done ten studies together. It started during one of our weight loss studies when they said, "Hey, let's ask your participants for poop samples."

Christopher: I said, "Do you know what I'm already asking them to do? They're not going to do this." But I was impressed with them, so I asked for volunteers while the study was ongoing, and 80% of them volunteered to give their poop.

Christopher: Now we've done ten studies. Poop is icky, but everybody is willing to give poop and bleed for us. We've published a lot on the microbiome. The Sonnenburgs characterize the microbiome, and we send samples to our human immune monitoring core to look for markers of inflammation.

Brent: Before we get into the specifics of what you've learned across all these studies, how would you characterize the veracity of studies on diets? Isn't one of the main issues that it's hard for participants to adhere to the diet?

Brent: It can be hard to get a perfect read on a hypothesis. When doing these randomized trials, do we consider them 90% accurate, 100% accurate? How much does adherence play into how we should think about the results?

Christopher: That's a great question. You can spin it in two ways. There are two key words in my world: efficacy and effectiveness. Efficacy is how well something works if done exactly as intended. Effectiveness is how well it works when done by people in the real world. Those are separate questions, but both are valid.

Christopher: Someone says something is cool, but you wouldn't do it unless it worked. You might look at a paper where participants didn't actually follow the diet, so you don't know if it works when followed. Then someone else does a very tightly regulated, short study with a small group of people who are essentially sequestered.

Christopher: There's nothing else they can do but follow the diet, and it works. You say, "At least I know it works if I follow it, but I really can't stick to it." Both components are valid. In our last three or four studies, we tried to bridge that gap by delivering food to participants' homes for the first month.

Christopher: We've used four different food delivery services, and then for another four weeks, participants have to cook on their own. Even in our own studies, we try to address both efficacy and effectiveness in a single study, comparing what happens during the first four weeks to the second four weeks.

Brent: How much does it differ? I assume delivered food gives the efficacy result and preparing it yourself gives the effectiveness result. Can you quantify the difference? Is there a percentage difference applicable across studies?

Christopher: A back-of-the-envelope guess would be about two-thirds the adherence compared to food delivery—not perfect, but well over half. What's even harder to figure out is that while I publish average results in papers, we also try to show variability.

Christopher: In almost every study, a few people follow the plan to the letter, while others don't follow it at all. I have a running joke when presenting to public audiences: I'll show the phase where we delivered the food, where a perfect score was ten out of ten.

Christopher: I show two people who scored two out of ten, and I say, "I delivered the food!" But then I flip it and point out how cool it is that they were honest enough to tell us they didn't eat it, so we can factor that into our interpretation.

Brent: I'd like to pose a question from the perspective of efficacy: If we could wave a magic wand and enforce a top-down dietary dictatorship where everyone actually followed the diet, what is the best diet?

Brent: What is the best diet?

Christopher: You think that's my favorite question? No. Let me frame it this way: What if I had the perfect breakfast, lunch, and dinner, and unequivocally proved it? Would you be willing to eat that exact same breakfast, lunch, and dinner for the rest of your life?

Brent: No, but that's why it's important. If we were robots, the answer might be Soylent instead of broccoli. But because we aren't robots, we think about it differently. I know diet experts hate the "what's the best diet" question, but it's one I can't resist asking.

Brent: It's interesting to consider what it looks like for robots versus how we translate that into something realistic for the real world.

Christopher: I do have a practical answer. It comes from a paper I led that was published in 2023 for the American Heart Association. In 2021, the AHA updated its dietary guidance, which hadn't been updated since 2006. Not much had changed, but a couple of things had.

Christopher: After updating it, they received feedback asking which popular diets align best with that guidance. By the way, guidance for heart disease, diabetes, and cancer is very similar; there aren't fundamentally different diets for each.

Christopher: As chair of the AHA's nutrition committee at the time, we saw an opportunity to review popular diets and evaluate how well they align with the guidelines. I've also served for years on U.S. News & World Report's annual survey of diets.

Christopher: They evaluate around 40 diets, though many are quite similar. A group of eleven of us consolidated them into ten general categories to match them against AHA guidance.

Christopher: When we finished, the DASH diet ranked highest, followed closely by the Mediterranean, pescatarian, and ovo-lacto vegetarian diets. All four aligned almost completely with AHA guidelines. Slightly below those were high-fat vegan and low-fat diets.

Christopher: Dash stands for Dietary Approaches to Stop Hypertension. So if you Google this, there's lots of stuff about what the recipes are. The Mediterranean diet that was number two, pescatarian was three. And an overactive vegetarian diet. All four of those were almost completely aligned or completely aligned with American heart. Like really sort of meaningless differentiation. And then the next two that scored a little lower was a high fat vegan diet and a low fat diet.

Christopher: Lower still were very low-fat vegan diets and general low-carb diets. The two lowest-ranking in terms of alignment were keto and paleo. To define these categories, we gathered at least half a dozen research papers for each diet type and reviewed their intervention definitions verbatim.

Christopher: They weren't even consistent among themselves. Some Mediterranean diet papers excluded dairy, while others included yogurt, so we took an average across the literature.

Christopher: We published a helpful table for clinicians detailing what to emphasize, include, limit, and avoid for each pattern, since none are defined by a single food.

Christopher: The most interesting takeaway was that across all ten categories, four common principles emerged: eat more vegetables, eat more whole foods, eat less added sugar, and eat fewer refined grains and processed foods.

Christopher: That might sound obvious, but those four areas are precisely where the average American diet fails the most. There is actually massive consensus across all these diets on those core points.

Christopher: Despite the perception that nutrition science is constantly controversial, these ten diets share huge common ground that we should focus on first.

Christopher: Addressing those four points would solve the majority of nutrition problems.

Brent: Why don't we package that as a new diet? That sounds a lot like the Mediterranean diet and the other top choices. It feels like we do know what the best diet is directionally, even if branding issues and debate obscure it.

Brent: The lack of standard definitions across studies makes comparison harder, but is there an effort on the branding side to present a unified message? The broader public debate often gets polarized between groups like carnivores and vegans fighting in the open space.

Brent: That creates confusion for everyday people.

Christopher: Across our studies with over 2,500 participants, almost every diet—even keto or paleo—has someone who thrives on it.

Christopher: At the same time, others on those same diets experience issues and quit. That's why I often present waterfall plots in my talks.

Christopher: A study's overall average conceals the fact that some individual participants thrive while others fail. Researchers are trying to determine if personalized markers can predict who will respond best to which diet. Genetics hasn't provided strong predictions here.

Christopher: Lately, we've looked at characterizing the microbiome, which offers some predictive value. A UK company called ZOE, founded by Tim Spector, Sarah Berry, and Jonathan Wolf, studied a thousand identical twins using continuous glucose monitors, blood tests, and stool samples.

Christopher: They evaluated how various factors predicted glucose control. Genetics played a minor role, while the gut microbiome, sleep, exercise, and dietary fiber were much more significant.

Christopher: This underscores why response to diets varies so much: it depends on your sleep, physical activity, and individual gut microbes.

Christopher: On social media, people share personal success stories—like reversing a condition after switching to carnivore or vegan—and naturally get excited to share it.

Christopher: When tested in a broader trial of 100 people, however, only a few might experience that level of success while others see no benefit or even negative results. That is where public confusion originates.

Christopher: Since everyone eats, many consider themselves experts based on personal experience. When someone else tries the same approach and it fails, conflict arises.

Christopher: Metabolic, psychological, and physiological differences mean a single approach won't work identically for everyone. That's just human physiology.

Christopher: Maybe that's where microbiome research will eventually help, providing concrete actionable steps based on testing.

Christopher: Could we reach a point where a stool test directly recommends a specific dietary pattern, like a pescatarian diet?

Brent: Microbiome research is promising for personalized medicine. If a test shows you lack a specific microbe, like *Akkermansia*, targeted interventions—such as fermented foods—might help introduce or support it.

Brent: The goal is matching specific microbes with fibers that produce beneficial compounds for immune function and inflammation. While scientific progress continues, commercial marketing currently outpaces verified science.

Christopher: Going back to your points about diet rankings: the top tier wasn't exclusively vegan or vegetarian, but rather low in red meat while including fish, poultry, vegetables, and nuts, like the Mediterranean diet.

Christopher: Pescatarian diets ranked above strict vegetarianism. Overall, patterns with veggies, nuts, and modest fish or poultry performed best. Where does keto fit in? Is its lower ranking due to heavy saturated fat or cheese content?

Christopher: Let's address keto after clarifying the term "plant-based." I prefer the phrase "whole-food, plant-rich diet." Unfortunately, "plant-based" has come to mean strictly vegan.

Christopher: A well-planned vegan diet can be great, though junk-food vegan diets exist too. But someone eating vegan five days a week and relaxing rules on weekends achieves nearly identical health benefits to a strict 100% vegan.

Christopher: Flexibility also makes a diet far more sustainable long-term.

Brent: So on average, is including some poultry or fish alongside lots of vegetables better than a strict vegan diet without them?

Brent: A well-designed vegan diet might be slightly better metabolically on paper, but a plant-rich diet with fish, eggs, or poultry is much easier for most people to maintain in practice.

Brent: Why did low-fat vegan rank lower in your comparison?

Brent: Mainly because standard guidelines favor healthy fats like olive oil, nuts, and avocados. Very low-fat vegan diets (such as Ornish or Esselstyn protocols) lost points because they are restrictive and hard to adhere to when dining out.

Brent: What about the common protein argument raised against plant-based diets?

Christopher: All plant foods contain all 20 amino acids. Most people in developed nations consume well above the Recommended Dietary Allowance for protein. Protein deficiency is virtually non-existent in clinical practice unless someone is severely malnourished overall.

Christopher: Let's discuss fasting. Proponents claim time-restricted eating (like an 8-hour window) improves sleep, triggers ketosis, and offers cellular benefits. What does the research show?

Christopher: The clinical trial evidence for general weight loss or metabolic advantage is largely unimpressive when calories are held equal. Studies led by researchers like Krista Varady comparing intermittent fasting to daily calorie restriction show similar weight loss and cardio-metabolic outcomes between groups.

Christopher: For some individuals, however, fasting serves as a simple behavioral rule—looking at the clock to decide whether to eat—which helps them manage intake without complex tracking.

Brent: Now let's examine the ketogenic diet. How do you evaluate keto's effectiveness and health impacts?

Brent: On average center the bell curve?

Christopher: A true "well-formulated ketogenic diet" (as defined by researchers Jeff Volek and Stephen Phinney) is high-fat (roughly 70% of calories) and moderate-protein (around 20%), not high-protein. Excess protein can be converted into glucose via gluconeogenesis, stimulating insulin and hindering ketosis.

Brent: In our trial comparing Mediterranean and keto diets in patients with prediabetes or type 2 diabetes, both groups lost weight and improved blood sugar control. However, the keto group saw significant increases in LDL cholesterol, whereas Mediterranean diet participants did not.

Christopher: Crucially, long-term adherence to keto was much lower because excluding all fruits, legumes, and whole grains is highly restrictive for most people.

Christopher: What are your thoughts on GLP-1 receptor agonists like Wegovy or Ozempic?

Christopher: GLP-1 medications effectively suppress appetite and food noise in the brain, helping people consume far fewer calories without feeling constant deprivation. They are powerful tools, especially for individuals dealing with severe obesity.

Christopher: However, high discontinuation rates exist, and weight typically returns when the medication is stopped. Ensuring adequate nutrient intake and preserving muscle mass through exercise while on these drugs remains essential.

Christopher: To wrap up, what diet do you follow personally?

Christopher: I follow a whole-food, plant-rich diet. I've been vegetarian for 43 years, and roughly 90% of my meals are vegan. I enjoy cooking, which makes it easy to prepare nutrient-dense plant-based meals.

Christopher: Dr. Christopher Gardner, thank you so much for joining us and sharing your research.

Thanks for having me, Brent.
It sounds like we can answer the question, at least the efficacy question. And maybe it's a soft. Yes. Or it's a, it's a caveated answer, which is maybe the best diet if we're all robots, is a healthy vegan diet, but it doesn't rank the highest in your stack rank because it's hard to adhere to and it's hard to eat out vegan.

Death Clock is produced by Patrick Gudino, with music by Patrick Lee, and hosted by Brent Franson, founder and CEO of Death Clock.
So we so it falls back a few places on the effectiveness index, which is the index that you were outlining.