But this reduction diminished over time
Labelling menus with the calorie content of individual dishes and drinks in fast food outlets is associated with a small immediate decrease in average calories purchased, according to a study using data from a large US restaurant company published by The BMJ today.
But this was followed by a gradual weekly increase over the next year, implying that calorie labelling alone “may not be enough to make sustainable reductions in calorie intake in fast food restaurants,” say the researchers.
Calorie labelling has been required in large chain restaurants in the US since May 2018 to help consumers make healthier choices and encourage manufacturers to reformulate dishes or increase the number of lower calorie options.
The policy is currently being considered in the United Kingdom, but evidence of its effects on calorie purchases is mixed and incomplete.
So a team of US researchers set out to evaluate the impact of calorie labelling on calories in meals purchased in a large restaurant franchise in the southern United States, where obesity rates are among the highest in the country.
Ahead of the new rule, the franchise labelled all menus with calories in April 2017 and provided weekly sales data from 104 restaurants during the pre-labelling (April 2015 to April 2017) and post-labelling (April 2017 to April 2018) period.
This gave the researchers access to nearly 50 million transactions across three years (two years before and one year after the change) for analysis.
Investigators calculated total calories for each menu item and grouped items into one of five categories: entrees (main courses), sides (including desserts), sugar sweetened beverages, low calorie beverages, and condiments.
After adjusting for the baseline trend, season and holidays, calorie labelling was associated with an immediate decrease of 60 calories per transaction, or 4% of total calories purchased.
However, this initial decrease was followed by a small weekly increase in calories per transaction over the next year so that, by the end of the study, the 60 calorie reduction had dropped to just 23 fewer calories for each purchase made.
This trend increase might happen faster among people with lower incomes, note the authors, although these results should be viewed cautiously.
They point to some study limitations, such as being unable to calculate calories purchased per person or measure meal modification (e.g. adding condiments), beverage refills, or how much of each meal was eaten. Strengths include the availability of a large sample over a three year period and findings that remained similar after further analyses, suggesting that they are robust.
“Before drawing conclusions on the overall effectiveness of calorie labeling as a nutrition policy, future research should be done to estimate the effects of labeling over a longer period, especially once restaurants have had sufficient time to reformulate their menus,” they conclude.
This paper contains other important messages for decision makers, say researchers at the University of Oxford in a linked editorial.
For example, as the data suggest, calorie menu labelling may inadvertently increase health inequalities, or restaurants might reformulate dishes to have fewer calories but more unhealthy nutrients such as salt or sugar. However, including additional nutritional information on labels could reduce the impact of these negative outcomes.
Although these results might be disappointing to some, they note that small changes to calorie intake can have meaningful effects at the population level. As such, they call for “a multifaceted, cross government approach” to tackle obesity in which calorie and nutrition labelling on restaurant menus should play a part.
Notes to Editors
Research: Estimating the effect of calorie menu labeling on calories purchased in a large restaurant franchise in the southern United States: quasi-experimental study
Editorial: Calorie labelling to reduce obesity
Journal: The BMJ
Funding: National Institutes of Health
Link to Academy of Medical Sciences press release labelling system: https://press.psprings.co.uk/AMSlabels.pdf
Peer-reviewed? Yes (research); No (linked editorial)
Evidence type: Quasi-experimental study; Opinion
Subjects: US fast food restaurant chain
Link to research: https://www.bmj.com/content/367/bmj.l5837
Link to editorial: https://www.bmj.com/content/367/bmj.l6119