The issue was highlighted at a recent conference of the Food & Health Forum, at which scientists questioned whether human clinical trials were appropriate for products aimed at 'healthy' individuals, and consumed over lifetimes.
"When it comes down to nutrition and measuring the positive health effects, the randomised control trial is, in general, not very useful," said Renger Witkamp, professor of nutrition and pharmacology at Wageningen University in The Netherlands.
"Basically we are trying to target a population that is apparently healthy," he said. The problem, he added, is that pharmacological and disease biomarkers are "hardly useful in nutrition".
"Health is dealing with the elasticity of the homeostasis [the body's ability to keep itself in equilibrium], the resilience; we are dealing with multiple biomarkers; we have to look at several at the same time."
He added: "New concepts in health are coming, based on the methodology of genomics. We should be more flexible in looking at the boundaries between health and disease."
Discussing the effects of wholegrain on cardiovascular health, Chris Seal, professor of food and human nutrition at the University of Newcastle upon Tyne, reported that most published research was based on observational studies.
This made: "establishing causal relationships very difficult", he said.
Seal added: "Changing the diet in free-living intervention studies is not easy: particularly in complex food-based systems." He asked if intervention studies should be weighted more than observational studies and if it was possible to design more effective intervention studies.
Professor Hans Biesalski from the Department of Biological Chemistry and Nutrition at the University of Hohenheim in Stuttgart, Germany, gave a critical review of the controversy over antioxidants and health claims and called on studies to show their efficacy.
"The majority of studies [to date] were with high dose supplementation," said Biesalski. "Here we are talking about pharmacology, not nutrition."