We would like to have our cake and eat it too. In a sense, that is just what non-nutritive sweeteners such as aspartame, sucralose, and stevia promise. Cut out sugar, a dietary villain, and replace it with a non-caloric sweetener and reduce your chance of putting on the pounds! The claim sounds good if it really works, and of course, if the sweeteners themselves don’t introduce a problem.
Although numerous studies have attested to the safety of such sweeteners, there have also been some that have raised the spectre of risk. As a result, we have been privy to a long-standing, smoldering controversy. Now a study published last year in the Canadian Medical Association Journal has added fuel to the fire, concluding that replacing sugar with non-nutritive sweeteners does not improve weight management. Furthermore, the use of such products is associated with high blood pressure, type 2 diabetes, heart attack, and stroke — collectively referred to as “cardiometabolic factors.” Sounds scary, but there is no need to panic.
This study was actually a “study of studies,” or a “meta-analysis,” meaning that the researchers scrutinized the published literature for relevant studies, hoping that the combined data would allow for a more meaningful conclusion than any individual trial would permit. Seven of the studies were randomized controlled trials (RCTs), and 30 were “prospective cohort” studies. In an RCT, one group of subjects is given the substance in question and the other is not, with neither the subjects nor the experimenters knowing who is in which group. After a set period of time, the health status of the subjects is analyzed for any difference between the groups. A “prospective cohort study” focuses on a group of similar individuals — defined as cohorts — who differ with respect to the factor being studied. In this case that factor was the consumption of non-nutritive sweeteners, which was followed for years to determine if it affected their condition in any way.
RCTs are acknowledged to be the “gold standard” in science, which in this instance showed no significant difference in the Body Mass Index (BMI), body weight, waist circumference or insulin resistance between the experimental and control groups. However, these trials were of relatively short duration and did not look for cardiovascular effects.
The cohort studies, on the other hand, lasted longer and revealed that non-nutritive sweeteners had an association with weight gain as well as with an increased risk of hypertension, heart attacks, strokes, and type 2 diabetes. Here, though, we have the issue of possible “reverse causation,” meaning that people who already suffer from these conditions are more likely to consume non-nutritive sweeteners. The researchers, however, have attempted to control for this by excluding studies in which the subjects had known, pre-existing conditions. It should also be mentioned that the consumption of sugar has also been associated with cardiometabolic risk, as well as with weight gain, so switching from non-nutritive sweeteners to sugar is unlikely to be a benefit.
What then do we make of this meta-analysis? The cohort studies cannot prove cause and effect, and the RCTs were short and did not evaluate cardiovascular risk. In the cohort studies no distinction was made between the different non-nutritive sweeteners which are very different in their chemical structures and are unlikely to have the same effects on the cardiovascular system. For example, the plant extract stevia is significantly different chemically from aspartame or sucralose. It is also possible that subjects who consume non-nutritive sweeteners then reward themselves by eating foods they would otherwise have shied away from and so this part of their diet is responsible for the observed negative effects.
Publication bias also has to be considered, meaning that negative studies about sweeteners are more likely to be published than studies that show no effect. Another problem is that the consumption of sweeteners varied greatly over the studies and there is no way to tease out any dose-response relationship. That is an important consideration because the key to evaluating the toxicity of a substance is to see if risk increases with dose. It is possible that guzzling large amounts of sweeteners poses some risk but moderate amounts are innocuous.
Taking all this into account, it would be inappropriate to conclude that non-nutritive sweeteners cause diabetes and cardiovascular disease. But admittedly, it is somewhat unsettling that a number of cohort studies have found such an association. Previous studies have suggested that non-nutritive sweeteners can alter the balance of bacteria in the gut and that this can have various physiological consequences, a possibility that needs to be further explored. So, while the link to cardiometabolic health is inconclusive, the conclusion that non-nutritive sweeteners do not prevent weight gain seems sound. If there are no significant benefits from consuming these substances, why do so? Water is a very good alternative to soft drinks, plain yogurt can be flavoured with fresh fruit and good coffee doesn’t need any sort of sweetener.
Joe Schwarcz is Director of McGill University’s Office for Science and Society, which is dedicated to the promotion of critical thinking and the presentation of scientific information to the public, educators and students in an accurate and responsible fashion.