Previous studies based on consumer interviews suggested the FSA-promoted system was the best-understood labelling scheme. But this week, the first report based on hard sales data was published in the World Health Organisation's Health Promotion International journal and it painted a very different picture.
Researchers tracked ready meal and sandwich sales at an unnamed UK retailer for a month before and a month after the introduction of the labelling, and found no significant difference indeed, the healthiest two sandwiches experienced the sharpest sales drop of any product.
"This study found the introduction of a system of traffic light labels had no discernible effect on the relative healthiness of consumer purchases," it said.
The report's co-author, Mike Rayner of Oxford University's public health department, was previously an advocate of traffic light labelling. In a 2008 report for the European Heart Network he and his co-authors concluded: "Traffic lights are helpful and can be applied across all foods and enable consumers to evaluate food."
Rayner had set out to confirm supermarket findings suggesting traffic lights had beneficial effects. However, he was only able to replicate these findings by being selective with the data. Further long-term research, based on actual sales data rather than interviews, was necessary to judge whether the controversial labels were useful, he said, slamming the FSA for not persuading Tesco and Sainsbury's to release sales data that could benefit public health. The FSA responded that it was in conversation with retailers.
A spokeswoman for Sainsbury's insisted the retailer was committed to traffic lights. "This research is on a rather limited range of ready meals and sandwiches, over a short period of time, whereas we now have traffic lights on about 5,000 own-brand items."
The Food and Drink Federation said the research vindicated its concerns about the traffic-light system's ability to improve nutrition.
"This report makes a valuable contribution to our understanding of the role of front-of-pack nutrition labelling," said communications director Julian Hunt. "Our concerns are well documented and this new study only serves to reinforce our belief that colour coding will neither improve consumers' food literacy nor help them make long-term behavioural changes."
Will the FSA halt its advice to ministers on traffic lights? (analysis; 7 November 2009)
Portion size: it may be on the label, but it’s not being read (1 August 2009)