No Health Benefit Seen from Small Soda Taxes

Health, diet, lifestyle and demographic data from a large national cohort of schoolchildren indicated that each 1% increment in so-called soda taxes reduced kids' mean body mass index (BMI) values by a nonsignificant 0.013 points, according to Roland Sturm, PhD, of the RAND Corp. in Santa Monica, CA, and colleagues.

Effects on drink consumption overall and in school were also barely detectable.

"Small taxes in the range of existing differentials are unlikely to have visible effects at the population level," the researchers wrote online in Health Affairs.

Sturm stated that, heftier taxes -- such as an 18% levy proposed in 2008 in New York state, or a two-cents-per-ounce tax suggested last month by Pittsburgh's mayor -- could have a much more significant effect.  "As the tax increases, the effect will not be proportional," he said. "It will be larger."

Sturm and colleagues also found that small taxes had statistically significant effects in certain subgroups, including African Americans, children already showing signs of excess weight, those from poorer households, and heavy TV watchers. As a result, they suggested, taxes could still have a useful public health benefit.

But Sturm said their most important effect may simply be increasing revenue, which could be used to fund programs targeting childhood obesity and poor health behaviors.

The analysis used data from the prospective Early Childhood Longitudinal Study and its Kindergarten Cohort, a group of more than 7,000 children first recruited in 1998.

Information on the youngsters' diets, including their consumption of sweetened drinks other than fruit juices, as well as their purchases of such drinks at school, was collected in 2004 when they reached fifth grade. Height and weight were measured regularly. The data also included sociodemographic information, children's weekly estimated time spent watching TV, and the frequency of vigorous exercise.

Sturm and colleagues matched the data on individual children with sales taxes on sodas in effect in their states of residence. In many states, such drinks are taxed at higher rates than other foods, and the difference was also included as a variable in the analysis.

Overall, there was little correlation between taxation and the fifth-graders' drink consumption or BMI.

For each 1% of soda taxation, the researchers found the following effects:

• Reduction in total soda drinks per week: 0.004 (not significant)
• Reduction in drinks bought at school weekly: 0.010 (not significant)
• Reduction in BMI: 0.013 (P between 0.05 and 0.10)

The effects were somewhat greater when Sturm and colleagues plotted drink consumption and BMI against the differential between soda taxes and sale taxes on other foods. But none reached the P<0.05 level except for BMI, which was reduced by 0.085 points for each 1% differential.

Among the 701 African-American children in the study, mean total and school-purchased drink consumption declined by 0.125 and 0.103 sodas per week, respectively, for each 1% increment in soda taxation. But there was no effect on their BMI.

Body weight was significantly affected by taxation, however, among the 2,900 participants who were heavier than average and considered at risk for becoming formally overweight. Each 1% in taxes was associated with a reduction of 0.033 BMI points in this group (P<0.05). Moreover, each 1% differential in taxation between sodas and other foods was associated with a 0.222-point reduction in BMI among the children at risk for overweight (P<0.05).

Children from households with income less than $25,000 and those who watched more than nine hours of TV weekly also drank fewer sodas overall and in school -- by 0.029 to 0.142 per week, for each 1% of tax (P<0.05). But as was the case with African-American children, there was no discernible effect on BMI.

In the Health Affairs paper, Sturm and colleagues suggested that structuring them as excise instead of sales taxes may boost their effectiveness.

"An excise tax is preferable... because it would be incorporated into the shelf price, making the higher costs more visible to consumers," they pointed out. They also indicated that a tax levied per ounce, such as the one proposed in Pittsburgh, would be easiest to implement.

Practice Pearls:

Explain to interested patients that sugary beverages contribute to childhood overweight and obesity, which, in turn, puts children and adolescents at risk for chronic diseases such as diabetes and heart disease.

Explain that increasing the costs of tobacco and alcohol has been shown to reduce consumption. Whether this strategy will also work for sodas has not been well studied, and this study indicated that small taxes on sugary soft drinks would have little health impact.

Sturm R, et al "Soda taxes, soft drink consumption, and children's body mass index" Health Aff 2010; DOI: 10.1377/hlthaff.2009.0061.