Artificial sweeteners and fatness

Another week, another observational trial showing the evils of artificial sweeteners.

The press releases and general lay public information from this trial are awful.  I know universities communication teams love controversial topics as they increase the hit count.  But this is garbage.

Observational trials are insensitive to pick up these effects, but they create a nice easy story.  You know, the "requires further investigation" type claims at the end of the paper.  Sure.  It also requires further investigation with examinations that are not just dataset milking piles of junk.

More to the point, this finding is nothing new.  Its most likely due to the fact that people know they are fat, therefore consume diet beverages - then treat it as permissive - I've been good and now i get to eat whatever.  Much like the gym goers who think because they did 20 minutes of cycling they deserve a triple fat mocha latte with a cookie.   Or like the previous research on telling people the product is low fat resulted in increased consumption.

Now, last year in the American Journal of Clinical Nutrition, there was an interesting meta-analysis published looking into this.  It was unique, in that it performed a meta-analysis on observational cohorts and also experimental trials.'

This paper can be found here.

It found that observationally, artificial sweetener consumption was associated with no real differences other than a significantly higher BMI (or significantly associated with a slightly higher BMI - as the authors say).

The meta-analysis of the experimental trials showed decreased bodyweight, decreased BMI, decreased fat mass and decreased waist hip ratio.

Why the difference?  Because observational trials are inadequate to decipher most of which they are asked to do.  But also, data mining of observational data sets (cough, Willett et al) will find enormous quantities of spurious associations that have nothing to do with anything.   Nothing like performing 5000 statistical tests on the same dataset to throw up some pointless garbage.

Just because fat people drink diet beverages doesn't mean that diet beverages caused the fatness.

Also, quit hyping your latest and greatest observational claptrap just because you want to advertise your University or Department.  Have some ethics.

Paper comment: Whey protein supplementation does not affect exercise training-induced changes in body composition and indices of metabolic syndrome in middle-aged overweight and obese adults.

I was going to post something else, but thought I would throw this up instead.

So, from the title you can see "Whey protein does not affect exercise training-induced changes in body composition"

Silly supplement companies, selling their whey protein for NO EFFECT.

Large study.  Lots more subjects than normally find in protein trials.  Middle aged, inactive people.  They should get some benefit from actually doing something, so whey sucks at doses ranging from 0, 20, 40 or 60g per day.

Man, they took all this protein and achieved nothing.

But, try and find the information that you would be interested in, and they hide it.  When it first came out I had to wait for the 'supplemental' information that was not available when it first came out.

The body composition changes.

 

So... nine months of exercise, with or without protein, and the fat inactive middle aged adults didn't really change body weight, lean body mass or fat mass.  Well, they lost a (significant) tiny bit of fat mass and gained an equally small amount of muscle.   NINE MONTHS!
But hey, they gained a minuscule bit of strength, which is within the range seen from learning how to perform an exercise.

These lovely folk got to use somebody elses money (Supported by the U.S. Whey Protein Research Consortium and NIH T32AG025671 and UL1RR025761), mess around with subjects for nine months, and even their training program is so modest that the people went away with little to no difference. Plus they get published in one of the leading nutrition journals.

Hard life for some.