In Defense of Artificial Sweeteners

F. Perry Wilson, MD MSCE
4 min readMar 16, 2022

A new meta-analysis suggests that, in terms of controlling weight at least, diet drinks are better than sugary drinks.

When I counsel patients who are trying to lose weight, there is something I always discuss — “Don’t drink calories”. The idea is that it is so easy to consume sweetened beverages (and alcoholic ones for that matter) and we don’t really get a sense of how many calories we’re taking in.

Some patients balk at the idea, saying they can’t stand the taste of water or just can’t bring themselves to drink it. While, as a nephrologist, this pains me deeply to hear, I often suggest going for low or zero-calorie flavored drinks instead of the sugary stuff.

And yet… I need to admit that recently I’ve been more nervous about that advice. A very nice study in Nature, for example, found that artificial sweeteners induce glucose intolerance and weight gain — in mice.

Several observational studies have suggested that the use of non-nutritive sweeteners — sucralose, aspartame, and so on — are associated with higher body weight and type 2 diabetes. Of course, observational studies in this space are tricky — are people gaining weight because they are drinking so called “diet” soda, or are they drinking diet soda because they are gaining weight?

Arrow of causality, turn turn turn, tell us the listen we should learn

Randomized trials, as ever, are they key to deeper understanding, but most trials in this space are relatively small. That makes a good case for this study, appearing in JAMA Network Open, which combines data from 17 randomized trials to determine what effects substituting sugary drinks with low- and zero-calorie drinks truly has.

So — what’s the bottom line? Should I ditch the Splenda in my morning coffee and drop in some sugar cubes?

It turns out the effects of drinking low-or zero-calorie drinks instead of sugary ones is modest, but overall beneficial, depending on the outcome you’re trying to achieve.

Randomized trials show that switching to low-cal drinks reduces body weight by about a kilogram, and BMI by 0.3 points. It also reduces body fat by about half a percent.

Effects on glucose homeostasis — hemoglobin A1C level and fasting glucose were not that impressive though.

The authors also compared sugar-sweetened beverages to plain old water drinking. I expected this analysis to show more dramatic benefits. After all, we’re all just ugly giant bags of mostly water.

Interestingly, the effects of switching to water were not as dramatic — and largely non-significant with respect to most outcomes evaluated.

So what do we make of this? If someone is a habitual drinker of sugar-sweetened beverages, is it preferable to switch to a zero-calorie flavored drink compared to plain water?

One possibility is that, in the trials where people are randomized to switch to water they aren’t as adherent. Just because we ask someone to drink water doesn’t mean they do it, and so there may be a tendency to “cheat” with sugar-sweetened beverages. However, if told that low- or zero-calorie flavored drinks are ok, maybe it’s easier to stick to the plan? This is essentially the argument you get from people who say that vaping is a good way to quit smoking. It may or may not be true.

It could also be that we just don’t have enough rigorous data to make a firm conclusion. Of the 17 trials examined, only 3 of them used water substitution as an intervention.

All-in-all, this data provides some reassurance that the zero-calorie sweeteners aren’t secretly exacerbating the obesity epidemic. I’d certainly rather my patients drink diet coke than regular coke. That said, these studies are necessarily short term — the longer-term effects of sugar substitutes, while perhaps not as bad as the long-term effects of sugar, must necessarily be worse than the long term effects of drinking water. Maybe this is the nephrologist in me talking again, but I doubt that there could possibly be a fluid better for the human body that good old H2O.

Except coffee, of course.

A version of this commentary first appeared on



F. Perry Wilson, MD MSCE

Medicine, science, statistics. Associate Professor of Medicine and Public Health at Yale. New book “How Medicine Works and When it Doesn’t” available now.