Sweetened beverage tax cuts kids' BMI in Seattle, study finds

In a recent study published in JAMA Network Open, researchers investigated whether implementing a tax on sweetened beverages was associated with changes in the body mass index (BMI) of children in Seattle.

Their findings indicate that the tax was significantly associated with a decrease in BMI among children residing in the Seattle area, suggesting that such taxes may effectively contribute to modest improvements in children’s BMI.

Study: Sweetened Beverage Tax Implementation and Change in Body Mass Index Among Children in Seattle. Image Credit: WH_Pics / Shutterstock

Background

To date, seven US cities have implemented excise taxes on sweetened beverages to improve public health by reducing sugar-sweetened beverage intake, the largest source of added sugar in the US, and to generate revenue for other health programs.

Previous studies have shown that these taxes increase beverage prices and reduce beverage purchases, potentially lowering calorie intake. However, the relationship between these taxes and self-reported sugary beverage consumption is often null, likely due to measurement difficulties and small study sizes.

Assessing health outcomes remains crucial as children, who consume more sweetened beverages, might be more affected by these taxes than adults.

Prior research on sweetened beverage taxes and children’s BMI is limited and shows mixed results: no association in Mauritius, a reduction in obesity among adolescent girls in Mexico, and a BMI decrease among children in three US cities.

About the study

This study used longitudinal BMI data from Seattle to examine the impact of taxes on children’s BMI, hypothesizing lower BMI gains post-tax.

This cohort study included children aged 2 to 18 who resided in urban neighborhoods of three Seattle counties and received primary care from two major hospitals or clinics between 2014 and 2019.

Participants were excluded if they had cancer, undergone bariatric surgery, moved out of the study area, or had extreme BMI values.

The exposure of interest was the tax on sweetened beverages that was implemented in Seattle in early January 2018. Data on weight and height were obtained from electronic health records, and BMI was calculated.

Researchers used BMIp95 (BMI as a percentage of the 95th percentile for age and sex) as the study’s primary outcome.

The study used two statistical models: a synthetic difference-in-differences (SDID) model for children with annual BMI measurements from 2015 to 2019 and a fine stratification average treatment effect (FSATE) weighted within-person change model for children with at least one pre- and post-tax measurement.

Both models adjusted for confounders, with the primary model emphasizing pretreatment trends and individual-level fixed effects. Analyses were conducted using Stata, with statistical significance set at α = .05.

Findings

The study’s primary SDID model comprised a sample of 6313 children (48% female and 52% male) with annual BMI measurements over five years, 28% of whom were from Seattle and 72% from comparison areas.

On average, the participants were 7.7 years old, with a racial/ethnic composition of 13% Asian, 10% Black, 10% Hispanic, 50% White, and 11% of multiple races. The FSATE-weighted model included 22,779 children.

Results showed that post-tax, Seattle children had a more significant reduction in BMIp95 compared to those in non-taxed areas, with an SDID estimate of -0.90 percentage points (95% CI, -1.2 to -0.60).

The FSATE-weighted model showed a similar but slightly larger reduction (β = -1.16 percentage points, 95% CI, -1.91 to -0.41).

Subgroup analyses indicated consistent reductions in BMIp95 across various demographic groups, including different ages, sexes, races, and insurance types, particularly among those with baseline overweight.

Conclusions

The study concluded that the tax on sweetened beverages was correlated with a statistically significant reduction in children’s BMIp95. This finding aligns with previous research, such as studies in Philadelphia, San Francisco, Oakland, and Mexico, which also reported reductions in children’s BMI following beverage tax implementations.

The study’s strengths include its use of longitudinal data, measured heights and weights, and robust statistical methods to control for pretax differences, thereby providing more reliable evidence than prior cross-sectional studies.

Limitations of the study include the lack of direct data on sweetened beverage consumption and the use of medical records with limited socioeconomic details. Additionally, while the SDID model offered internally valid estimates, it required all children to have the same number of outcome measurements, which reduced the sample size and generalizability.

Despite these limitations, the study provides strong evidence that taxes on sweetened beverages can provide practical policy tools to target BMI reductions in children. Future studies can explore this association in other cities with similar taxes to confirm these findings.