Displaying calories on restaurant menus: scenarios for the UK

Over 9,000 heart disease deaths could be avoided by 2041 in England if all gastronomy establishments displayed calorie disclaimers on their menus.

UK makes it mandatory to display calories on menus

Over one in four adults in England is currently suffering from obesity, and trends suggest that this percentage is set to increase.

New menu legislation came into effect in the UK on 6 April 2023. Under the novel law, restaurants and takeaways must report the calorie content of each dish, on location and online, if businesses have more than 250 employees. Similar legislation is under consideration in Wales and Scotland, and has been gradually introduced in countries such as the US (since 2019) and parts of Australia.

Early estimates of this rules' impact suggest that the policy could prevent or postpone approximately 730 deaths from cardiovascular disease between 2022 and 2041. It is estimated that the health benefits would be greater if the legislation were applied to all English catering establishments, leading to the avoidance of approximately 9,200 deaths from cardiovascular disease over the next 20 years under this scenario.

Previous studies conducted in countries such as the UK, US and Canada suggest that indicating calories on menus leads people to order meals with around 47 kcal less and businesses to reduce the average calorie content of their meals by 15 kcal. This study is the first to model the impact of menu calorie labelling on obesity and deaths from cardiovascular disease in England, and how the effect varies across different socio-economic groups.

Can calories on menus reduce the impact of cardiovascular disease?

The authors studied the estimated effect on obesity rates and deaths from cardiovascular disease from 2022 to 2041, with a mandatory calorie labelling on menus in England for two scenarios:

  1. The current situation, in which only large catering establishments with 250 or more employees are obliged to provide calorie information (this type of establishment accounts for 18% of outlets);
  2. The introduction of mandatory calorie information on menus in all catering establishments in England.

These two scenarios were compared to a baseline scenario with no regulation at all.

Without any regulation on the indication of calories on menus, the model estimates that the trend in cardiovascular disease will result in approximately 830,000 deaths by 2041 (range 600,000-1,200,000). However, it is estimated that the current policy prevents about 730 of these deaths (range 430-1,300) and that, if the policy were extended to all activities, about 9,200 deaths could be prevented (range 5,500-16,000), almost 13 times more than under the current policy.

The base case assumed that the prevalence of obesity in England in 2041 would be 27%. The model estimated that the current policy would reduce obesity prevalence by 0.31 percentage points (within a range of 0.10-0.35) over the next 20 years, but full implementation of the policy would reduce obesity prevalence by 2.65 percentage points (within a range of 1.97-3.24).

According to the authors, the results suggest that extending the requirement to display calories on menus to all British catering establishments could play an important role in future government strategies to support people to make healthier choices to combat obesity. In addition, the authors encourage the government to continue and strengthen the strategy to prevent obesity in England with a wide range of policies, such as calorie labelling, combating junk food marketing and increasing taxes for the soft drinks industry, that will reduce both obesity and the shocking health inequality gap in society.

The authors point out that their study has some limitations, including the fact that the calorie reduction data associated with the policy within the model was taken from US studies, which may not be comparable to the UK population. Furthermore, they point out that the study only modelled adult obesity, so it is unable to examine the impact on childhood obesity.

Comments and criticism

In a commentary, Dr David D. Kim of the University of Chicago, who did not participate in the study, stated that: "The evidence gained from conventional study designs is sometimes insufficient to inform policy decisions. They tend to focus on short-term health outcomes (e.g., changes in weight or biomarkers), may not represent a heterogeneous population, and are unlikely to evaluate all policy-relevant options." Modelling and simulation can help to fill this evidence gap by informing multiple relevant processes, testing a multitude of plausible scenarios that would be impractical and unfeasible in trials, quantifying the magnitude of expected and unintended consequences, and having the ability to adjust and refine designs prior to a trial or actual real-world implementation.

Some researchers criticised the basic hypothesis of the study. The authors of the research hypothesised that the implementation of calorie labelling on menus would reduce energy consumption by 47 kcal (95% CI 15-78) per meal out, drawing these estimates from the Cochrane review and meta-analysis of randomised controlled trials by Crockett and colleagues. The authors rely on this assumption for their predictive statistical study. According to critics, this study was not designed to test the hypotheses of the aforementioned 2018 review, and they point out that pre-pandemic and pre-crisis conditions may not accurately reflect current eating behaviour. They also raise doubts about the alleged uniformity of the effects of calorie labelling in different demographic categories.

Others point out the complexities and uncertainties involved in evaluating the effect of calorie labelling. They question the assumption that calorie labelling automatically leads to a reduction in calorie consumption, pointing out that some studies suggest instead an increase in calorie consumption in response to such a procedure (people are starting to switch off, some groups may be reasoning for better value for money).

Finally, concerns are raised about the financial impact and feasibility of the new labelling for small food businesses, as well as the lack of consideration of potential negative effects, such as triggering or reinforcing eating disorders.  Indeed, the current law has already attracted the interest of British eating disorder associations, as this regulation may be a barrier to eating out with family and friends for people recovering from an eating disorder.  If the policy of displaying calories on menus were to be extended to all catering businesses, the risk could increase further.

"All models are wrong, but some are useful"

Prof. Kevin McConway, Professor Emeritus of Applied Statistics, The Open University, said: 'One might wonder why it is worthwhile.  Statistical models, particularly predictive models, are often the subject of bad press because things do not turn out as predicted by the model.  However, many statisticians (myself included) are wont to quote an eminent 20th century Anglo-American statistician, George Box, who said: All models are wrong, but some are useful.  Models inevitably simplify what happens in the real world, so they misrepresent some aspects of it, just as a map does not represent exactly what is on the real ground. But models can give us an idea of what has happened, or might happen, in ways that can be very useful, just as a map can be.  In this case, policy makers have to decide whether or not to change the rules governing mandatory calorie labelling for food provided and consumed away from home.  They cannot observe what will happen with their policies 20 years in the future. A good, useful model can help them understand what might work and what might not, and can enable them to compare the impact of different policy choices in ways that are essentially impossible to do by other methods.  One can be fairly confident in stating that the predictions of this modelling study will not exactly happen and, with hindsight, may turn out to be rather inaccurate. But, if nothing else, the exercise will indicate which features seem to be important and where the greatest uncertainties lie."

Sources
  1. Colombet Z, Robinson E, Kypridemos C, Jones A, O'Flaherty M. Effect of calorie labelling in the out-of-home food sector on adult obesity prevalence, cardiovascular mortality, and social inequalities in England: a modelling study. Lancet Public Health. 2024 Mar;9(3):e178-e185. doi: 10.1016/S2468-2667(23)00326-2. PMID: 38429017.
  2. Kim DD. The role of simulation modelling in public health policy evaluation. Lancet Public Health. 2024 Mar;9(3):e150-e151. doi: 10.1016/S2468-2667(24)00027-6. PMID: 38429013.
  3. Science Media Centre. Expert reaction to modelling study looking at menu calorie labelling, and obesity and deaths from cardiovascular disease in England. 28 february 2024