A new systematic review published in the International Journal of Obesity looks at the totality of evidence investigating whether consuming low energy sweeteners (LES), such as aspartame, sucralose, or stevia, is a net benefit or detriment for weight control. In addition to providing some clarity on the answer, the review also provides some insight into how different kinds of evidence address such questions.
LES vs sugar or water
The question may at first seem obvious – if you consume a beverage sweetened with an LES instead of sugar you will be avoiding up to several hundred calories. Those calories add up quickly. Twelve ounces of cola with sugar has 138 calories, for example. These are foods that people generally consume on a daily basis, and so avoiding these unnecessary calories should reduce total energy intake and help reduce or maintain weight.
The human body, however, is a complex system, as is human psychology, and so we have to consider the law of unintended consequences. It is possible, for example, that when people drink a diet beverage they feel they have earned the right to consume more calories elsewhere. This phenomenon is called compensation, and there is evidence for this effect.
It is also possible that consuming a food or beverage that tastes sweet but contains no or few calories will trick or confuse the brain, separating the sensation of sweetness with the ingestion of calories. This may lead to a craving for more calories. There are also studies showing the existence of sweet receptors in the GI tract, and activating these receptors may stimulate appetite.
Types of evidence
There are basically three types of research that address the question of whether or not consuming LES results in a net increase or decrease in total caloric intake. The first is animal studies, mostly in rats. Rats can be fed water sweetened with sugar or an LES and then given access to as much food as they want to eat (ad libitum).
The advantage of rat studies is that the experimenters have total control over the animals. They can record everything they consume, control everything they do, and record the results. The disadvantage of rat studies is that rats are not people.
There are two types of studies of LES in humans, cohort studies and controlled experiments. In cohort studies groups of people who consume or don’t consume LES are followed with their caloric intake and weight tracked. The advantage of such studies is that they are plausible to do for a long period with many subjects, and they are a real-world observation. The big disadvantage, however, is that it is difficult to make any conclusions about cause and effect. Specifically, if people who are overweight consume more LES, do they consume LES because they are trying to lose weight, or are they overweight because they consume LES?
The best type of data are controlled experiments in which people are randomized and blinded to either LES, sugar, or unsweetened water. The advantage to this type of data is that you can make confident conclusions about cause and effect. The disadvantage is that they are more difficult to conduct, and therefore may be small or short term.
The systematic review
The authors of the current review looked at all three types of studies. It was their intent to do the first systematic review looking at the totality of evidence addressing this question. Here is what they found:
Rat studies generally show no advantage to LES in terms of calorie intake, and some even show a disadvantage:
In 62 of 90 animal studies exposure to LES did not affect or decreased BW. Of 28 reporting increased BW, 19 compared LES with glucose exposure using a specific ‘learning’ paradigm.
For the animal studies the bottom line is that most showed no effect on body weight or energy intake. Some of the studies did show a decrease in weight and caloric intake, but these studies tended to use doses of LES 10-100 times what a human would typically consume.
The “learning” paradigm is an interesting subset of studies, some of which do show increased weight in rats fed LES vs sugar with a moderately sweetened feed. The authors conclude that there does appear to be an effect here, but the problem is its applicability to humans. The experimental setup is contrived, exposing rats to a food source they are not used to. One hypothesis is that the LES sweetened water may make the food more palatable to the rats, so they consume more.
Incidentally, most often when I see a headline declaring the LES cause weight gain, it is this type of animal study. Let’s get to the human studies to see if these rat studies are likely to be applicable.
Twelve prospective cohort studies in humans reported inconsistent associations between LES use and body mass index.
In total there were six comparisons in adults and five in children. The random effects model showed no change in BMI with LES consumption. However, there was a high level of heterogeneity among the studies.
The bottom line with these types of studies, it seems, is that they are a wash. Results are all over the place, consistent with either random noise or various confounding factors that differ from study to study.
Human experiments were divided into short and long term studies. The short term studies looked at LES consumption on the caloric intake of a single meal. There is some variability in these studies depending on specific methods used, but the authors give this bottom line:
In sum the results of these short-term studies comprise a large body of evidence showing that consumption of LES in place of sugar reduces overall EI [energy intake] acutely, with no indication that LES increase appetite.
So short term there is an advantage to consuming LES vs sugar, but this leaves open concern for long term compensation effects. The authors point out that these short term studies do show compensation, meaning that consuming LES was associated with consuming more calories in the following meal, but those calories were less than the calories avoided by not consuming sugar.
The long term studies were mostly in overweight or obese subjects, followed subjects from between 10 days and three years, were mostly in adults, and some were in the context of a weight loss diet. The authors found:
In all cases, the use of LES led to a relative reduction in EI, and greater loss (or reduced gain) of BW. Notably, there was no example of a sustained exposure intervention trial where LES use led to a relative increase in mean EI or BW. This was supported quantitatively by the results of the meta-analyses of BW change, indicating lower relative BWs in LES intervention arms. Furthermore, outcomes were similar in studies with children and adults, and followed a similar pattern whether participants were blinded or not blinded to the intervention. Consumption of LES-sweetened beverages also reduced BW relative to consumption of water.
The results are fairly consistent, differing only in the magnitude of the effect – consuming LES vs sugar or water was associated with a reduction in energy intake and body weight. I was a bit surprised about the advantage vs water, as water is also zero calorie. This implies that drinking a sweet beverage vs an unsweetened beverage allows for lower overall calorie consumption.
Conclusion: Low energy sweeteners probably help with weight control
The results of this systematic review and meta-analysis are quite compelling. In my opinion, human experimental studies trump observational studies, studies in animals, or basic science studies looking at possible mechanisms.
In this case, despite confusing results from many types of studies, the best evidence consistently indicates that consuming LES results in reduced energy intake and reduced weight compared to either consuming sugar-sweetened beverages or even water. There does appear to be room for further rigorous studies, but this is the best conclusion we can currently make based upon all the evidence, and it is fairly consistent and robust.
The story of research into the question of low energy sweeteners is also a massive cautionary tale for the average consumer. The media, self-help books, and nutrition and alternative gurus have been handing out conflicting recommendations to the public, largely based on extrapolation from possible mechanisms, animal studies, and observational studies.
In retrospect, and not only with the question of LES but with many health questions, these types of studies should be considered with a heavy dose of skepticism. There is a reason why skeptics and proponents of SBM don’t find such studies definitive, because history has shown they are often misleading.
What is compelling is controlled blinded experiments looking at the relevant net health effects of an intervention. Until we see consistent results from rigorous studies of this type, it is best to reserve judgement. I don’t find extrapolations from basic science or animal studies to be compelling – history has shown they are simply not reliable. However, such studies do serve a useful purpose in designing clinical trials and informing our choices about which interventions are worth studying.
Unfortunately, wild extrapolation from basic science is common in nutrition self-help books. Diets are justified based upon studies showing the effects of insulin, or the presence of receptors. The human body, however, is simply too complex to be able to draw a straight line from these starting points to a net clinical effect. There are simply too many compensatory mechanisms and feedback loops in biology.
We always need clinical studies looking at net health outcomes, and they need to be rigorous and repeatable.
In the case of LES and weight control, the clinical studies are giving us a consistent answer. Initial common sense actually holds up in this case – not consuming extra calories in the form of sugar-sweetened beverages or food is a good way to reduce calorie intake and manage weight.