Don’t buy the lazy nutrition science about artificial sweeteners

In 2023, the apparently authoritative World Health Organization (WHO) recommended that sugar-free sweeteners “not be used as a means of achieving weight control or reducing the risk of noncommunicable diseases.” Sugar-free sweeteners are found in diet sodas, yogurts and snack bars, and many other foods.

The guide was a surprise. After all, the very purpose of sugar-free sweeteners, which contain little or no calories, is to help consumers manage their weight and reduce the risk of disease by replacing sugar. Excessive sugar consumption is a major contributor to diabetes, obesity, heart disease, certain cancers, and a variety of other health problems.

In its report, the WHO cited evidence that long-term use of sugar-free sweeteners is associates with a higher risk of type 2 diabetes, cardiovascular disease and death. How are sweeteners such as aspartame, advantame, cyclamates, neotame, saccharin, sucralose and stevia linked to the negative health effects they are supposed to prevent?

There is nothing that fits causal explanation, but there is a decidedly human explanation: the science of lazy nutrition.

The limits of observational studies

The WHO made its recommendation after reviewing hundreds of published studies. The problem is that the vast majority of these studies are observational, in which researchers track behaviors and health outcomes in large populations. In these studies, subjects tend to self-report their food intake, a process fraught with inaccuracy. More importantly, observational studies cannot determine cause and effect. Do sugar-free sweeteners cause diabetes, or are people at risk of diabetes simply consuming them as they try to cut out some sugar? Finally, there are numerous variables that researchers cannot possibly control in these studies that could skew the results.

Randomized controlled trials (RCTs) tell a different story about sugar-free sweeteners. These rigorous studies control for confounding variables by randomly assigning people to a treatment or control group, and can determine cause and effect. They show that sweeteners modestly benefit weight loss and help control blood sugar, without the adverse effects seen in observational research. The disadvantage of RCTs is that they are shorter in duration, often only a few months. Well, negative effects I could appear after longer use and we could not tell from these RCTs.

But we also can’t tell from observational studies, which only measure correlation and not causation. Observational studies like this are why the public is regularly bombarded with advertising health stories linking virtually any food to a negative health outcome. Associations are easy to find and become a source of potentially viral news.

This state of affairs bothers Steven Novella, a Yale neurologist and popular science communicator.

“Making broad dietary recommendations based on observational or preliminary data causes ‘alarm fatigue’ in the public,” he wrote. “If every week they hear about another thing they’re supposed to avoid, it becomes impossible to manage.”

Encourage better research

One solution is for scientists to stop muddying the waters of knowledge with shoddy observational studies. In the case of sugar-free sweeteners, they could be pushing people to consume more sugar, which is arguably worse for health.

“Filling the research gap with studies that do little to help us understand the cause-and-effect relationships of our real-life dietary decisions does little to advance understanding; in fact, it sows confusion,” Drs. Anupam B. Jena and Christopher Worsham wrote last year in the News from New York. (Jena and Worsham are both professors of health care policy at Harvard Medical School.)

However, the status quo may be difficult to dislodge. RCTs are expensive and time-consuming, whereas observational trials are cheap and relatively easy to conduct. RCTs require recruiting participants, establishing and administering diet plans, and regularly measuring subjects’ health outcomes. With observational trials, researchers can simply download data from already available databases such as NHANES or the UK Biobank and look for correlations. Because academia regularly rewards quantity of publications over quality, the incentives favor observational essays.

For change to happen, it may have to start at the top, where science is funded. Government agencies such as the National Institutes of Health, which distribute billions in research grants, should begin prioritizing RCTs.

“We’ve wasted enough resources and caused enough confusion, and now we need to refocus,” John Ioannidis, a professor of epidemiology and population health at Stanford, said in an interview. “Funds, resources, and effort should be spread over fewer, better-designed randomized trials.”