Public health historian David Merritt Johns has written an unusually thorough and lucid piece in The Atlantic on this unlikely research phenomenon: a series of scientific studies shows a health benefit to eating ice cream. And the researchers have been trying—and mostly failing—to find flaws in the results.

What’s really going on here?

Nutrition health research is fiendishly difficult. It’s prone to every imaginable kind of bias, ranging from data reporting biases of the test subjects, experiment design biases, researcher biases, cultural biases, population selection biases, and on and on.

Unfortunately, we’re unable to (for example) grow 100 diabetic humans in a petri dish under identical conditions, inoculate 70 of them with ice cream, 30 of them with a perfect ice cream placebo, and observe the results for 40 years. We’re stuck with imperfect methods, whose strengths and shortfalls Merritt describes briefly but well.

Science, fortunately, offers robust tools to help control for bias. These tools let us go back to the data with hypotheses like, “maybe the people who don’t eat ice cream are actually less healthy because many of them have pre-existing problem; they don’t eat ice cream because their doctors told them not to.” This particular bias is called “reverse causation.” It’s one of the many that the researchers probed for. And yet: when they controlled for this one and others, the results did not go away.

What do the studies tell us?

Moderate ice cream consumption might reduce your chances of developing insulin resistance, type-2 diabetes, obesity, and heart disease. In these regards, the effects of ice cream appear equal to, or even slightly better than, those of yogurt.

What’s still Unclear?

• How certain we should be about these results
• If the results are causal and not merely correlative
• The mechanism of any positive health benefit

Is this fringe science?

No. Several groups of researchers, driven partly by normal scientific diligence, partly by fear of further muddying the dangerously muddy waters of popular nutrition science, have attempted to find flaws in the studies. They’ve scrutinized, probed for biases, and compared with older and newer studies. The results, while still not a statistical slam-dunk, keep holding up.

Nevertheless, some of the scientists directly involved in the studies are trying to keep the results quiet. They likely want to avoid the risks of overconfidently or prematurely refuting a long-standing orthodoxy. And they probably don’t want their results mischaracterized, to the detriment of public health, by lazy journalists or charlatans. Or, um … people like me.

What’s my take?

That this is interesting. That if your overall diet and lifestyle are healthy, and even if they’re not, you shouldn’t worry about eating some ice cream a few times a week. Which has always been my take.

I’m not going to go full-on clickbait and declare ice cream a superfood. In fact, if you call anything a superfood, even kale or spirulina, I’m going to stop reading. There is no ingredient (or even meal) that’s going to contribute much to your vitality or imminent demise.* It’s the sum total of your your diet and your habits that matter. Cheese steak and gelato are just small pieces of this picture. Unless for you they’re very large pieces—then you might have a problem. Even then, the genetic cards you’ve been dealt, for better or worse, probably overshadow most of these choices.

In Closing

If you’re familiar with my recipes, you’ll notice that I aim for a more intense experience—distinct, vibrant flavors, and a texture balanced to deliver those flavors fast and long. The hope is that you can substitute intensity—and quality—for quantity. If a single quenelle of ice cream can take you on a technicolor, ayahuasca-like journey into the deepest jungles of your mind, you probably don’t need a triple sundae.

So eat ice cream in good health. Make it the good stuff, so you won’t be tempted to eat too much. And as always, further research is warranted.

Selected Research Papers 

Jean-Philippe Drouin-Chartier, Yanping Li, Andres Victor Ardisson Korat, Ming Ding, Benoît Lamarche, JoAnn E Manson, Eric B Rimm, Walter C Willett, Frank B Hu, Changes in dairy product consumption and risk of type 2 diabetes: results from 3 large prospective cohorts of US men and women, The American Journal of Clinical Nutrition, Volume 110, Issue 5, November 2019, Pages 1201–1212, https://doi.org/10.1093/ajcn/nqz180

Chen, M., Sun, Q., Giovannucci, E. et al. Dairy consumption and risk of type 2 diabetes: 3 cohorts of US adults and an updated meta-analysis. BMC Med 12, 215 (2014). https://doi.org/10.1186/s12916-014-0215-1

Pereira MA, Jacobs, Jr DR, Van Horn L, Slattery ML, Kartashov AI, Ludwig DS. Dairy Consumption, Obesity, and the Insulin Resistance Syndrome in Young Adults: The CARDIA Study. JAMA. 2002;287(16):2081–2089. doi:10.1001/jama.287.16.2081

*For the sake of argument, I’m not counting mercury, benzene, botulinum toxin, etc. as “ingredients.”