麻豆传媒

Philadelphia's beverage tax on sugary drinks was not associated with changes in pediatric weight outcomes 2 years after it took effect, researchers found.

Among 2- to 18-year-olds with BMI measured both before and after the $0.015/oz tax was implemented in 2017 on sugar-sweetened beverages (SSBs) and artificially sweetened beverages, standardized body mass index (zBMI) dropped by only a nonsignificant 0.004 (P=0.13) between youth in Philadelphia and those in surrounding counties who served as controls without the tax, reported Emily Gregory, MD, MHS, of Children's Hospital of Philadelphia, and colleagues.

Obesity prevalence didn't change either, with an odds ratio of 1.02 (95% CI 0.97-1.08) for BMIs in the 95th percentile or higher with versus without the tax, according to the findings published in .

"Sweetened beverage taxes are widely implemented outside the U.S. but there are still relatively few inside the U.S.," Gregory told 麻豆传媒 in an email. "Philadelphia is the largest jurisdiction to implement a sweetened beverage tax so evaluating health outcomes related to this new policy seemed important."

Philadelphia's tax led to a 30% increase in price and a 35% decrease in overall sales of sweetened drinks.

"The evidence suggests that taxes can be highly effective in changing consumption and reducing the burden of diseases associated with consuming these products," wrote Annika Janson, MD, PhD, and Liselotte Schäfer Elinder, MSc, PhD, both of the Karolinska Institute in Stockholm, in an . "But as illustrated by the study by Gregory et al, taxes at 50 cents per liter on SSBs and artificially sweetened beverages, while reducing consumption, are simply not enough to curb obesity in children."

They called for a "comprehensive package of prioritized interventions addressing healthy diet and physical activity across multiple settings to prevent obesity."

Indeed, four prior studies looking at beverage taxes and weight outcomes found "modest" BMI reductions for youth in Seattle and adolescent girls in Mexico, no BMI changes for youth in Mauritius, and reductions in low birth weight across U.S. cities that adopted beverage taxes, Gregory and colleagues noted.

"In general, evaluations of sweetened beverage taxes and pediatric weight outcomes have demonstrated small effects, sometimes limited to certain subgroups," Gregory told 麻豆传媒. "I think our findings are in line with this other literature. Weight is multifactorial and policy changes that address one component of healthy lifestyle may be important but not sufficient to move this outcome."

Although the weight-related outcomes of the Philadelphia Beverage Tax were "minimal," there were some other benefits -- small reductions in dental cavities and adult weight, she added. The tax "also generated over $180 million for the city that has largely gone to support pre-kindergarten and renovation of parks and recreational facilities."

Her group's study used electronic health record data on 136,078 individuals ages 2 to 18 years with one or more BMI measurement pre-tax (2014 to 2016) and one or more BMI measurement post-tax (2018 to 2019) in a pediatric health system with 28 primary care offices. Among them, 32,920 lived in Philadelphia and were exposed to the tax, while the rest lived in surrounding communities that did not have the tax.

In the sample of children in Philadelphia, the unadjusted pre-tax mean zBMI was 0.46, and 16% had a BMI in the 95th percentile or higher. The sample was 69% Black, 16% white, 15% multiple or other races, and 8% Hispanic ethnicity. Additionally, 71% received Medicaid, and 79% lived in census-tracts with Yost index of 1 or 2 indicating lower socioeconomic status.

While the primary outcomes showed no impact of the tax, there was a significant difference in the secondary outcome of zBMI change for youth with baseline BMI 85th percentile or higher compared with controls (−0.020, 95% CI −0.028 to −0.012).

Cross-sectional analysis of a larger group of 2- to 18-year-olds who had one or more BMI measurement at any time from 2014 to 2019 showed similar results. Among these 258,584 children, the difference in zBMI change was −0.004 (95% CI −0.009 to 0.001, P=0.10) between Philadelphia and the controls. And the tax was associated with an odds ratio of a BMI in the 95th percentile or higher of 1.01 (95% CI 0.95-1.07).

Subgroup analyses suggested that white children in Philadelphia had greater zBMI decreases compared with the controls (panel −0.034, 95% CI −0.040 to −0.027; cross-section −0.018, 95% CI −0.024 to −0.012). However, there was no difference in zBMI change for Black youth.

The cross-sectional data showed no difference in standardized BMI change with the tax for children ages 2 to 5 (−0.013, 95% CI −0.027 to 0.001) but an increase for 6- to 12-year-olds (0.027, 95% CI 0.020-0.033) and a decrease among those ages 13 to 18 (−0.030, 95% CI −0.039 to −0.022).

There were no differences when stratifying by sex.

"Though we found changes in certain subgroups, the magnitude of changes was small, inconsistent, and of limited clinical significance," Gregory and colleagues noted. "Overall, these findings are consistent with prior literature, which have shown no associations or modest associations between SSB taxation and weight in certain contexts."

Limitations included that the research team was not able to match on dietary factors, that data from a single health system may not be broadly representative, that data were observational, and that BMI is not a perfect measure of adiposity.

Comment