Does removing sodas from school work?

A new study, behind a lovely paywall, attempts to quantify the effect of California's school competitive beverage policy. The main point is that there was a decline in obesity prevalence, but only for schools in the wealthiest neighborhoods.

It’s not a bad study, I don’t guess. I think the point that policies affect different groups differently is probably spot-on, and something we should always look at.


1) The biggest problem is that there is no control group. I could go on a tirade about the sorry state of obesity research, but this is just critical. Their time point of change—2004/2005—is exactly when obesity trends declined nationwide. Let me say the lack of a control group again. It is simply not possible to attribute the change to the policy. Period. There was a change the same time as the policy, but there are plenty of other reasons I could argue also “caused” the effect.

2) What’s the plausibility of the effect? It’s a policy that said, essentially, you can only sell drinks that are half crap and half juice (so just a caloric as a soda). A second policy, which more clearly limited sugar content, went into place in 2007, and there is no effect of that on their outcomes.

3) They don’t look at harms. I don’t love the idea of Coke having a role in schools. But as long as we are unwilling to fund schools appropriately, the financial impact of these programs can’t be overlooked. What’s more is that these programs are most likely to help the poorest neighborhoods. Not only do poor kids not get the same positive effect, their schools also lack the resources to fill in the gaps of the lost income.

4) Finally, prevalence of overweight/obesity is not a great outcome. Change in z-score is hard because it’s not longitudinal, but if there is more severe obesity in minority or poor groups, then there will naturally be less transition into the healthy weight group.

I really want to find a policy that works. But I don’t think the obesity “epidemic” can be pinned on the devil of sugar sweetened beverages, much less cured with a SSB policy based only at schools. It is the nature of researchers to parse out interventions to their lowest part. But the complexity of obesity simply won’t yield to parsimony. Obesity won’t be cured with a cooking class, swapping out the coke for juice, or a school vegetable garden. (Oh, god, the zealotry of the school vegetable garden people.)

Ironically, one small thing like a beverage policy is unlikely to improve obesity, but it does have the power to be harmful. Just as it takes a lifetime to raise kids to be happy and well-adjusted, it takes only a single traumatic event to completely change trajectories. It’s easy to pass a policy banning beverages. It’s not easy to include in that policy coverage of the costs of that decision.