JENA: Can I just ask a clarifying question? Is the United Kingdom the same thing as England? Is that the same thing as Britain or are we talking about three different things
KANFER: I believe the United Kingdom encompasses, Northern Ireland, Wales, Scotland, and England. But I will fact check myself on that.
That’s my producer Julie Kanfer and she was right about the United Kingdom, by the way. You might be wondering why we’re talking about this at all, and the answer has to do with a clever natural experiment.
We know Americans consume a lot of added sugar, on average around 57 pounds every year. What it’s doing to our bodies is a little less clear. But a new paper gets us closer to solving the mystery. It was published in December 2022 by the National Bureau of Economic Research and is appropriately titled “The Sweet Life.” The co-authors explore a relationship that’s difficult to measure, but one that’s vitally important, given how much sugar we all consume.
I’m Bapu Jena and this is Freakonomics, M.D. Today on the show: what are the long-term health effects of eating sugar? And, back to the United Kingdom: how can a food rationing program that took place there more than 70 years ago?
JENA: So Julie, before we jump into our conversation, let me give you some quick background. Here’s how the U.K food rationing program worked during World War II: People were issued coupons that entitled them to buy specific quantities of restricted foods. Things that were processed and had lots of sugar, but also meat, cheese, butter, bread. The government was trying to guarantee everyone fair access to certain items, and they were also trying to prevent shortages and starvation. Also, because Britain is an island nation, it imported most of its food, and those imports were under siege by the German navy and air force throughout the war. The rationing program continued for a few years after the war ended in 1945, as the U.K. was slowly recovering.
KANFER: And why did this study catch your eye? Cause it caught my eye too.
JENA: The basic idea was trying to use this national policy in the U.K. to figure out what is the causal effect of sugar intake on long-term health and economic outcomes. There were these restrictions that were lifted over time. So bread and flour in July of 1948, and then biscuits, jam, canned, dried foods, 1950. And then bang, September, 1953, sugar and sweets, they went off ration. It’s a really interesting historical twist of fate and the rationing provides a nice natural experiment to see what happens when people consume less sugar than usual for a fixed period, and then go back to normal sugar consumption. Question is, what kind of health effects do we see? I like that kind of stuff in general, but you could argue we think we know a lot about the impact of sugar on our health from decades and decades of studies. But what, I think most people don’t appreciate is that the level of evidence, while it is really enormous in terms of volume, is not typically very high quality.
JENA: What do I mean by that? So a lot of the research comes from animal models, which is useful to understand the physiology of how sugar affects the body, how diseases might result from increased exposure to sugar. But it’s a huge jump to say something about how sugar affects the human body, by just looking at small animals. The second thing is there are studies that look at randomization of humans to different types of diets. For example, high or low sugar diets or processed sugar versus not. But those studies tend to be really small, which makes them not very generalizable. And the follow up period is, typically very short. These people are not being followed for years or decades to understand what long-term effects are of sugar. And then the third bucket of studies are really these associational or observational studies where you simply look at people who have more sugar intake versus less sugar intake. Maybe you try to account or adjust or control for some other variables that are different between those two groups. And then you attribute any differences in outcomes, let’s say rates of diabetes, to the differences in sugar intake. But the people who eat a lot of sugar are different than the people who don’t. So how do you know that it’s really the sugar intake versus other fatty foods that they might eat versus exercise versus occupation? So it’s not the sugar that’s causing it. That’s what attracted me to this particular study. It was a nice natural experiment to get at the effect of one particular nutritional element, in this case sugar, on health and non-health outcomes over a long period of time.
KANFER: And we should really care about sugar for all the reasons that, you said. We know that there are links to health problems. But also people really like sugar. So the ration was lifted and sweets and chocolate sales increased by more than 150 percent in one year. There was a very dramatic reaction by the people in the U.K. who had not been able to access as much of these sugary snacks as they would’ve wanted to. As soon as they could, they really went for it.
JENA: And what’s really impressive about this study is they also got data from national food surveys from the 1950s, and they showed this large and sharp increase in sugar consumption from something like below 40 grams to more than 80 grams of sugar per person, per day. And then the next question is, all right, how do you use that shock , to tell us something about the long-term effects of sugar exposure on health outcomes?
KANFER: Because just identifying it is the first step, right? So then how do you figure out what to look for?
JENA: They basically looked at people who were born during rationing and they compared them to people who were born just after rationing. And then they looked at data from the English Longitudinal Survey of Aging. And these are people who are between the ages of 50 and 65, who were born in the years around the U.K. rationing. The researchers wanted to compare rates of certain diseases like diabetes and also markers of chronic inflammation in those two groups more than 50 years later. So they looked at data for each group between 2008 and 2018, those who were exposed to the sugar rationing, and those who weren’t.
KANFER: And Bapu, why would you want to look at someone’s early life exposure?
JENA: There is some evidence that our early exposures to different things — environmental factors, one of which could be sugar — not only affect the way our metabolism might develop, but might also affect the way that our genes are expressed. But it also might just change our habits. Suppose you see that early exposure to something like sugar or television creates this habit-forming behavior where you want sugar more later in life, or you’re more interested in watching T.V. later in life. Let me tell you a little bit about what they found They saw that the prevalence of diabetes was about 50 percent higher in that group that was exposed to more sugar when they were young. And, it wasn’t just diabetes, they found elevated rates of cholesterol arthritis. And they also looked at measures of laboratory markers of chronic inflammation. and levels were about 33 percent higher in people who had been exposed, to more sugar. And the other thing that is interesting is they didn’t find much difference in the total caloric intake of these individuals much later in life. This is people who had similar total calories, but more of those calories were coming from sugar. So it does speak to this broader question which is, what is the composition of calories that we get as opposed to the total number of calories, that we take.
KANFER: And really none of it is terribly surprising to this point. So it would’ve, I think, really blown everyone away if they were like, actually the people who suddenly had all this access to sugar were healthier. What else did they find though, because I think that the next bit is interesting.
JENA: There’s two other things that they found, which I think do advance the science, The first is that they also looked at, measures of human capital accumulation. We can think of that in economics as how much education you get and economic wellbeing. What they found there was that on average, the cohorts who were exposed to this period of greater sugar, they were less likely to complete post-secondary education by about 20 percent. They did not find any effect on the probability of being employed, but that exposure to sugar was associated with the reduced likelihood of working in a skilled profession and having above median wealth by the time you were, let’s say 50 to 65. If you believe these results as being causal, the argument would be that it’s related to the health behaviors that develop, the health outcomes that then develop, and the effects of health on education and wealth.
KANFER: Does this tell us anything new or do you think it kind of adds to an existing literature?
JENA: I think it probably adds to an existing literature because we know from other studies, how health affects education and employment and wealth. I think what was elegant here was that they focus again on a very simple thing, which is sugar.
KANFER: And just in terms of how people crave sugar compared to other nutrients, so sugar tends to be slightly more addictive
JENA: There’s actually studies that look at how the addictive properties of sugar compared to something like cocaine. Right? And anybody who’s got a sweet tooth would appreciate that.
KANFER: The reaction of my children when I take away dessert is evidence enough for me.
We talked about the creative way this study’s co-authors, Paul Gertler, and Tadeja Gracner, figured out the long-term effects of early-life sugar consumption. But how else could you try to answer that question, or similar ones about nutrition? And what other signals could you look for in this unique natural experiment? That’s coming up, after the break.
* * *
KANFER: Think about how else you could use this data, what else you would do with it.
JENA: I’ve got some more ideas if you’ll humor me.
There’s usually more than one way to answer a research question, and also, more than one answer. A lot of my time is spent trying to figure out the best approach. Because it’s possible that even with some really cool data, you could miss the mark entirely. So, what else could we do with the information from the U.K food rationing program?
JENA: I think they answered the big question here. I mean, they could look at other health outcomes besides ones related to chronic disease. For example, rates of A.D.H.D. might be different in cohorts who are exposed to sugar rich diets because of how, sugar affects the way that we think and behave. Maybe you would see higher rates of A.D.H.D. in adulthood among people who are exposed to these sugar rich diets when they’re kids.
It makes me think of whether or not you could use the exact same framework that they looked at to measure spillover effects. So, for example, do these preferences that young kids develop affect older siblings who weren’t affected by the rationing? They generate these preferences for sugar. Does any of that spill over into the older siblings? Because, for example, now there’s just more sugary food in the house. If that happens, then we’re actually understating the negative effect of sugar by looking at the estimates from their study. The other thing is there something intergenerational going on here? Are those adults who are exposed to high sugar diets, who now, in their fifties and sixties intake more sugar, what happens to their kids? So you could totally imagine that U.K. policy in 1953 might have long-term effects, But there’s at least three other ways that you could look at very similar sorts of issues in more modern day.
A lot of schools have vending machines, but some schools don’t. So you could imagine looking at variation across schools in access to vending machines, or even taking photographs of the vending machines to figure out what is the nutritional content of the things in those machines to look at how well that correlates with child health outcomes. And you could look at schools that introduce policies to allow vending machines versus schools that in the same period of time and the same area do not. You could look at kids that are exposed to taxes on sugary beverages versus not. And then look at the long-term effects of that exposure to a tax, which reduces your likelihood of using a sugary beverage perhaps
The last idea I had is imagine you had data on families in which one child was newly diagnosed with say, type one diabetes. It’s plausible to think that if a family has a child that’s diagnosed with type one diabetes, that they would then have less sugary foods in the household, right? Let’s say these are families with three kids. The middle kid has diabetes. The oldest kid is exposed to a sugar rich environment. Then, the middle kid is diagnosed with diabetes. The youngest kid then grows up in an environment where there are perhaps less sugary foods in the house. The prediction would be that there would be a gradient, in sugar related health outcomes between the first child and the third child in families where there is a middle child with type one diabetes. And you would see no such gradient in families in which the middle child does not have type one diabetes. I literally just came up with that. Are you gonna gimme like some credit for that or no?
KANFER: Well, I get — no, I, I give you credit of course, but it makes me think that you can use that with any food sensitivity or allergy. So a child has a gluten intolerance. How does that affect the diet of everyone else in the house or the other sibling? Or let’s say there’s a nut allergy? I think you should do some kind of study involving these approaches.
So, what about the study — called “The Sweet Life” — that kicked off this conversation?
KANFER: Let’s say you’re somebody who works in the nutrition field and you’re looking at this study. What are you going to think to yourself as far as making guidelines or how people should be forming their diets?
JENA: I think the specific information of this study may not move the needle in terms of our, cumulative knowledge and understanding about the impact of sugar. But to me, that’s totally fine. To be honest, the field of nutritional epidemiology it’s like almost like a nutritional wasteland. You can’t really infer anything from these studies because they aren’t causal at all, but they’re always in the news and, journals love to publish them. What I really loved about this study was that it introduces this way of economic thinking of using natural experiments to study the effect of a nutritional behavior on outcomes. And it allows us to study the effect, not just in the short term, but also longer term. So that’s where I see the value of this particular paper.
And I’ll just add this: You may have heard about another recent paper that linked the popular sugar substitute, Erythritol, to increased risk of cardiovascular events, like heart attack and stroke. I know I did, because it was all over the news. That research took a lot of different approaches, including analyzing animal data and clotting in blood samples. It also looked at rates of cardiovascular problems in actual human beings and while they did account for a handful of factors that predict cardiovascular risk, the research didn’t ultimately fully account for the fact that people who consume products with sugar substitutes are different from people who don’t. Their diets, exercise habits, incomes, and a lot of other factors likely vary. If you fail to consider those factors, you may get an unreliable answer about whether sugar substitutes are helpful or harmful. Because when the right methods are applied, these types of relationships can easily reverse. Anyway, that’s it for today’s show. I want to thank my producer Julie Kanfer for joining me, and thanks to you, of course, for listening. Does this new research make you want to change YOUR sugar consumption? How closely do you follow nutrition studies in the news? Send us an email at BAPU at Freakonomics dot com. That’s B-A-P-U at Freakonomics dot com.
Coming up next week on the show: Nobody wants to be in the hospital. If you find yourself there, though, could the slightest change — a sunnier room perhaps, or a softer pillow — make you feel better? Find out what the research says, and how we might reframe the way we think about these amenities. That’s next week on Freakonomics, M.D.
* * *
Freakonomics, M.D. is part of the Freakonomics Radio Network, which also includes Freakonomics Radio, No Stupid Questions, and People I (Mostly) Admire. All our shows are produced by Stitcher and Renbud Radio. You can find us on Twitter at @drbapupod. This episode was produced by Julie Kanfer and mixed by Eleanor Osborne. Lyric Bowditch is our production associate. Our executive team is Neal Carruth, Gabriel Roth, and Stephen Dubner. Original music composed by Luis Guerra. If you like this show or any other show in the Freakonomics Radio Network please recommend it to your family and friends. That’s the best way to support the podcasts you love. As always, thanks for listening.
JENA: As long as one listener does a study, you could — just, just put me in the acknowledgements. I’ll be — put us in the acknowledgements actually, this is — I credit you for this idea.
- “The Sweet Life: The Long-Term Effects of a Sugar-Rich Early Childhood,” by Paul Gertler and Tadeja Gracner (NBER Working Paper, 2022).
- “Outcomes Following Taxation of Sugar-Sweetened Beverages: A Systematic Review and Meta-Analysis,” by Tatiana Andreyeva, Keith Marple, Samantha Marinello, Timothy E. Moore, and Lisa M. Powell (JAMA Network Open, 2022).
- “Foods Sold in School Vending Machines are Associated with Overall Student Dietary Intake,” by Alisha J. Rovner, Tonja R. Nansel, Jing Wang, and Ronald J. Iannotti (Journal of Adolescent Health, 2011).
- “Evidence for Sugar Addiction: Behavioral and Neurochemical Effects of Intermittent, Excessive Sugar Intake,” by Nicole M. Avena, Pedro Rada, and Bartley G. Hoebel (Neuroscience & Biobehavioral Reviews, 2008).
- “Is Poor Nutrition a Supply Problem or a Demand Problem?” by No Stupid Questions (2022).
- “There’s A War On Sugar. Is It Justified?” by Freakonomics Radio (2017).