Evaluation of dietetic product innovations: the relative role of preclinical and clinical studies.

A variety of systems are used to establish efficacy of food ingredients. Immortal human cell lines have the advantage of rapid throughput and often have the ability to point to mechanisms of action. Transgenic and natural variants of animals (usually rats and mice) have proven to be extremely useful in elucidating effects in vivo, although extrapolation of results to humans has risks. Animal models are also useful in establishing safety and toxic levels of ingredients. Human trials have the most relevance to society. Types of evidence for efficacy rise from improved status level in subjects as a result of eating food (long-chain polyunsaturated fatty acid, levels in erythrocytes), change in surrogate markers as a result of eating food (plasma cholesterol or glutathione peroxidase activity), change in a physiological outcome (such as visual evoked potential acuity or heart rate variability) through to the highest level of evidence, a change in a clinical outcome (improved global development, reduction in infections) established in randomized controlled trials. Ultimately, there is a need for tests of pragmatic interventions that can easily be incorporated into usual dietary practices of the culture in which it is tested.