Reduction in purchases of sugar-sweetened beverages among low-income Black adolescents after exposure to caloric information

Abstract

Objectives: We examined the effect of an intervention to provide caloric information about sugar-sweetened beverages (SSBs) on the number of SSB purchases.

Methods: We used a case-crossover design with 4 corner stores located in low-income, predominately Black neighborhoods in Baltimore, Maryland. The intervention randomly posted 1 of 3 signs with the following caloric information: (1) absolute caloric count, (2) percentage of total recommended daily intake, and (3) physical activity equivalent. We collected data for 1600 beverage sales by Black adolescents, aged 12-18 years, including 400 during a baseline period and 400 for each of the 3 caloric condition interventions.

Results: Providing Black adolescents with any caloric information significantly reduced the odds of SSB purchases relative to the baseline (odds ratio [OR] = 0.56; 95% confidence interval [CI] = 0.36, 0.89). When examining the 3 caloric conditions separately, the significant effect was observed when caloric information was provided as a physical activity equivalent (OR = 0.51; 95% CI = 0.31, 0.85).

Conclusions: Providing easily understandable caloric information--particularly a physical activity equivalent--may reduce calorie intake from SSBs among low-income, Black adolescents.

FIGURE 1—

Study design: a store-based intervention to reduce sugar-sweetened beverage consumption among low-income Black adolescents, Baltimore, MD, April–October, 2011.

FIGURE 2—

Beverage purchases from 4 corner stores in a low-income, predominantly Black neighborhood of Baltimore: a store-based intervention to reduce sugar-sweetened beverage consumption among low-income Black adolescents, Baltimore, MD, April–October, 2011. Note. SSB = sugar-sweetened beverage. These estimates are the predicted probabilities from 8 separate logistical regression models predicting the odds that a purchased beverage was that particular type. These estimates are, therefore, adjusted for gender, store, time of day, time of year, time of week (weekend vs weekday), and average beverage price.* P < .05