Food for Thought: What Does the Data Say About Food Dye Safety?

Introduction

0:00Hello, and welcome to the Harvard Data Science Review Podcast. I'm Liberty Vittert-Capital, and along with my co-host and editor-in-chief, Shally Meng, today we're digging into your dinner plate.

0:16From bright blue cereal to shelf-stable snacks, food dyes, preservatives, and ultra-processed foods, they are everywhere. But are they safe, are they necessary, or are they maybe even harmful? In this episode, we talk with leading experts in food science and public health to separate the facts from the fear. Lisa Lefferts is an environmental health consultant and former senior scientist at the Independent Center for Science in the Public Interest.

0:47She is the primary author of the successful petition to ban REDD-3 and also served on FDA's Food Advisory Committee when it considered synthetic food dyes in 2011. And Professor Marion Nessel, an American molecular biologist, nutritionist, and public health advocate. She is the Pollack Goddard Professor of Nutrition, Food Studies, and Public Health at New York University. What does the data actually say about these controversial ingredients?

1:17And are recent legislative bans driven by science or perhaps something else? Stay with us as we break down the science, the politics, and the public perception behind what we eat. Thank you, Mary and Lisa, for joining us. And let's just get started.

Food Dyes Explained

1:36What are food dyes and the preservatives chemically? And why were they introduced in the first place? Well, the synthetic food dyes include blue one, blue two, red three, red 40, yellow five, yellow six, and green three. And red three first was approved, I think it was 1907. So they've been around a long time. They were originally derived from coal tar. And the bright colors, you know, attract people, especially children, to consume the

2:09food. So they're just cosmetic additions to the food supply. They don't add flavor or anything like that, just color. So Marion, because we're a data science podcast, you know, a lot of this is about the studies that lead to these bans that have been happening across the United States. So, you know, I mean, as we know, several US states have proposed or have passed bans on specific food dyes and additives. And I think what I want to know is what is the actual scientific evidence that these substances

2:42are truly harmful to human health? Because, you know, we hear in the news that I think it was red dye three that the ban was off of a study from 30 years ago on some rats. And, you know, 20% of them had these, you know, if they made red dye three, I think it was 5% or 4% of their diet, that 20% of them got benign thyroid tumors. But, you know, for me to ingest, to have red dye three be 4% of my diet, I'd need to have, you know, 100 cups of candy corn a day or whatever, you know, some crazy amount.

Scientific Evidence

3:13So, could you just, could you illuminate what these studies are that have shown us that this is actually really bad for us? Yeah, this goes back to the early 1980s when a physician named Benjamin Feingold was very concerned about hyperactivity and behavioral problems in children and attributed them to the fruit dyes. And so, there was research that started then and linked to hyperactivity. And parents would say that when their kids ate foods that contained fruit dyes, their kids

3:49would be bouncing off the walls. So, at the time, in the 1980s, there was a controlled clinical trial that was done on a very small group of children who were given two drinks, one with a lot of fruit dyes and one with no food dyes so that the drinks looked the same, they tasted the same, but one had a lot of dyes in it and one that didn't. And there were six children in the study, and there was one child in the study who reacted to the drink every single time it was given.

4:22And the parents could tell which drink had the dyes because of this child's behavior. That was not true of the other five. So, that's, in a nutshell, is the problem. Not all children reacted, but some do. And in general, it's very, very difficult to do studies on children, and it's even more difficult to do studies on substances that are consumed in very, very small amounts.

4:52So, the other studies are all animal studies. And those tend to show that some of the dyes are potentially far-synogic. The research, in my view, is iffy. Very, very difficult to get down. It's very hard to do, but these things are completely unnecessary. They have no health function. They have nothing except the cosmetic function. And a sales function, that has been well-documented. You can do studies that show that kids think that foods with bright colors taste better,

5:27even if there's no difference between them. So, from a psychological standpoint, these dyes are really, really important. And when the companies have tried to get rid of them, sales grow dead. That's the problem. I see. That's really quite interesting. But a follow-up question on this is, as you mentioned, right, these are all based on even clinical trials, which is obviously golden standard for doing this type of studies. They're still, in a way, to just find out this as a phenomenon.

5:59But what are the scientific kind of understanding why these dyes could cause the problem? Is there any scientific understanding, chemically or biologically? So, the very best assessment that's ever been done on this issue was done by the state of California. And it was a state-of-the-art, really top-notch, systematic review of all of the evidence. The human evidence, the animal evidence, and the mechanistic evidence. With synthetic food dyes, there are 27 clinical trials that met California's strict criteria,

6:34which is really unprecedented and unheard of for additives. You never have that kind of data. Now, each of the trials you could pick apart. They're often small. They have weaknesses. But combining them together, California discovered that the best conducted studies were the ones that were most likely to find an effect between synthetic food dyes and neurobehavioral effects. And that assessment concluded that synthetic food dyes cause

7:04neurobehavioral effects in some children. Now, cause is not a word that scientists use lightly, as you well know. But they felt confident using the word cause, and I feel confident using that word cause, for several reasons. One is, we do have these clinical trials in children that show that some children, not all children, are sensitive to synthetic food dyes. We don't really know exactly why some children are affected and others aren't.

7:35But the best study we have indicates that it may be due to some polymorphisms, or shall I say, different variations of a neurotransmitter enzyme and how the neurotransmitter is broken down. So that seems to be the reason, but there's only one study and it hasn't been replicated. The other reason that we are confident in the data on the neurobehavioral effects is that all three lines of evidence, the human, the animal, and the mechanistic,

8:06all converge to the same conclusion that yes, synthetic food dyes can cause neurobehavioral impacts in some children. From the animal studies, we can see microscopic changes in the brain structure. We can measure effects on neurotransmitters. So we're pretty confident in those results. Now, you mentioned REDD3 at the outset. Yeah, the study in REDD3 was done in the early 80s, and FDA gave the industry 36 extensions

8:38of the deadline to make its case that REDD3 was not a concern for people. Industry had argued that, you know, it was really due to these high doses they were using that caused a hormonal imbalance. And the hormonal imbalance led to the cancers, and a lower dose that people were exposed to would not cause cancer. But after 36 extensions, the industry failed to convince FDA that that was the case. And so FDA banned REDD3 from cosmetics and externally applied drugs in 1990.

9:15And at that time, it said it would take steps to also ban REDD3 in food, but it never did. And so the Center for Science and the Public Interest filed a petition urging FDA to ban REDD3 in food and ingested drugs, and they finally did just recently. And that ban goes into effect in, I think it's 2027, although this administration has said they might try to speed that up. So some people have said, well, the mechanism wouldn't operate in people,

9:48but I'm just not buying it. FDA scientists back in 1990 talked about the kinds of studies they would need to see to be convinced that even though it caused cancer in rats, that it would be safe for people. And those studies were never done. And I know that because I have looked at every single study that's ever been done on this topic. We had to include those in the petition on REDD3. You know, I have to ask a follow-up to this because I have no sort of dog in this fight, except that I like Froot Loops. And I am having a baby in a couple of weeks. So I've been looking

10:22into this stuff recently. And I looked at the California studies and I got to say, and I say this as a statistician, nothing else. It felt really subjective to me and that it was parent and teacher evaluations. The parents saw some difference. The teachers saw absolutely no statistically significant difference. And I didn't buy it. It didn't do it for me. Now, Marion, your thing that it's just not necessary, so why would you do it, is a better argument than I've really seen anywhere else. But am I missing something? Like, what have I missed here, Marion?

10:58The data from the 1980s have shown exactly the same thing, that there seem to be some children who by parental or other kinds of evaluations and some biological measures seem to react to this, and that there's no evidence at all that these things are healthy. That's another way of putting it. So if you're looking at the precautionary principle, which is what they use in Europe, these things have no value for health, nutrition, or safety. They're totally cosmetic. Their only

11:36value is to the food industry. They have no value for human health. And they're drawing suggestions over a 50-year period that they may be harmful to at least some segment of the childhood population. Why are they still in the food supply? It doesn't make any sense. They're not necessary. Get rid of them. And it's true that trying to measure changes in behavior, it's not like, you know, measuring body weight, right? It's tricky, especially in kids, and the setting can matter, etc. But I thought what was

12:12interesting in the California analysis is that they looked at some of the better studies that had validated measures of behavior, not just, you know, what a parent reported, for example. And the better studies, the larger studies, were more likely to find an effect than those that didn't use those kinds of objective computer-validated measures of behavior. So, yeah, it has been hard to tease out. But besides the California assessment, there have been eight or nine independent scientific

12:48evaluations of the evidence that also conclude that eliminating synthetic food dyes can help with behavior problems for some children. But I also completely agree that it's not like these are life-saving drugs where you're willing to tolerate maybe some side effect or adverse effect to get the benefit you want. These are completely unnecessary chemicals. You can use real colorful fruit and vegetables to color the food. You don't need to have these synthetic food dyes at all.

13:22As someone who has taste tested them, I will, I'll say I do like the synthetic dyes more. But I'll, but I mean, as we know that, as Marianne, you said that psychology, they're teaching you to like those. Yeah, you should do that double blind where you can't speak.

Ultra-Processed Foods

13:36Yeah, exactly. So, I think that brings me to these other things that have been getting a really bad rap lately, which is the term ultra-processed foods. And that term gets thrown around all the time. So, is there sort of a scientific consensus on what that even means? And are all ultra-processed foods inherently bad? Because like, when I think of ultra-processed foods, I think of the worst of the worst. But there's a lot of these things that are, you know, they're actually super clean, super healthy

14:08bars and stuff. Not a lot. I disagree. Not a lot. The concept comes from Brazil, and it's a relatively new concept from 2009, where ultra-processed foods are defined as a specific category of foods that are industrially produced, refined with industrial additives. You can't make them in your home kitchen because you don't have those additives. You can't get them. They're designed to be irresistibly delicious, if not addictive. And they're designed for profit, and they're very effective at that.

14:40The definition covers things like commercially produced chips and bars and all those kinds of things. And the concept has been heavily criticized as excluding healthy foods. But if you push on what they mean by healthy foods, it turns out to be commercially produced whole wheat bread, commercially produced yogurt, plant-based milk, plant-based meats, and infant formula, which I think goes into a separate category of its own. So there aren't a lot of exceptions. But the food

15:14industry has focused on the exceptions because the concept is really easy to understand. And there are now thousands, literally thousands of studies that have associated ultra-processed foods with poor health outcomes. And there are now two extremely well-controlled clinical trials, very short-term, but done in metabolic wards where people were locked up and couldn't lie or cheat about what they were eating, that show that these foods induce people to eat a lot more calories without realizing it. So I think

15:48it's a powerful concept. If you want to lose weight, don't eat ultra-processed food. And it's not very hard to figure out what it is. I see. So from, because this is a more data science podcast, so thinking a little bit more quantitatively, when you mentioned, you know, when Liberty mentioned ultra-processed food, you mentioned the minimum processed food before we talk about these additives. Are any of those things have any kind of a quantitative definition? Like if you're adding this much, then the concept is kind of very

16:22fluid. Depends on who interprets. No, you have to read the ingredient list.

16:29I see. It's based on the ingredients. I see. But is there some national standards that's there to consider something? Not that I'm aware of. I mean, are you asking if there's one bad ingredient, does that make an ultra-processed? I think the answer is yes. But usually there are lots. I mean, I got to ask this though. You know, I mean, critics of these laws would say that there isn't real empirical data and evidence behind this stuff, but they're really precautionary.

17:00And so, you know, I mean, are there other instances? And I feel like there's a lot of public fear around these food dyes, preservatives, ultra-processed food. It's all over social media. I mean, it's really come to a forefront, you know, kudos to the movement because they've done a great job of really bringing this conversation to the forefront. But I mean, are there other times where sort of fear has driven policy like this?

17:24I don't think we've ever had anybody in government who was concerned about all of our attitudes. I mean, it's hard to get my head around it. You know, when President Trump introduced RFK Jr. as his candidate, he talked about the industrial food complex. I nearly fell off my chair. You know, where did that come from? And here's RFK Jr. talking about how the food industry is poisoning America. I mean, this is astonishing. Although his idea of poisoning America and mine

17:57may be somewhat different. As I said, I don't think color adenance is the most important issue in child health. I think calories are the most important issue in child health and that and marketing into children. I assume this issue is not a U.S. issue. So what are the studies in other countries? Do other countries have these issues or do other countries using these diets, preserves? And is there any kind of a, you know, cultural difference? What do we know beyond the United States?

18:29Yeah. So in Europe, there are warning labels on foods that contain certain synthetic food dyes, including red 40, yellow 5, and yellow 6, which are the top three synthetic food dyes in the United States. And in Europe, the foods say that the dye may have an adverse effect on activity and attention in children. And as a result of that warning label, most manufacturers reformulated their products just to eliminate the dyes. Because, you know, if you're a food

19:00company, the last thing you want to do is put a warning label on your product. So they just reformulated. But they sell the exact same products in the United States, which they may have reformulated for the United States also, or they may not have. Gradually, companies have been reformulating also for the United States. But, you know, it used to be that you could go to a McDonald's in the UK, get a strawberry sundae, no synthetic dyes, but buy one here. And there was. Now McDonald's has reformulated. And, you know, some companies, when they reformulated,

19:36nobody noticed. Like, for example, Kraft reformulated its iconic macaroni and cheese to eliminate yellow 5 and yellow 6. And they ran an ad campaign that said, you know, we changed the formula and nobody noticed. And it's still that bright neon orange. So, you know, there have been instances, I think, where consumers have noticed a difference, but it depends on the product. For many products, you can reformulate and not even see any difference. You know, in West

20:10Virginia, which is where I live, it was the first state, you know, sort of passed legislation to ban synthetic dyes statewide. And it gave people a lot of questions about, I mean, there's a lot of data that shows that this is really going to disproportionately affect lower income populations who really rely on shelf stable, affordable foods. A lot of these food companies have said that they're going to just have to pull their products off the shelf. What's going to happen next year when these are banned and people can't afford food?

20:42They're not going to pull the products because the products make too much money. They're going to reformulate the products just like they do in Europe. They're selling the same products in Europe. And, you know, the dyes are just such a teeny, teeny, teeny fraction of the cost of making a food. It doesn't make any cost difference to take out the dyes altogether or to substitute something else. They like to use dyes because it's easy. You know, you always get the same color.

21:14It's just easy to do. It might be slightly trickier to use some of the naturally derived colorings. But again, it's not going to affect the cost of the product, which is mainly marketing and a lot of other things, you know, the primary ingredients. It's such a tiny amount of dye that you really can't plausibly argue that taking the dye out is going to increase the price of the product. I just don't buy that. No, I think it's probably exactly for the reason it costs almost nothing to use it, but it has

21:49benefited on a cell. That is a stronger incentive for the industry to use this, at least for those profit oriented people. Right. So yes, you're absolutely right. Yeah. To simply use those formulations, recipes that they have developed for other countries and just start using them here. So it's not, it's not rocket science and we can do it and we can, you know, provide those products. Plus, you know, remember the ban doesn't go into effect in West Virginia immediately. I believe it's not until 2027 that that ban takes effect. So there is a little bit of

22:24time to make the adjustment. It's different for schools. It happens sooner for schools. For schools, it's right away. That's right. But now the foods that are subsidized by the government, very few of those contain any synthetic food dyes. I mean, barely none. There's plenty to choose from that don't contain synthetic food dyes. So this has been like really looked at in other states like California. And, you know, it was concluded that this really is not going to be a problem for schools to make that switch. I think that brings us to our magic wand question. And we ask everyone

Magic Wand Question

22:58a magic wand question at the end of every episode. So our magic wand question is if there was one thing you could either make people do or you could ban, what would it be? So for example, my belief is that one of the biggest issues is weight in the United States. You know, if you gain, what is it? If you gain a pound of excess body weight annually over 15 years, you, I think you increase your cancer risk by 40% or something. So for me, it would be that everyone had to be a certain weight, you know,

23:29weight limit for everybody if they could, if they could do that magically for public health concerns. So what would be a magic wand for you? Either something you make everyone do or that you ban? Magic wand for public health. Two things. Get money out of politics and stop marketing and ban marketing and drugs from children. I love that. All right. Lisa, what's yours? I mean, I'll go with Marion's. The only thing I'd add is that, you know, it's really not that

24:02complicated. We should all be eating like half of our plate should be fruits and vegetables and the other half can be a combination of whole grains, nuts, dairy products, lean meats, etc. So if we could somehow help people to make that happen, that's what we all need to be eating rather than putting all this effort into these packaged foods, ultra processed foods that are really no good for us. So help make those more accessible, easier to find, to purchase, more available. That would be my

24:36magic wand.

24:40Well, thank you to both of you for these very informative sessions. And we learned quite a bit. And I did not realize all these things really has no benefit whatsoever. And so, which I thought it would make it taste better, but it doesn't seem to be the case. And, but I hope they have something natural. Don't knock my fruit loops, Shally. They psychologically taste better. All right. I have to run a real double blunt experiment on you to figure out whether you really like the, oh, it's just the psychological effect, which is important. I'm not saying that's

25:14not important, but it's thanks to both of you. And thanks for the work you do to ensure that we're all healthy, particularly for our children. Thank you very much. I'm Liberty Vitter Capito, and on behalf of Shally Meng and our guests, thank you for joining us. And a special thanks to our producers, Rebecca McLeod and Tina Tobey-Mack, and assistant producers, Arianwen Frank, Gavin Yang, and Belle Reilly. This was the Harvard Data Science Review.

25:45Everything data science and data science for everyone.