Most nutrition advice starts by telling you what not to do. Don't consume salt, eggs, butter, or fish. Don't drink. Avoid red meat.
It's not a fun way to eat. And according to Aaron Carroll, a doctor, columnist, and director of the Center for Health Policy and Professionalism Research at the Indiana University School of Medicine, it's an overly restrictive approach that is also not supported by the evidence.
In his new book, The Bad Food Bible: How and Why to Eat Sinfully, Carroll explores the history of nutritional advice, from media-driven fad diets to faulty government nutrition guidelines, and explains how to judge solid nutrition science from junk studies. He rounds up the evidence on butter, salt, dairy, gluten, alcohol, red meat, and more, and finds that much of what we think we know about healthy eating is based on shaky research or plain old myth. Ultimately, he argues that we should enjoy food, not fear it.
Over email, Carroll and I discussed the enduring appeal of dietary moralism, government's poor track record with nutrition advice and food policy, what nutrition science says about Thanksgiving, and his professional medical opinion about my ideal diet of potato chips, fried chicken, and whiskey.
Reason: Something I really appreciate about your book is that it's not moralistic or restrictive. Although you do issue some warnings about certain behaviors to avoid, a big part of your message is that it's actually fine to consume most food and beverages. It's a book that repeatedly says, sure, it's okay to eat or drink that, at least in moderation.
That's quite a bit different from a lot of the diet advice we see, which tends to be heavily restrictive and focused on what you should avoid consuming. And it's also different than the moralism found in a lot of hand-me-down health wisdom, which is all about which foods are inherently good and which are inherently bad.
What's the appeal of restrictive diet moralism? Why does it persist—and in many cases spread? It can't be because it's pleasurable to eat that way!
Aaron Carroll: For as long as I can remember, nutritional advice has always been about telling me I'm doing something wrong. It was always telling me I was eating the wrong things. Don't eat cholesterol. Don't eat fat. Don't eat carbs. You have to eat something.
I think there are likely a few reasons for this. One is that some people think that making people feel shame is a motivating factor. Another is that we tend to think that being overweight or obese is somehow your "fault" and that you, therefore, are to "blame". Too often we equate being overweight with a moral failing.
We also shouldn't discount the financial drivers. There's lots of money to be made by making people feel afraid, and it certainly works in food.
Reason: On a related note, I want to thank you for giving your official medical professional's blessing to my all fried chicken, potato chips, and whiskey diet. Oh wait, that's probably not right either. How do you balance the twin extremes of diet restrictionism and permissiveness? It seems like a hard thing for people to think about.
AC: There's always the danger that people will take what I'm saying and translate it into a license to eat anything they want all of the time. That's not what the book says. I would, however, say that the occasional dinner of fried chicken, potato chips, and whiskey isn't going to kill you either.
Too many medical professionals think that people can't hear nuanced messages. I get emails from them all the time saying we have to tell everyone to eat very low salt diets—even if they benefit few people and might be bad for some—because if we don't, then the people who need to restrict salt won't. I think most people are smarter than that. I think they can handle the nuance.
There are some guidelines in the last chapter of the book that would allow for a wide range of eating habits, including many "bad" foods, that would still be considered healthy.
Reason: Another thing I like is the way you handle the "study says" part of the healthy eating tips business. Every chapter debunks myths about food and drink related health risks, many of which have been spread using dubious, or dubiously interpreted, science. It really seems like you are taking aim at the entire diet advice industrial complex. What accounts for the huge amount of bad advice about eating that we see everywhere—from government sources and mainstream publications and even in scientific journals?
AC: Some of it is an overreach of science. Humans are complicated animals. The reasons we eat certain foods, or do anything for that matter, are much more complex than other species. Yet so much nutrition research is predicated on small studies of mice for short periods of time.
It's not just mouse research either. A lot of the human research done is observational, which is fine for looking for potential links between behaviors and outcomes, but isn't so good at figuring out causality. The number of good, long-term randomized controlled trials in nutrition is very, very small.
There are professional conflicts of interest all over the place, too. Many academics have made careers out of promoting things like a "low-fat" diet. When you get used to the grant money and prestige of being part of the consensus, it's very hard to change paths.
There's plenty of blame to go around with respect to the media as well. They trumpet results from flawed studies far too often. They also take each new study, no matter how small, as overturning all previous research without considering it in context. Well researched articles that consider all of the evidence, old and new, are harder to write and not nearly as click-baity.
Reason: Let's talk specifically about the government for a moment. Libertarians have been arguing for a while that the government has historically gotten an awful lot wrong when it comes to nutrition. The most notorious examples are the promotion of low fat diets and the—related—demonization of both saturated fats and salt.
Your book contains many examples in which advice or policy from official institutions has been either wrong or counter productive: Beyond mistaken diet advice, you talk about the USDA's promotion of milk and dairy products, restrictions on genetically modified foods in Europe and Vermont's state-level GMO (genetically modified organism) labeling, the World Health Organization connecting red meat and cancer. A lot of the mistakes went on for years, even decades, as contrary evidence piled up. You also write that nutrition guidelines are starting to catch up, at least in some instances.
It seems to me that, even limiting myself to examples in your book, the track record of government when it comes to diet and nutrition advice is not great. Doesn't the history here suggest that we should at least be cautious, a little skeptical—or maybe even very skeptical—when it comes to both government nutrition guidelines and policy interventions designed around that advice? Or, to put it another way, how can we be certain that we've gotten it right this time?
AC: If you're looking for me to argue against this, you've got the wrong guy. I've written multiple columns—and this book!—on how we've gotten food policy wrong, and how the USDA and other parts of the government continue to promote nutritionally questionable things. This includes support for the dairy industry, the demonization of eggs, and the many, many iterations of the food pyramid. During World War II, one of the seven food groups was "butter and fortified margarine." I think the policies around GMOs, around menu labelling, etc. have just more to do with politics than science.
That said, I think government is no better or worse than many other organizations when it comes to questionable food policy. Many professional medical organizations are just as guilty of being on shaky footing.
Reason: Relatedly, isn't there another problem with the sorts of nutrition guidelines we often see from official organizations, which is that a proper diet is going to be different for different individuals? It depends on all sorts of idiosyncratic personal factors: health history, activity levels, personal tastes, what you are optimizing for.
In your book, you note that one can eat a very broad range of foods you can eat in the context of a healthy diet. But a lot of nutrition advice, whether from official sources or from diet gurus, seems designed to push people into a fairly narrow range. The inherent problem with general guidelines seems even stronger when you're dealing with government guidelines, since they end up having so much influence. Without going so far as to say that no nutrition advice can ever be broadly applicable, it strikes me that there are real limits, and I wonder if you agree and what you think those limits are.
AC: Unfortunately, it's much easier to tell people what not to do than to tell people what to do. But I think you're noting the flaw with medical guidelines, in general. We have a general sense—because we have studies—of what to do in particular situations for particular people, usually at high risk. We get into problems when we start broadening those recommendations into other groups to catch a wider net. The benefits go down while the harms remain the same.
It's easy to spot people we know are eating really unhealthily and say, "don't do that." The problem is that we don't have the evidence base to tell reasonably healthy people—and most people fall into that group—what to do. I'm amazed when I talk to healthy, fit friends and they relate to me that their doctor just told them to make a drastic change in their life. Why?
You can make a lot of money by selling a new fad diet, but they are almost never evidence-based. They are always super-prescriptive, as well. In general, I think there are many, many diets that can fit into a "healthy" eating pattern. I try and get into those in the last chapter of the book. I doubt that will make me a lot of money.
Reason: Poor journalism on nutrition science is something you deal with a lot. Too many media reports are based on low quality studies, or single studies that are contradicted by other evidence. Your book guides readers through how to make better judgments about study quality, explaining differences between observational studies and randomized trials, and why reviews that aggregate multiple high quality studies tend to be best of all.
This is quite helpful. But it also seems, honestly, like a lot of work, especially for non-oddballs who don't enjoy that sort of research and don't have a lot of time to track down and read the studies themselves. What are you supposed to do if that describes you—besides, of course, buy your book? I almost wonder if, for many people, the best advice is to basically ignore the majority of media reports about diet and nutrition. But full on diet-science nihilism doesn't seem ideal either.
AC: You stole my good answer, which was to buy my book. But the longer answer is still similar. Find people you trust to report on the science, and not because they tell you what they want you to hear. The best sources in this area will tell you why you should or should not believe a study, not just what the study says. But if you need broader guidelines, if the news article you're reading is about a large, randomized controlled trial of human beings in the real world, I would pay attention. If it's not, then I'd check to see if it places the new study in context with everything that's come before. If it doesn't do that—I'd ignore it.
When my friends ask me this question, I always tell them, somewhat sarcastically, that they should ask me. I like to track down and read the studies. I even put a lot of them in convenient book format.
Reason: Cancer risk is something that comes up a lot in nutrition advice and in your writing. And in a lot of cases, your message in basically just: Don't worry about it too much. Even when studies do show increased risk factors, the increase in risk is often so small as to be almost meaningless. So when it comes to food and drink, is it safe to say that most relatively healthy people more or less shouldn't worry about cancer risks?
AC: Yes. One of my favorite studies ever was a 2013 systematic review that collected the research on 40 common ingredients selected from a cookbook. They found more than 260 studies that focused on at least one of the ingredients. They basically showed that you can find evidence to show that pretty much all of them could both prevent and cause cancer. If you cherry-pick, you can say almost anything you want about many ingredients and cancer.
Of course, some ingredients have more evidence that's bad than good. Even then, the absolute risks are small. Perhaps the best example is processed red meat, which the WHO thinks will raise your relative risk of cancer by 18 percent. That sounds scary. That's still debatable, but even if it's true, then it would mean that if I declared today that I would eat an extra three pieces of bacon every day for the next 30 years, my absolute risk of colon cancer might go from 2.7 percent to 3.2 percent. That's… not that scary. And I'm not going to eat that much bacon, so it's likely much less.
Reason: As you may have picked up, my biggest question when it comes to nutrition science is basically just: How certain can we really be? The science is complex, the studies are often weak or contradictory, and individual bodies have so much variation that even good studies can sometimes only provide us with limited information. There's been a lot of bad diet advice over the last few decades, and a lot of it has been based on what looks like scientific evidence.
And yet, like you, I am also reasonably confident that there's a lot we can and do know—about some of the risks with trans fats and sugar, for instance. So I'm curious, what are you least certain about when it comes to nutrition? What do you think the prevailing scientific consensus in 2017 is most likely to have gotten wrong, or to have not fully understood?
AC: I think I'm least certain about saturated fat. There are still some people who think that saturated fat is no big deal. I'm not so sure. I think the evidence base against it is still much weaker than many say, but I wouldn't go eating unlimited amounts of the stuff.
Unsaturated fats, on the other hand, appear to be much less harmful than many originally thought. Even the most anti-fat groups now seem to be saying that you should replace saturated fats with unsaturated fats as much as possible, but not with carbs which is, of course, not what they always said.
As to what we might be getting wrong—I think salt might be my top choice. There's a growing body of evidence that says that for many people, very low salt diets might not be a good idea. A lot of advocacy groups haven't grappled with that yet. I also think the anti-gluten fad will pass at some point.
Reason: Finally, Thanksgiving is coming up. It is, I think most would agree, the greatest food holiday. The entire celebration is built around the preparation and consumption of an extravagant, often gigantic meal. It is not quite a celebration of gluttony, but it is very much not a holiday concerned with moderation. So, what sort of advice can you give us about Thanksgiving dinner? What does diet science have to say about the plate—okay, plates—full of turkey, buttered rolls, and mashed potatoes that I plan to consume on Thursday?
AC: Science doesn't say much at all, I think. It's one meal a year—and, in my opinion, one of the best. I let it go that night. I eat extra pie, lots of mashed potatoes and gravy, and as much turkey as I can stomach. What you do on that one day isn't nearly as important as what you do on the other 364. It's a holiday. Enjoy it. That's what makes it special!
(This interview has been lightly edited for clarity.)