Skeptoid PodcastSkeptoid on Facebook   Skeptoid on Twitter   Skeptoid on Stitcher   iTunes   Google Play

Members Portal



Get a Free Book



Over-interpretation of studies in the anti-aspartame movement

by Josh DeWald

November 16, 2012

Share Tweet Reddit

Donate Despite thirty years of research demonstrating aspartame to be safe, and no scientifically plausible reason to be concerned in the first place, there continues to be a large community on the Internet, as well as in the general population, that insists aspartame is dangerous. There are even a couple of documentaries available (one of them Sweet Misery, I have discussed elsewhere). I would like to focus this brief article on how that group has inflated the results of an actual study that is effectively neutral in its conclusions.

A "weak" aspartame/cancer study

Very recently a study (entitled "Consumption of artificial sweetener— and sugar-containing soda and risk of lymphoma and leukemia in men and women") from Brigham and Women's Hospital came out which looked at 22 years of data on consumption of soft drinks (both sugar- and aspartame-sweetened) and incidence of lymphoma and leukemia. Although the study is getting touted by the anti-aspartame community as convincing evidence demonstrating that aspartame causes these cancers, the actual study results are significantly less forceful. Their conclusion (emphasis mine):
Although our findings preserve the possibility of a detrimental effect of a constituent of diet soda, such as aspartame, on select cancers, the inconsistent sex effects and occurrence of an apparent cancer risk in individuals who consume regular soda do not permit the ruling out of chance as an explanation.
They were only able to tease out effects for lymphoma if they analyzed men and women separately, and even then only for men (RR 1.3 even then, the low end of the relative risk was 1.01, which means no increased risk). In fact, in men consumption of regular, non-diet soda was associated with lymphoma. For leukemia, on the other hand, they could only get an effect if they combined the sexes. The low end of increases risk (RR 1.4) there actually started at 1.00 or, in other words, no increased risk. Unless they had planned it from the start, it strikes me as odd that they would use different analyses for each of the illnesses, as that ends up feeling like pure data mining to find an effect.

So in essence the study found nearly no effect. In fact, Brigham and Women's Hospital, where the research was performed and promoted, ended up apologizing after it was published for pushing such a "weak" study (their words) to the media for reporting. It looks like it is scheduled to get published in the December 2012 issue of the American Journal of Clinical Nutrition, though interestingly the table of contents for the November issue of the journal lists an editorial commenting on the article.

The anti-aspartame community responds

It should come as no surprise that this study was not taken for what it is by the anti-aspartame folks out there. Instead, headline after headline says something along the lines of "Aspartame linked with cancer in 22-year study". Sometimes the word "Landmark" is thrown in for extra effect. For example, one blog goes out of its way to quote from the results, except for studiously avoiding the part about regular soda also being linked. That same article also cites a cancer study which has been roundly criticized by both the FDA and the European Food Agency (EFSA).

Industry scare tactics?

Where it gets extra interesting is the bit of conspiracy-thinking that suggests that the reason the school came out with an "apology" and possible retraction was due to industry scare tactics, and not because its results show almost no effect. None other than Joe Mercola had this to say (emphasis my own):
After all, can you imagine the liability the food and beverage industries, not to mention virtually every public health agency in the US, would face were there convincing evidence that aspartame is carcinogenic? They simply cannot afford such evidence to be accepted.
As should be clear now, to call this "convincing evidence" is speaking well above what the evidence actually seems to demonstrate. There is also that hint of conspiracy theory that the only reason the hospital issued the release was through some unspecified action of the food and beverage industry. Actually finding a strong link between aspartame would, I think, garner enough attention to get lots of funding to look "deeper" into other products that have been deemed by the Internet to be unsafe.

But at the levels we are talking about in this study, even fruits and dairy products have been found to be linked with lymphoma. For example, in one study dairy and fried meat were associated (OR of 1.5 between lowest and highest) with NHL. Another study, looked at tobacco (3.5), barbecued meat (1.7), salted meat (2.4) and beer (5.5). The high end risk factor of those in some cases goes well into the 20s.


So was aspartame linked with leukemia and lymphoma? It seems so, but weakly, and so was regular soda. And both had lower risk numbers than other dietary factors that apply to the average person (even one avoiding soda). In other words, it seems that, as the cliche goes, nearly everything is in some way associated with various cancers. So to claim that aspartame has been found to be a particularly insidious culprit is simply untrue.

1. Schernhammer, Eva S., Kimberly A. Bertrand, Brenda M. Birmann, Laura Sampson, Walter W. Willett, and Diane Feskanich. "Consumption of artificial sweetener—and sugar-containing soda and risk of lymphoma and leukemia in men and women." The American Journal of Clinical Nutrition (2012).
2. Chang, Ellen T., Karin Ekstrm Smedby, Shumin M. Zhang, Henrik Hjalgrim, Mads Melbye, ke st, Bengt Glimelius, Alicja Wolk, and Hans-Olov Adami. "Dietary factors and risk of non-Hodgkin lymphoma in men and women." Cancer Epidemiology Biomarkers & Prevention 14, no. 2 (2005): 512-520.
3. De Stefani, Eduardo, Luis Fierro, Enrique Barrios, and Alvaro Ronco. "Tobacco, alcohol, diet and risk of non-Hodgkin's lymphoma: a case-control study in Uruguay." Leukemia research 22, no. 5 (1998): 445-452.
4. US Food and Drug Administration. "FDA Statement on European Aspartame Study". CFSAN/Office of Food Additive Safety. April 20, 2007. Accessed 6/13/2010
5. European Commission Scientific Committee on Feed. "Opinion of the Scientific Committee on Food:Update on the Safety of Aspartame". December 4, 2002. Accessed 6/13/2010.

by Josh DeWald

Share Tweet Reddit

@Skeptoid Media, a 501(c)(3) nonprofit