Are Artificial sweeteners turning sour?

The World Heath Organisation has recently upgraded their assessment of the artificial sweetener aspartame as possibly being carcinogenic. It is now classified as level 2b on the International Agency for Research on Cancer 4-point scale. To put this into context, level 1, the highest level of certainty, is occupied by smoking, alcohol and radiation, and level 2a- probably carcinogenic, is occupied by red meat consumption and night shift work.

Aspartame has often been regarded as a controversial substance, partly due to its handling by our metabolism which results in the formation of formaldehyde and methanol, albeit at tiny concentrations.

There are number of meta-analyses and reviews on the heath risk and benefits of non-sugar sweeteners (NSS) including an extensive World Health Organisation Review carried out in 2022. There are several other excellent reviews, though some have been funded indirectly by the soft drink industry.

This is a difficult area, particularly as the potential effect sizes are tiny when compared with cigarette smoking, for example which, can increase the risk of lung cancer by 2500%, whereas most of the potential effect size through NSS consumption, may be a matter of only 10% or more.

Most data can only come from observational data. Randomised trials have been undertaken but can only be short term in nature. No individual can spend their whole lives in a nutritional randomized study. There has sometimes between a contradiction between the short-term randomized control findings and the long-term findings when it comes to weight loss for example, which seem favourable for NSS in the shorter-term randomized control studies.

It may not clear from studies whether any observed effects are mediated by other factors such as volume of fluid intake, total calory consumption, sugar consumption and other lifestyle and behavioural factors. Researchers are well aware of possible confounding factors, but it may be difficult to control for all them outside of a randomized control trial.

When looking at the WHO statistics for cancer for all Non sugar sweeteners, there is a risk in assessing multiple cancers simultaneously, which may yield spurious findings by chance through the effect of multiplicity.

In medicine, if we find the probability of a study finding to have arisen by chance to be less than 1 in 20 we conclude that it was not due to chance, but due to a real effect. However, if we ask 20 questions than we might expect to get a spurious outcome in one of the answers.

The incidence of two cancers were increased in those with NSS consumption; bladder and laryngeal cancer, however the incidence of 2 cancers were found to be reduced; lung and ovarian cancer, though in single studies. It is likely that no one will believe these latter two findings to be genuine partly as there no known mechanism of action and secondary it is perhaps human nature to dwell on the risks. There also seems to be an increase in obesity related cancers form one cohort study. Of all the cancer, bladder cancer is the most studied and the pooled date suggests that there may be a 30% increased risk from NSS consumption. To put this into context, in the UK, the annual risk of developing bladder cancer is about 1 per 1000, if the effect size is correct form the pooled studies, then this could increase to 1 per 700 per year.

In the latest summary guidance, the WHO has stated that there may be an increased risk of hepatocellular cancer particularly from aspartame.

So where does this all leave artificial sweeteners? Will our taste for artificial sweeteners turn sour?

Certainly, there is a lot more epidemiological evidence to be gathered.

If we compare artificial sweeteners to smoking than it is unlikely that anything like the level of harm that is caused by smoking will ever be discovered. However, the situation with NSS is very different to smoking in that they were conceived to confer widespread health benefits by virtue of reducing calory consumption, and smoking is a completely unnecessary activity, whereas fluid hydration is most definitely not. If these health benefits are not acquired or if they prove detrimental to health, then it is likely that many will be dissuaded from consuming them or return to natural sugars, not that that would pose any problem for the soft drink industry. It is also worth emphasising that sugar consumption isn’t good for the health and there are some studies which suggest a role in carcinogenesis.

Another disappointing feature of the epidemiological data, is that NSS seems to be associated with increased obesity, hypertension and stroke in the long run but no one is sure if this is genuine causality or confounded by other behavioural factors which were not accounted for in these studies.

It is unlikely that NSS will be banned particularly when considering that cigarettes and alcohol, which are far more damaging are not prohibited, but there maybe changes in practice due to people’s expectation, our understanding of the risk, consumer forces and the potential development of newer alternatives. Rather we will need to make our own decisions.

It is well known that calory restriction can potentially increase lifespan. It makes sense that if we give our metabolism less (unnecessary) work to do that it will fare better in the long run and perform faultlessly for the long run. But we can only take shielding ourselves from the environment so far and will be motivated by many things that lead to living a fuller life. I think it’s reasonable that we should expect at least a little flavour in our drinks from time to time.

References 

1) Summary of findings of the evaluation of aspartame at the International Agency for Research on Cancer (IARC) Monographs Programme’s 134th Meeting, 6–13 June 2023

2) IARC Hazard classification 

3) Health effects of the use of non-sugar sweeteners: a systematic review and meta-analysis

4) Non-sugar sweeteners and cancer: Toxicological and epidemiological evidence

5) Dietary glycemic index, glycemic load and cancer risk: a meta-analysis of prospective cohort studies