Study Says Sweeteners Prompt Food Cravings? Nope.

Reporting on non-caloric sweeteners has more in common with reporting on religion than science reporting. We see a constant churn of reporting about studies by people who are looking for evidence to support a belief that sweeteners must be bad for you. The latest headline from NPR on this subject tells us these sweeteners “may prompt food cravings, especially in women and people with obesity.”

But there’s one little problem. The study Allison Aubrey is reporting on did not actually measure food cravings. The researchers looked at brain scans, but not food cravings. In fact the study doesn’t say a single word about food cravings. Nor did the research even look at weight gain. It was the study of neurological responses after a single exposure to water – plain, with sucrose, or with sucralose (Splenda®).

A Randomized Crossover Study

The study was not a bad design. It was a randomized crossover study. That means that everyone got each of the three exposures: water plain, with sugar, or with sucralose. The objective was to see what differences they could find in neurological responses, metabolic responses, and eating behaviors after each of these exposures. To observe eating behaviors, researchers served subjects a buffet of food after the experiment was seemingly over. They recorded precisely what each participant chose to eat.

Brain Scans, Yes; Eating Behaviors, Not Really

The results did show some difference in brain scans – for women and people with obesity. But when it came to eating behaviors, they did not find much. Buried in Supplement 2 under eAppendix 2, you will find the following:

“Participants consumed fewer total calories (β=-2.00, 95% CI: -2.84 to -0.68, p<0.001) following sucrose vs water. There was no difference in ad libitum total caloric (β=-0.63, 95% CI: -1.62-0.52, p=0.25) intake after the sucralose vs water conditions.”

This appendix also notes that “females consumed fewer total calories” after sucralose than after water. This inconvenient detail does not find its way into the paper. Perhaps that’s because it’s not really a significant finding. Or perhaps because it also gets in the way of the narrative the paper presents – an increased response to food cues after sucralose.

Perspective

In a tweet about this study, Tamar Haspel notes that eating behaviors might be more important than brain scans – if you want to know whether sweeteners lead a person to eat more. She calls this thread “BORING” and “WONKY.”

For a dispassionate view, we turn to the invited commentary by Stephanie Kullmann:

“Before recommending or discouraging the use of NNSs [non-nutritive sweeteners] as part of a healthful diet, further studies of the effects of NNSs on a variety of neurobehavioral and metabolic outcomes are warranted.”

In other words, this is interesting research, but it tells us nothing about the real-world effects of sweeteners on real people in real life. It’s all about brain scans and neural responses. It serves up clues about directions for more research.

But none of that goes into stories by reporters looking for issues with sweeteners. If you want clicks and listeners, you need to say there’s a problem. These sweeteners “can leave people with increased food cravings,” says Alison Aubrey in her report for NPR. Nevermind that the researchers did not actually measure food cravings. A pesky detail.

Click here for the research, here for the invited commentary, and here for Aubrey’s reporting.

Splenda, photograph © Austin Fausto / flickr

Subscribe by email to follow the accumulating evidence and observations that shape our view of health, obesity, and policy.

October 9, 2021