Background

In May 2018, federal regulations mandated that restaurant chains with ≥20 locations nationwide label their menus with calorie information. Evidence of this policy’s effectiveness for reducing calorie purchases is mixed and incomplete. A large franchise of a national fast-food company located in Louisiana, Texas, and Mississippi began calorie labeling in April 2017 in anticipation of the federal rule, providing a rare opportunity to conduct a long-term, quasi-experimental study.

Methods

We obtained all weekly sales data from the franchise from April 2015 until April 2018. We used an interrupted time series and linear mixed models to estimate the change in calories/transaction immediately following labeling implementation, as well as the change in trend one year after implementation compared to the baseline trend. We conducted analyses overall, by category (entrees, sides, and sugar-sweetened beverages), and stratified by sociodemographic characteristics of restaurant ZIP codes.

Results

Over three years and among 104 restaurants, there were about 243 million items purchased across 49 million transactions, with a mean of 1490 calories/transaction (SD: 149). After labeling implementation, we observed a decrease of 60 calories/transaction (95% CI: 48, 72; approximately 4%), followed by a post-implementation trend increase of 0.71 calories/transaction per week (95% CI: 0.51, 0.92) independent of the baseline trend over the next year. The results were generally similar by category. We observed a stronger post-implementation trend increase in calories/transaction in lower-income (trend change=0.95 calories/transaction per week; 95% CI: 0.63, 1.26) than in higher-income ZIP codes (change=0.50 calories/transaction per week; 95% CI: 0.22, 0.77; P-interaction=0.05).

Conclusions

Calorie labeling was associated with a small decrease in calorie purchases following implementation in a franchise of fast-food restaurants. This reduction diminished over one year of follow-up.