Another Dimension of Health Care Reform: Discouraging Soft Drink Consumption →

This article by Sharon Begley, “Can it! Soda studies cite stronger link to obesity,” discusses the strongest evidence so far that discouraging the consumption of sugary soft-drinks could reduce obesity. (The studies actually look at the effects on weight of substituting diet soft drinks for sugary soft drinks.)  

Many people think of discouraging sugary soft drink consumption as something that would appeal mainly to the political left. But Modeled Behavior suggests in a tweet at least one way to discourage soft drink consumption that might appeal to the political right, saying this: 

A federal ban on using food stamps to buy non-diet soda seems like a natural policy for Romney.

Let me say something about diet soft drinks. My understanding is that there is currently not enough evidence to say definitively what the effects of diet soft drinks are, though the recent studies cited in Sharon Begley’s article suggest that it causes less weight gain than sugary soft drinks. Based on the shreds of scientific evidence and reasoning I have read, my hypothesis is this:

Sweetness itself, regardless of how the sweet taste is generated may trigger an insulin response, getting the body ready for food. If food is not forthcoming, that readiness for food will make one feel hungry. Drinking diet soft drinks with a meal is OK, since in that case getting the body prepared for food was appropriate. But drinking diet soft drinks as a snack is likely to lead to weight gain.  

Be careful. I am not saying this is known. It isn’t. As far as I know, the evidence just isn’t there. But if I were an obesity researcher, this is the hypothesis I would be investigating. I would be delighted for any references to evidence for or against this hypothesis, and encourage those who are obesity researchers to pursue it, if they are not already. Because the hypothesis involves the detailed timing of consumption diet soft drinks, it would be hard to get good evidence from non-experimental data.