Not meaning to necro, but I always remembered this thread. A few years ago my wife and I started drinking lot's of diet soda with aspartame. We quit two weeks ago after I found these:
I realize these tests are on rats, but we don't feel like risking it. We basically drink water and coffee now. Occasional juice, beer, and wine. I'm still on the fence about Splenda and Stevia. My new strategy for our health is to just stop being addicted to sweet.
The results of this mega-experiment indicate that APM is a multipotential carcinogenic agent, even at a daily dose of 20 mg/kg body weight, much less than the current acceptable daily intake. On the basis of these results, a reevaluation of the present guidelines on the use and consumption of APM is urgent and cannot be delayed.
also, according to the can of coke zero i am drinking, 20 mg/kg is ~40 cans of diet per day for a 150lb man.
A meta-analysis has the greater statistical power; systematic review trumps cherry picked primary studies.
That's if all studies included in your sample were testing the same thing right?
My inclination here is that the EFSA reviewed a large number of studies on the effects of Aspartame on humans, and found them to be sound. The majority of them, if not all, indicate that there can not be a caustic relationship between cancer and aspartame so they say here is an ADI go for it. Which is entirely logical. However, there are studies in controlled environments (unlike X number of humans over X years that are exposed to changes in lifestyle, chemical exposure, etc over the years), that definitively show aspartame given to rats over time causes them to develop cancers.
Is it wrong of me to mentally discard the studies done on humans because I think it is nearly impossible to prove anything like that? I guess that's cherry picking. I don't know, someone tell me I'm an idiot with good logical reasoning so I can go back to drinking my filthy diet soda. I miss it.
@pirateninja By definition, a (good) systemic review must only consider studies that meet a predefined set of criteria, so, yes, they must all study the same thing. Also, in biology, we always consider human data to be stronger than any pre-clinical data (mice, rats, cell lines), simply because they are only models for how things interact in humans - in cancer bio, you can barely sit through a meeting without person x reminding person y that "mice aren't people". As Thrax (i think) mentioned above, the dose also makes the poison - 100 mg/kg daily (the only treatment at which there was a statistically significant positive result, from table 1 in your first paper) is considerably more than a human would consume. There are infamous studies showing that rats get cancer from peanut butter at doses like that (google Bruce Ames). Also, as a nitpicky point, anything using Sprague-Dawley rats makes me concerned. They are notorious for getting random tumors, which is why in the paper, they specifically mention the historical rate of specific tumors in their SD rat colonies - most of the numbers of affected rats from the treatment group also fall with in that range as well, despite being statistically significant in this particular study.
Comments
also, according to the can of coke zero i am drinking, 20 mg/kg is ~40 cans of diet per day for a 150lb man.
My inclination here is that the EFSA reviewed a large number of studies on the effects of Aspartame on humans, and found them to be sound. The majority of them, if not all, indicate that there can not be a caustic relationship between cancer and aspartame so they say here is an ADI go for it. Which is entirely logical.
However, there are studies in controlled environments (unlike X number of humans over X years that are exposed to changes in lifestyle, chemical exposure, etc over the years), that definitively show aspartame given to rats over time causes them to develop cancers.
Is it wrong of me to mentally discard the studies done on humans because I think it is nearly impossible to prove anything like that? I guess that's cherry picking. I don't know, someone tell me I'm an idiot with good logical reasoning so I can go back to drinking my filthy diet soda. I miss it.
By definition, a (good) systemic review must only consider studies that meet a predefined set of criteria, so, yes, they must all study the same thing. Also, in biology, we always consider human data to be stronger than any pre-clinical data (mice, rats, cell lines), simply because they are only models for how things interact in humans - in cancer bio, you can barely sit through a meeting without person x reminding person y that "mice aren't people". As Thrax (i think) mentioned above, the dose also makes the poison - 100 mg/kg daily (the only treatment at which there was a statistically significant positive result, from table 1 in your first paper) is considerably more than a human would consume. There are infamous studies showing that rats get cancer from peanut butter at doses like that (google Bruce Ames). Also, as a nitpicky point, anything using Sprague-Dawley rats makes me concerned. They are notorious for getting random tumors, which is why in the paper, they specifically mention the historical rate of specific tumors in their SD rat colonies - most of the numbers of affected rats from the treatment group also fall with in that range as well, despite being statistically significant in this particular study.
I feel better about it now, back on the coke zero train we go.