- Matthew Thompson, senior clinical scientist1,
- Carl Heneghan, reader in evidence based medicine1,
- Deborah Cohen, investigations editor2
- 1Department of Primary Care Health Sciences, University of Oxford, Oxford OX2 6GG, UK
- 2BMJ, London WC1H 9JR
- Correspondence to: M Thompson
Back in January 2006, the European Union decided to adopt legislation to assess health and nutrition claims related to foods.1 The EU regulation aims to ensure that “claims made on food labelling and advertising regarding nutrition and health are meaningful and accurate, and can thereby help consumers in making healthy diet choices.” From the end of 2012, all claims used to market and advertise a product will need to be approved.
The body responsible for evaluating the scientific basis of health claims is the European Food Safety Authority (EFSA). Their remit includes health claims related to the “roles of nutrients or substances in growth, development or functions of the body, psychological or behavioural functions, or slimming or weight control.”2 This includes claims related to sports supplements and sports drinks.
After the legislation was passed, individual member states supplied the European Commission with a list of over 44 000 health claims. After duplicate or overlapping claims were removed, these were narrowed down to a final list of 4637 claims; EFSA has evaluated 2758 of them and has published scientific opinions on 341.1
How does EFSA undertake scientific evaluations?
EFSA uses five panels to evaluate the claims, one of which is the panel on dietetic products, nutrition, and allergies (NDA).3 Scientific evaluations “should be scientifically substantiated” by “taking into account the totality of the available scientific data, and by weighing the evidence,” in addition to reviewing the quality of the evidence. Evaluations are expected to show:
The claimed effect of the food is beneficial for human health
A cause and effect relation between consumption of the food and claimed effect (such as strength, consistency, specificity, dose-response, and biological plausibility)