Artificial sweeteners appear in the news almost constantly. If we believed the headlines, they would be a huge threat to our health, supposedly much more dangerous than the sugar they replace.
Read more Trump signed a new sanctions law against Russia
However, this is not always confirmed by scientific evidence. Still, we like to hear such messages, so they are reported almost every week, regardless of what the data actually show.
The latest hype about artificial sweeteners is a perfect example of this. For several months, headlines have claimed that artificial sweeteners accelerate cognitive decline.
According to a recently published study, there are six such “dangerous chemicals,” including commonly used sweeteners like xylitol and aspartame, which supposedly can lead to faster deterioration of our brain functions.
Hundreds of media outlets reported on this study. Although the study was published in September 2025, it still appears in headlines in the second half of 2026. All this is especially surprising because the study actually does not show at all that artificial sweeteners cause any problems.
The study itself is a very simple observational study.
The authors used a cohort of Brazilian civil servants called the ELSA dataset. It is a long-term study in which, over time, brain function was assessed using a standard set of tests.
At the beginning of the study, participants also completed detailed questionnaires about their lives, including the so-called food frequency questionnaire. It shows what people say they eat, not what they actually eat.
This is important – we will return to this.
The authors examined the relationship between reported sweetener consumption – calculated based on those food frequency questionnaires – and cognitive decline, which was assessed using four different scales.
Over eight years of the study, it was found that people in the third who consumed the most sweeteners experienced slightly greater brain aging compared to people in the third who consumed the least sweeteners.
This study faces the usual problems typical of such observational studies.
One of the main problems is that food frequency questionnaires do not actually measure precisely how much food a person consumes. They show what people think they eat, but we know that we are often unreliable judges of our past habits.
Most such questionnaires ask about what a person ate over the past 12 months, but people rarely can remember their dietary habits that accurately for so long.
The analysis also assumes that people’s diets did not change during the study.
The authors could rely on only one food frequency questionnaire, so if people’s dietary habits changed during this period, they had no way to detect it.
As noted in letters to the journal, there are also serious questions about the biological plausibility of the claim.
Some sweeteners are natural substances, one of them is actually a very rare form of sugar, and all their molecules differ greatly. Therefore, it would be very unlikely that all these different chemicals would affect the brain in the same way.
There is also the problem of reverse causality – meaning that people whose brain functions had already begun to decline due to diseases such as diabetes or heart disease at the start of the study might have been more likely to consume artificial sweeteners.
This problem is difficult to correct in the type of analysis presented in this study.
It is also worth noting that at least one of the analyses performed by the authors found no association between sweetener consumption and brain aging. In a series of statistical models presented in the supplementary appendix, the authors showed that when analyzing all cases, no association between sweetener consumption and brain aging was found.
This analysis includes all participants of the initial cohort, while the most publicized analysis excluded some people for various reasons, such as those with Parkinson’s disease or those reporting unusually high calorie intake.
Read more Lithuanian 3×3 youth team – in the semifinals of the world championship
