London School of Hygiene & Tropical Medicine

Introducing a small levy of 10 pence per drink to the price of sugar-sweetened beverages (SSBs) sold in Jamie’s Italian restaurants across the UK is likely to have contributed to a significant decline in SSB sales, according to new research published in the Journal of Epidemiology & Community Health.

The study was led by the London School of Hygiene & Tropical Medicine with the University of Cambridge, and funded by the National Institute for Health Research. After adjusting for general trends in sales it found that adding a 10 pence levy to SSBs sold in 37 Jamie’s Italian restaurants, combined with activities such as re-designing menus, offering new lower sugar drinks and related publicity, was associated with an 11% decline in sales of SSBs per customer 12 weeks after the levy was introduced. A decline in sales of 9.3% per customer was still observed six months after the levy was introduced. The authors say further research with a longer follow-up is required to assess whether this is sustained. Reductions were greatest in restaurants with higher SSB sales per customer.

Consumption of sugar-sweetened beverages is associated with obesity, type 2 diabetes, cardiovascular disease and tooth decay. Decreasing consumption of SSBs can reduce body weight and weight gain in children and adolescents. In the UK, SSBs are thought to account for up to half of the excess calories consumed per day by children. Adults consume an average of 50 calories per day from SSBs.

Cutting consumption of SSBs is therefore seen as important in improving public health but the most effective way of encouraging this change in behaviour is less clear. Financial measures, alongside wider strategies, are one option but there is limited evidence on the effectiveness of such measures on either the sale or consumption of SSBs.

In September 2015 Jamie’s Italian, a chain of restaurants founded by chef Jamie Oliver, added a 10p levy to the price of its non-alcoholic SSBs. At the same time, the chain reorganised the non-alcoholic beverage menu into two sections: SSBs and ‘other’ beverages which included fresh fruit juices, bottled waters and diet cola. In addition, fruit spritzers (fruit juice mixed with water) were added to the main non-alcoholic beverage menu. The menu also explained the decision to implement the levy and that proceeds would go directly to a Children’s Health Fund which supports children’s health initiatives. Using sales data from 37 Jamie’s Italian restaurants, this study explored the effects of these changes on sales of all types of SSBs.

Steven Cummins, Professor of Population Health at the London School of Hygiene & Tropical Medicine, who led the study, said: “Obesity, type 2 diabetes and cardiovascular disease are among the most pressing global health challenges facing the world today. Evidence suggests that excessive consumption of sugar-sweetened beverages is an important contributor to these potentially life-threatening conditions but we still don’t have a clear answer on how best to encourage people to consume fewer of them.

 “Our study showed that a combination of the levy, menu changes and clearly explaining to customers why it was introduced and that the proceeds would go directly to a worthy cause, looks to have had a relatively large effect on consumer behaviour given the small size of the levy. This type of ‘complex intervention’ has also been shown to be successful in economic studies of levies on alcohol.”

The study also found there was a general decrease in the numbers of non-alcoholic beverages sold per customer, with the exception of fruit juice, which increased by 22% after six months. Sales of diet cola and bottled waters also declined.

Professor Cummins said: “A possible reason for this decline could be that more people were choosing tap water, but data on tap water orders was not available as it was not recorded on the restaurant’s sales system. Overall, our study suggests that a small levy on sugar-sweetened drinks sold in restaurants, coupled with complementary activities, may have the potential to change consumer behaviour and reduce the consumption of these drinks which are associated with major health risks.”

Dr Laura Cornelsen, Assistant Professor in Public Health Economics and MRC Career Development Fellow at the London School of Hygiene & Tropical Medicine, co-led the study. She said: “This study raises interesting questions about how fiscal interventions may work beyond the effect of the price increase. For example, how big a role did the menu change or the text introducing the levy play? Taxes and levies on sugar-sweetened drinks are regularly framed as health related measures and such framing might be having an effect of its own, including on both the quantity and the range of beverages sold.”

Professor Martin White, Director of the NIHR's Public Health Research (PHR) Programme and a study author, is conducting an evaluation of the impact of the UK Government's forthcoming levy on sugary drinks with Professor Cummins and others.

Professor White said: "This is an important piece of research which explored how a fiscal intervention and associated publicity affected consumer choices in a national restaurant chain. Further research is needed to work out how such levies can best be used in different settings and circumstances. Obesity, heart disease, diabetes and tooth decay are serious conditions that have a huge impact on health. Population interventions of this sort could be vital in combating them."

The authors acknowledge limitations of their study including the fact that by taking advantage of a change that was already happening, they could only directly observe the changes that occurred, not necessarily the causes. The study was also undertaken in just one chain of restaurants.

Publication

Laura Cornelsen, Oliver T Mytton, Jean Adams, Antonio Gasparrini, Dalia Iskander, Cecile Knai, Mark Petticrew, Courtney Scott, Richard Smith, Claire Thompson, Martin White, Steven Cummins. Change in non-alcoholic beverage sales following a 10-pence levy on sugar-sweetened beverages within a national chain of restaurants in the UK: interrupted time series analysis of a natural experiment. Journal of Epidemiology & Community Health. DOI: 10.1136/jech-2017-209947.