Soda, schools and childhood obesity

This article discusses a new report from the University of Maryland that shows the impact of removing soda and high sugar content fruit drinks from schools is probably negligible in terms of limiting the number of children who are obese. The study was funded by...The American Beverage Association.

Despite its funding source, it is still a valuable exercise to understand how the study was conducted, which is helpfully described in the article (the study does not appear on the web yet as far as I could find):

"The authors of the current study stated there is little scientific evidence to support such a policy, and set out to analyze the existing literature. To this end, they used the tool of 'risk analysis,' which, Forshee said, 'has not been applied as widely as it should be in nutrition policy.'

"The idea was to apply the tool to a controversial area in nutrition policy (soft drink sales in schools) as a way to demonstrate that it might have wider utility.
Forshee and his colleagues used two federally funded data sets, including the National Health and Nutrition Examination Survey 1999-2000 (NHANES), and one data set from the National Family Opinion consumer research firm.

"The researchers used the largest association between school soft drinks and body mass index (BMI) they could find, which was a .24 unit change in BMI for every one serving change in soft drink consumption.

"Even using this upper-end figure, Forshee said, 'there was no statistically significant association and, in fact, regular carbonated soft drinks accounted for less than 1 percent of the variance in BMI.'

"Consumption of soft drinks from school vending machines was also quite low, with estimates ranging from half an ounce to two ounces per day per student. Adolescents drank five times as much at home." (Italics added)

The italicized portion reveals the authors have used an "average" in a misleading manner. I state this because, earlier in the article, we had learned that childhood obesity was in about 10% of the children population of the US in the late 1980s, but had grown to 15% by the year 2000. (Separate the question of how that trend has been measured. Let's just assume its accuracy for today.)

With that relatively small, but growing population of obese children in mind, let's remind ourselves that a soda can is not 1/2 an ounce to 2 ounces. A soda can contains 16 ounces or sometimes 20 ounces. Further, note later in the article where it states:

"With current consumption of such drinks at 2 servings a day for males and 1.2 a day for females, it's unlikely that such a policy in schools will make a 'meaningful difference in BMI distribution of the population,' the authors wrote." (Italics added)

Hmmmm...If our public school children are only consuming two servings a day, then a small group of children are getting most of those ounces of soda from those vending machines. Therefore, a ban on soda and high sugar content fruit drinks could have a beneficial impact on the consumption of sugar if 1 of the 2 servings occurs at school. Further, assuming there is a relation between obesity and drinking soda, which epidemiologists will know more than I, the authors, if they were not being paid by the American Beverage Association, might have been able to see that the following conclusion is at least as reasonable as theirs:

Banning soda in schools is not likely to adversely affect any children, who may not be drinking soda in any event at school. However, for those children who drink even one can of soda per day from vending machines at school, the beneficial effect of a ban on soda vending machines at schools could reduce, by half, such children's sugar consumption.

In all, I am glad the American Beverage Association helped move the debate forward. The authors of the study should be commended for at least reviewing additional data. However, the authors' conclusion is sadly a rather poorly reasoned conclusion.

For our family, we have no soda in our house. We let the kids have a soda as a treat perhaps two or three times a month in a restaurant. As for fruit drinks, we have become more discerning than we were. We try now to choose fruit juice that contains less than 30 grams of sugar per 8 oz. serving. Coca-Cola has 27 grams per 8 oz. serving, however. See: Coca Cola's web site here, which claims that orange, apple and grape juices have essentially the same sugar content. I originally wrote on my blog at 6:30 a.m. this morning that Coca Cola was being misleading here. However, this evening, when looking at our newly purchased orange and apple juice, I have to say I was wrong and Coca-Cola is correct. The differences are not that great after all. I thought the label on our jucies was based upon 16 ounces, but it's based upon 8 ounces. Thus, the apple juice we normally buy has 19 grams of sugar at 8 ounces per serving, which is about 10 grams less than Coca-Cola. On the other hand, the orange juice we purchased today has just as much sugar as Coca-Cola.

In the end, however, banning soda and fruit juice vending machines from public school grounds is still a reasonable policy. It will probably be most helpful to those children most at risk of obesity and may lead to more nutritional habits in at least some of our nation's children.

(Edited and corrected in next to last paragraph)