Soft-drink epidemiology

If you’ve been following my occasional posts on nutrition, you’ll know that I’m no fan of epidemiological studies as ways to explore human metabolism.  It’s not enough simply to follow a group of nurses for a while and say “the ones that ate meat got fat; therefore, meat is bad for you” — there are hundreds of correlates involved and separating out a single variable so glibly — and attributing it to fundamental metabolic truths —  is just plain bad science.  That’s like studying the behaviour of an ant colony by looking at imagery from a weather radar station; you might find that ants do things differently when it rains, but you shouldn’t expect any insights into the colony’s social structure.

On the other hand, epidemiological studies are great for looking at population effects.  While an uncontrolled survey-based study might not tell you anything about fructose metabolism in the liver, it has a lot to say about whether populations in general tend to exhibit a correlation between soft-drink intake and obesity.  In other words: We know pretty well that, all else being equal, drinking a bunch of Coke is likely to make you fat; however, all else is not equal and before we start in on Pigovian taxes or other, more direct methods, we should check to see whether increased soda intake really is correlated with obesity in the general population.

Turns out not.  Aaron Carroll reports that:

You should click through and RTWT, but I’ll excerpt the main result:

See a pattern? No? Me neither. For 6-11 yo females, being in the “soft drink” cluster gave you the lowest chance of being obese. Same for 12-18 yo males. But it’s all over the place. All of those p values on the right hand side show that in none of the age/sex groups was BMI classification associated with beverage cluster.


Soft drinks were found to be significantly associated with a higher odds of overweight and obesity in boys 6-11 yo. No other relationships were seen for any beverages (including soft drinks) and any of the other age/sex groups. Looking at the prevalences in the first table I showed you that’s not surprising. But I find it fascinating that this relationship is totally missing in the 12-18 yo males. This means that either it’s a totally new phenomenon in young boys, who will continue to be obese, or this relationship disappears as they grow older.

Should this be taken as evidence that chugging gallons upon gallons of Coke won’t make you fat?  Not at all.  Depending on how much demand your metabolism has for liver glycogen, and how the rest of your diet meets that glycogenesis demand, it might not make you fat — see for example Michael Phelps — but if all you do is crank up your Coke intake because “studies show” it won’t make you fat, you’re likely to get fat(ter).

What this study does show is that you can’t predict which kids are going to get fat by looking at nothing but their soda intake, and therefore that soda control isn’t likely to have much of an effect — if any — on obesity rates.  People whose metabolisms are broken and who crave simple sugars from Mountain Dew are going to get them somewhere else, just like the ones who get their simple sugar fix while avoiding sodas do already.  (Is simple carb intake and metabolic dysfunction the reason people get pudgy?  My guess is it’s a big part of it, but not quite the whole story.)


0 Responses to “Soft-drink epidemiology”

  1. Leave a Comment

Leave a reply; use raw HTML for markup. Please blockquote quotations from the post or other comments.

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

anarchocapitalist agitprop

Be advised

I say fuck a lot



Statistics FTW


%d bloggers like this: