expert opinion

The relation between intake of a food constituent and a beneficial health effect

August 1, 2011

file

The Hohenheim Consensus Conference on Scientific Substantiation of Health Claims: Evidence-based Nutrition

“The public’s increasing awareness that nutrition and dietary components contribute significantly to personal well-being and health, and to public health overall, opened up opportunities and a need to make claims about the benefits of particular foods or of foods components. In consequence, there was an increase in claims about the relations between the consumption of particular food categories, foods, or food constituents and a specific benefit to health. The quality of the evidence underpinning such claims was variable, although some claims derived from generally accepted knowledge, which may not be as sound as nutritional science accepts. Others were based on selective and non-robust extrapolations from observational studies. Thus a range of products appeared spanning a spectrum of traditional food products that had been fortified or enriched with specific components to completely novel products. These developments coincided with national and international reviews of reference intakes for nutrients. These reviews demonstrated the limitations of the information from which reference intakes were induced, and they highlighted a need to strengthen the science base.

There emerged a movement, evidence-based nutrition, akin, but not necessarily analogous to evidence-based pharmacology and evidence-based medicine. The primary goal of nutrition is to maintain, or if possible to improve, health. This is an essential difference from pharmaceuticals, which are generally developed to treat, cure, or prevent disease. The term beneficial physiologic effect refers to the demonstrable effect(s) of a food or food constituent for which a health claim is made. Beneficial physiologic effects go beyond traditionally accepted nutritional effects and their validity should be substantiated for the general population or a population subgroup. This requirement to demonstrate a beneficial physiologic effect(s) of a food or food constituent in the general healthy population presents an enormous scientific challenge to nutrition science. In fact, this task is more complicated than assessing the efficacy of most pharmaceuticals, for which effects on clinical or surrogate endpoints are often easier to establish. The question is how to develop new research strategies in nutrition science aiming to demonstrate beneficial physiologic effects in persons who are healthy.

The assessment of the relation between intake of a food or food constituent and a beneficial health effect is a complex process. It can be performed based largely on the outcome of randomized controlled trials (RCTs), and it is not disputed that well-conducted RCTs offer the strongest support for cause-and-benefit relations. However, this is the case only if RCTs meet several essential criteria. Furthermore, RCTs are not always possible or available and are seldom capable of addressing the broad field of interactions of the human physiology. RCTs are a difficult tool for nutritional studies, unless the food component of interest has much in common with a pharmaceutical compound. Therefore, the presence or absence of positive RCT data should not be a gate-keeper criterion in the evaluation of a health claims assessment process.

Several case studies were discussed to address under what conditions the totality of the current scientific data and supportive evidence from generally accepted sources and recognized scientific institutions could find more suitable alternatives to RCTs. While we await, for example, the outcomes of ongoing RCTs assessing the efficacy of carotenoids in the prevention and treatment of age-related macular degeneration (AMD), the potential public health benefits, scientific plausibility, and excellent safety records of lutein and zeaxanthin are supporting the recommendation that dietary consumption and supplementation with these compounds should be encouraged. One may understand why an eye care specialist, in the absence of convincing evidence from RCTs but in the presence of a biologically plausible rationale and significant body of supporting evidence (albeit not from RCTs), might recommend antioxidant supplements that contain the macular carotenoids in view of the lack of other available putative or proved measures against AMD. The patient with intermediate or advanced AMD, who may have already lost vision in one eye, often wishes to actively participate in decreasing the risk against further visual loss and is unwilling to wait for a “conclusive” evidence base. Similarly, the “worried well” who have early signs of AMD or who have a strong family history of AMD are unlikely to be willing to wait for definitive RCTs applicable to them, especially in light of the excellent safety record of carotenoids. In the absence of definitive RCTs, we must view and assess the totality of the scientific data and weight of evidence as currently available, identifying and acknowledging such evidence from respected sources and scientific institutions, to find more suitable alternatives to RCTs to make important public health recommendations in a timely manner.

Many antioxidants exist in food or food supplements and might exert their antioxidant effects after ingestion within a complex pattern. Related to their nature as antioxidants, nearly all are claimed to decrease reactive oxygen species (ROS) and subsequently decrease oxidative damage and compensate oxidative stress. A major paradigm that justifies the use of antioxidants is that ROS exposure of cells and tissues results in oxidative damage, impairment of function, and at least a disease state. It has been shown in some studies that ROS are responsible for the development of different diseases such as cancer, neurological diseases, coronary heart disease, and other chronic degenerative diseases. The major question with respect to antioxidants and their “evidence of benefit” is whether the detection of an impact of antioxidants on oxidative stress might be sufficient to claim a beneficial effect or whether a combination of a disease-related marker and the antioxidative effect is needed. In cases of degenerative disease, the time from oxidative damage to occurrence of a disease-related biomarker is too long (years) to design a realistic human intervention study. Moreover, the development of a degenerative disease is not triggered by one effect alone. Dietary intakes, in particular of antioxidants, are a key determinant of gene expression, in part by their involvement in genomic stability. Long-lasting degenerative diseases, such as cancer, coronary heart disease, chronic inflammation, and accelerated aging, appear to be caused in part by damaged DNA which can be measured. Functional indices such as DNA damage may be useful biomarkers for defining adequate and optimal intakes of antioxidant micronutrients. These methods are particularly suitable for evaluating in relatively small-scale studies and can provide valuable information on the physiologic and metabolic mechanisms related to adequate and optimal intakes and the risk of long-latency chronic disease.

Research in the field of nutrition has entered a new stage, where it deserves its own nutritional methodo-logies that are distinct from and more diverse than evidence- based medicine. It requires redefining health in terms of an individual’s potential to adapt to internal and external stimuli rather than using the pharma-ceutical paradigm. New ways of addressing nutritional and functional effects of foods and food components are being developed and offer opportunities for future research. Any health claims assessment system that is not able to accommodate the wide spectrum of health effects should be adapted to reflect the science available. It should be capable of stimulating these developments toward new research and the development of innovative products and useful information to the benefit of the consumer, including but not limited to public health messages.”

July 2011

Biesalski H. K. (University of Hohenheim, Germany), Aggett P. J. (Lancaster University, UK), Anton R. (University of Strasbourg, France), Bernstein P. S. (University of Utah, USA), Blumberg J. (Tufts University, USA), Heaney R. P. (Creighton University, USA), Henry J. (Oxford Brookes University, UK), Nolan J. M. (Whitfield Clinic, Waterford, Ireland), Richardson D. P. (DPR Nutrition Limited, UK), van Ommen B. (TNO, The Netherlands), Witkamp R. F. (Wageningen University, The Netherlands), Rijkers G. T. (St. Antonius Hospital, The Netherlands), Zöllner I. (Baden-Wuerttemberg State Health Office, Epidemiology, Germany). 26th Hohenheim Consensus Conference, September 11, 2010. Scientific substantiation of health claims: Evidence-based nutrition. Nutrition. 2011.