* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Approach of the US Food and Nutrition Board to daily
Survey
Document related concepts
Food safety wikipedia , lookup
Plant nutrition wikipedia , lookup
Calorie restriction wikipedia , lookup
Malnutrition wikipedia , lookup
Obesity and the environment wikipedia , lookup
Malnutrition in South Africa wikipedia , lookup
Gastric bypass surgery wikipedia , lookup
Food politics wikipedia , lookup
Food coloring wikipedia , lookup
Food studies wikipedia , lookup
Academy of Nutrition and Dietetics wikipedia , lookup
Saturated fat and cardiovascular disease wikipedia , lookup
Human nutrition wikipedia , lookup
Transcript
European Journal of Clinical Nutrition (1999) 52, 786±791 ß 1999 Stockton Press. All rights reserved 0954±3007/99 $15.00 http://www.stockton-press.co.uk/ejcn Approach of the US Food and Nutrition Board to daily nutrient requirements: `A useful basis for the European discussion on risk assessment of nutrients?' Report on a Workshop organized by the European Academy of Nutritional Sciences (EANS) and TNO Food and Nutrition Research Institute, 11 December 1998, Brussels JH Brussaard1*, H van den Berg1, RJJ Hermus2 and P Walter3 1 TNO Nutrition and Food Research Institute, Zeist, The Netherlands; 2Vice President of EANS, TNO Nutrition and Food Research Institute, Zeist, The Netherlands; and 3President of EANS, University of Basel, Switzerland Objective: To discuss the approach of the US Food and Nutrition Board (FNB) to daily nutrient requirements and its relevance for Europe. Setting: A workshop for experts from academia and regulatory bodies, together with members from the FNB Dietary Reference Intakes Subcommittee on Upper Reference Levels of Intake, and from the B-vitamins Expert panel. The workshop was organised by the European Academy of Nutritional Sciences and TNO Nutrition and Food Research Institute (The Netherlands). Conclusions: Classical approaches to recommended dietary allowances (RDA) are no longer satisfactory, because they do not take into account newly emerging science with regard to intakes beyond the RDA. There is an urgent need for global harmonisation of criteria used for assessing adequacy of intake and for harmonisation of terminology. For sound advice to consumers, the development of tolerable upper intake levels is necessary. The discussion of principal issues relating to criteria for nutrient adequacy should be pursued on an international level. Sponsorship: Roche Vitamins Europe Ltd, Birsfelden, Switzerland Descriptors: nutrient requirements; dietary reference intakes; recommended dietary allowance; tolerable upper intake level Introduction In 1992, the Scienti®c Committee on Food (SCF) of the European Community expressed its opinion on nutrient and energy intakes for the European Community, at that time twelve Member States. The report provided nutritional recommendations to be used for different goals, including nutrition labelling and community programs on research and nutritional aspects. Excessive intakes were only discussed brie¯y. Experts of several Member States did not share the recommendations expressed by the SCF. National recommendations for daily nutrient requirement were issued by Member States in 1993 and thereafter, for example by Italy, Spain, the Nordic countries, while at present national recommendations are in preparation (for example, Germany, Austria, Switzerland, France, The Netherlands). However, the development of a (new) concept and of guidelines with respect to the safety and upper levels of intake of these nutrients was found to be necessary because the current activities regarding harmonisation on EU food legislation for vitamin and mineral supplements, and the addition of these essential nutrients to food. There*Correspondence: Dr JH Brussaard, TNO Nutrition and Food Research Institute, Dept of Consumer Research and Epidemiology, P.O. Box 360, 3700 AJ Zeist, The Netherlands. Received 31 March 1999; accepted 8 April 1999 fore, the European Communities may consider a new assessment of the 1992 recommendation timely and appraise new scienti®c evidence and insight in nutrients and their function. A similar discussion had already started in the US and Canada. After preparatory workshops organised by the Food and Nutrition Board (FNB) (Anonymous, 1994; Lachance, 1994), the Institute of Medicine (National Academy of Sciences, USA) installed a Standing Committee on the Scienti®c Evaluation of Dietary Reference Intakes (DRIs), chaired by V. Young. Expert panels were charged to review the scienti®c development regarding groups of nutrients, and two additional subcommittees were formed, one on Upper Reference Levels of Nutrients (chaired by I. Munro) and one on the Use of DRIs (chaired by S. Murphy). The ®rst two panels have issued prepublication reports: `Vitamin D, calcium and related minerals' (1997) and `Folic acid and other B-vitamins' (1998). To understand and discuss the concepts and rationale behind the FNB approach and to facilitate discussions between European experts from academia, industry and regulatory authorities, the European Academy of Nutritional Sciences (EANS) and TNO Nutrition and Food Research Institute organised this workshop (December 1998). In this workshop members of the FNB DRI Subcommittee on Upper Reference Levels of Intake, and from the B-vitamins Expert panel gave an overview on the European risk assessment of nutrients JH Brussaard et al concepts and activities in the USA, while European experts presented an overview on the past and current activities in Europe. In this paper, the presentations and resulting discussions are summarised. possible to deliberate more comprehensively and deeply, and slashing criticism afterwards is prevented. General concept and approach of the US Food and Nutrition Board (introduced by Prof. Vernon Young) Key issues in the development of a model to derive tolerable upper intake levels (UL) are the concept of safety versus risk, the limitations of traditional models and the unique characteristics of nutrients. The LJL refers to the maximum level of long-term daily nutrient intake that is unlikely to pose a risk of adverse health effects to almost all individuals in the speci®ed life-stage and gender group. It refers to all pathways of (oral) exposure combined, and is based on an evaluation conducted using the methodology for risk assessment. It is not intended to be a recommended level of intake, as there is no established bene®t for healthy individuals associated with nutrient intakes above the recommended dietary intake. In the USA, the main reason for a revision of the recommended dietary allowances (RDAs) (latest revision 1989) was the apparent evolution in the application of RDAs. RDAs are those levels of intakes judged by the FNB to be adequate to meet the known nutrient needs of practically all healthy persons. Thus, RDAs were to be applied to groups of healthy people and not individuals, and therefore set at levels that exceed the needs of most individuals to encompass the individual variability in nutrient requirements. However, RDAs are often considered to be the level of an essential nutrient that a healthy individual should consume on average over a period of time to ensure adequate and safe intake. It was felt that future RDA documents needed to provide more details about the actual derivation of the recommendations and give more explicit guidance in using the values for nutrition policy and other de®ned uses. Furthermore, suf®cient new knowledge has accumulated for selected nutrients and for certain subgroups, that would support an update and extension of the RDAs. Finally, the evolution of concepts for intake recommendations played a role. Especially, the consideration of reduction of risk of chronic disease as a criterion for adequacy of intake is new provided suf®cient data exist on ef®cacy and safety. However, incorporating the latter criterion appeared rather dif®cult because of lack of suf®cient data on several topics, such as the speci®city of effects as related to the intake of a single nutrient, dose-response relationships, information on the variability of the effective dose and the interaction between nutrients. Another problem appeared to be the insuf®cient information available for several subgroups such as adolescents and children. In developing DRIs,** the criteria to assess adequacy are of paramount importance. When international harmonisation of recommended intakes is pursued, this is one of the essential elements to agree upon. For example, in assessing the requirement for folate, reduction of risk for chronic disease was evaluated as criterion for adequacy. However, insuf®cient data were available for a quantitative assessment and the committee reverted to other classical indicators for adequacy such as the erythrocyte folate concentration. On the other hand, in the assessment of the calcium requirement, the scienti®c evidence was judged to be strong enough to use desirable calcium retention during growth and minimising bone loss during adulthood as a criterion for adequacy. An essential element of the FNB approach is given by the transparency of the process. After review of the available database and comments from experts, further advice is being solicited through workshops, meetings and correspondence; ®nally, the National Research Council reviews all information. In this way, it is ** [In the new FNB reports on nutrient requirements, the term DRI is used as a collective term for estimated average requirement (EAR), recommended dietary allowance (RDA), adequate intake (AI) and tolerable upper intake level (UL) (Anonymous, 1997a)] A model for development of tolerable upper intake levels (introduced by Dr. Ian Munro) Safety and risk Munro emphasised the important difference between safety and risk. Safety can be seen as an intellectual concept, not an inherent biological property. A safe level refers to a point on a continuum, and is affected by economic, political, and cultural in¯uences. Risk on the other hand, is de®ned as the probability of an adverse effect occurring at some speci®ed level of exposure. Risk assessment is viewed as a scienti®c exercise, less in¯uenced by value judgments. It is a systematic means of evaluating the probability of occurrence of adverse health effects in humans from excess exposure to a nutrient (Anonymous, 1997b). Limitations of traditional models Traditional models for the development of upper levels of intake focus on establishing safe, rather than tolerable upper intake levels, and rely heavily on animal data following the concept of establishing an acceptable daily intake (ADI). Typical documents required for approval of ingredients include studies on acute and sub-chronic and long-term toxicity, on carcinogenicity, multi-generation teratology and genotoxicity, as well as studies on absorption, distribution, metabolism and excretion. From this database a no-observed-adverse-effect-level (NOAEL) is derived and the ADI (mg=kg human body weight) is calculated by dividing the NOAEL (mg=kg animal body weight) by the appropriate safety factor. This factor is usually set at 100, to compensate for the extrapolation from animal to man, `intra-human sensitivities' (children, elderly, gender) and differences in test population size. Sporadic intake in excess of the ADI is not considered a risk in view of the cautious principles used in setting such levels of intake. If the same concept were applied to essential nutrients, it is not unlikely that for some nutrients the ADI might come down to an intake below the RDA, as has been shown for zinc. Therefore, for essential nutrients other procedures have to be followed. Unique characteristics of nutrients For nutrients no, or insuf®cient, dose-response data are actually available. There are only very few clinical studies on the relation between dose and biomarkers of adverse 787 European risk assessment of nutrients JH Brussaard et al 788 effects and the relevance of animal data obviously is rather limited. Case reports on medical applications of nutrients describe high intakes of nutrients, but the resulting data should not be applied to the general population. Furthermore, only few chronic studies and few surveillance studies with nutrients are available in man and animals to have a NOAEL evaluated. Besides, the quality of many of the available animal studies can be criticised, due to lack of published details, for example on age and weight of the animals, and the composition of the diet. Another drawback is that available databases often concentrate on supplement intake, but not on total intake. Finally, there are signi®cant differences in bioavailability of nutrients from different sources and mode of application. Risk assessment for nutrients Risk assessment models are essentially a way of organising data. A risk assessment model widely used in public health and regulatory decision making is that proposed by the National Research Council (NRC, 1983a,b) Using this approach the steps of risk assessment as applied to nutrients are as follows. Step l. Hazard identi®cation. This involves the collection, organisation and evaluation of all information pertaining to the adverse effects of a given nutrient in humans. It investigates causality, the relevance of experimental data, mechanism of toxic action, quality and completeness of the database and identi®cation of distinct and highly sensitive population groups. The result is a summary of the evidence concerning the capacity of the nutrient to cause one or more types of toxicity in humans. Step 2. Dose-response assessment. In this step, the relationship between nutrient intake (dose) and adverse effect (in terms of incidence and severity) is determined. It involves data selection, including critical endpoints and identi®cation of NOAEL or, where the NOAEL cannot be determined, the lowest-observed-adverse-effect-level (LOAEL). Also other sources of uncertainty have to be evaluated such as inter-individual variation in sensitivity, extrapolation from animal data to humans, and the use of subchronic NOAEL to predict chronic NOAEL. As a result of this step, the UL is calculated, possibly different ULs for various life-stage groups, by dividing the NOAEL by the relevant uncertainty factor (UF). The range between UL and NOAEL may be smaller or wider, depending on the level of uncertainty. The UF for nutrients is typically less than 10 depending on the quality and nature of the data and the observed adverse effects involved. Also a smaller UF may be used when the adverse effects are extremely mild and reversible. At this moment, extrapolation from adults to children is mostly done on a body weight basis, while obviously a function related basis would be better. This is one of the issues still under discussion. Another open discussion at this stage of the process is whether it is necessary to have ULs for all nutrients, or whether for some nutrients it may be suf®cient to de®ne another value based on a speci®c physiological indicator. Step 3. Exposure assessment, that is an evaluation of the distribution of total daily nutrient intakes among members of the general population. Step 4. Risk characterisation. This step summarises the conclusions from steps 1 through 3 and evaluates the risk. Generally, the risk is expressed as the fraction of the exposed population having nutrient intakes in excess of the estimated UL. Where possible, the magnitude of the excess is identi®ed and scienti®c uncertainties are described. It should be noted that risk is assessed as such and not weighed against possible bene®ts. Also, a discussion of recommendations for reducing risk is not part of the risk assessment process, but belongs to risk management. Case illustrations for upper tolerable levels of intake ± vitamin B6, folic acid and niacin (introduced by Prof. Robert Russell) The `NRC format' for risk characterisation for nutrients has recently been applied to the B-vitamins. For vitamin B6, the critical effect level on which the tolerable upper intake was based was sensory neuropathy. For this critical end point, a NOAEL of 200 mg=day was identi®ed and an uncertainty factor of 2 was selected because of the small number of studies available which used doses of pyridoxine less than 500 mg=day. Thus, an upper tolerable level of intake of 100 mg=day was obtained for vitamin B6. As with all nutrients, the upper levels for children and adolescents were calculated based upon the upper levels of adults, adjusted for body weight. The highest mean intake of vitamin B6 (including supplements) according to NHANES III was 9 mg=day and the highest reported intake at the ninety-®fth percentile was 21 mg=day (in pregnant women). Therefore, it can be concluded that the risk of adverse effects resulting from excess intake of vitamin B6 from food and supplements appears to be very low at the highest intakes noted above. For (synthetic) folic acid, the critical adverse effect indicator used was precipitation or exacerbation of neuropathy by folic acid in vitamin B12 de®cient individuals, although there is only limited, but suggestive evidence. It is important to note, that the upper tolerable level of intake was set not for food folate but rather for synthetic folic acid. Due to lack of data from the literature, the LOAEL rather than the NOAEL was used as a basis. A lowest observed effect level of 5 mg of folic acid was derived from 100 reported cases of folic acid treatment in dosages of 5 mg or more causing neurologic progression in patients with vitamin B12 de®ciency. However, the masking of a vitamin B12 de®ciency state by folate must be considered as lack of appropriate neurological examination of the patient. An uncertainty factor of 5 was used based on the severity of the possible neurologic complications (erring on the side of safety) and because of the fact that a LOAEL rather than a NOAEL was used. Thus, the tolerable upper level of intake for adults was set at 1 mg=day for synthetic folic acid. It could be argued, that this level may not be compatible with levels of folic acid possibly associated with bene®cial effects on cardiovascular risk. Presently, the conclusion of the FNB is that there is no dose-response relation for the association between dietary folate and blood concentrations of folic acid or homocysteine. However, ongoing studies on cardiovascular risk prevention use daily supplements of 2.5 mg folic acid, which is higher than the present UL. In the case where the results of these studies are positive, the UL for folic acid will certainly need to be re-considered and potential solutions such as addition of vitamin B12 to food along with folate might have to be addressed. For the moment, however, the UL remains at 1 mg=day. The highest ninety-®fth percentile for total folate intake was found to be 983 mg=day (non pregnant women); European risk assessment of nutrients JH Brussaard et al from food alone it was 438 mg=day, indicating a low level of risk. Another approach to risk characterisation in this case is to start from the level of food forti®cation. Mandatory folic acid forti®cation in grain cereals in the USA is currently 140 mg=100 g cereal. According to the food guide pyramid the upper level of cereal consumption is 11 servings=day, leading to a supplemental intake of folic acid of 440 mg=day. Therefore, the intake of synthetic folic acid is likely to exceed 1 mg=day only in combination with supplement use. Another consideration is that among those groups where a high intake of cereals is likely, for example, adolescents, the prevalence of vitamin B12 de®ciency is extremely low. Most of the data on adverse effects from niacin are from studies involving patients with high lipid levels who were treated with pharmacological preparations of nicotinic acid. The FNB panel considered that the generic name niacin might be interchangeable with nicotinic acid. The most sensitive but benign adverse effect reported is ¯ushing, resulting either in withdrawal from treatment or a change in the dosing pattern. Using this critical effect, a LOAEL of 50 mg=day was derived from the available database. However, the uncertainty factor chosen was small (1.5) because of the transient and rather benign nature of the ¯ushing effect; thus, the tolerable upper level of intake for niacin was de®ned as 35 mg=day. The highest mean intake of niacin found was 39 mg=day among men aged 31 to 50, with on average 28 mg from food and 11 mg from supplements. The conclusion therefore is that a small percentage of the population is likely to exceed the upper tolerable level of intake for niacin. The situation in Europe (introduced by Prof. Ruud Hermus and Dr A Somogyi) Contrary to the situation in the USA, European countries usually have their own national intake recommendations for energy and nutrients apart from the SCF recommendations expressed in 1992 (SCF Report, 1993). As illustrated in Table 1, there are large differences in recommendations between countries. This can be explained ®rst by differences in the conceptual framework for recommendations which are an element of nutrition policy, originally aimed at providing enough food for the population. More qualitative aspects of nutrition policy such as the risk reduction of chronic diseases have only recently been discussed as possibly important in establishing recommendations. The recommended intake values from different countries are sometimes based on original scienti®c data reviewed by expert committees, in other cases on a review of the recommendations from other national or international expert groups, or on a combination of both approaches. Table 1 RDAs for adults of selected nutrients in various countries (mg=day) European Commission, 1992 France, 1992 Germany, 1991 The Netherlands, 1989 Nordic countries, 1996 UK, 1991 Vitamin C (men and women) men women Calcium (men and women) 45 80 75 70 60 40 9.1 10 10 9 10 8.7 19.6 18 15 15 12 ± 18 14.8 700 900 900 700 ± 900 800 700 RDA Recommended dietary allowance. Iron Secondly, the starting point for setting recommendations generally is an assessment of the physiological requirement of a nutrient, plus margins for variability, uncertainty and bioavailability. This leads to several differences as, for example, the width of the safety margins chosen differs between countries. Another factor of uncertainty is that the period over which a requirement is established is seldom de®ned. The most important cause of differences, however, form the criteria used to de®ne adequacy, depending on whether they are based on prevention of de®ciency, tissue saturation, body stores, the maximisation of a physiological effect (for example, enzyme activity) or the minimisation of the risk of a chronic disease. Furthermore, there are differences with respect to classi®cation into age groups, for which the scienti®c reason is often less clear. Finally, application of the recommendations is different in the various countries. This can be illustrated by comparing the situation in three European countries, The Netherlands, the UK and the Nordic countries, respectively (Netherlands Food and Nutrition Council, 1992; DHSS,1991; Nordic, 1996). There are more or less explicit differences in the de®nition of values, the criteria for adequacy used, the utilisation aimed at, and the time period concerned. In the Netherlands, RDAs were developed for use in food supply planning and for the evaluation of food consumption. Based on data on the individual minimum requirement of a nutrient, its bioavailability and the metabolic interactions of nutrients, an average minimum requirement was assessed. This average minimum requirement together with data on the variations in the minimum requirement (usually 2 s.d. of the average minimum requirement) was used for the de®nition of the adequate level or range of intake. These values (average minimum requirement and adequate level of intake) are suitable for the evaluation of food consumption data. RDAs, especially in view of food supply planning, were based on the adequate level=range of intake, combined with data on dietary habits, acceptance, etc. In this way, RDAs ful®l the nutritional needs of practically all healthy persons in a population. In the UK, the Panel on Dietary Reference Values recognised that earlier de®nitions of recommended dietary intake values gave rise to a lot of misunderstanding. In trying to reduce the chance of misunderstanding it was decided to introduce new terms and to set a range of reference values (not recommendations!) based on the distribution of requirements for each nutrient. These reference values are to be used in a variety of ways such as the evaluation of dietary surveys, or for food labelling or to provide guidance on diet composition and meal provision for groups and individuals. The dietary reference values (DRVs) are de®ned as follows: the estimated average requirement (EAR) indicating the mean requirement, the reference nutrient intake (RNI) indicating that level above which intakes will be almost certainly adequate and the lower reference nutrient intake (LRNI), indicating the level below which intakes will be almost certainly inadequate for most individuals. In the UK, as well as The Netherlands, prevention of clinical signs of de®ciency, maintaining nutrient balance and allowing for a certain level of nutrient storage were used as criteria of adequacy. The claimed bene®cial effects of higher intakes of some nutrients were considered to be not suf®ciently scienti®cally based to be acceptable as criteria for adequacy. 789 European risk assessment of nutrients JH Brussaard et al 790 The Nordic countries followed a slightly different approach. In 1996, they published a set of Nordic nutrition recommendations (NNR) to be understood as basis for the planning of a diet which satis®es the nutritional needs (that is, covers the physiological requirement for growth and function) and is a prerequisite for overall good health and contributes to risk reduction of diet-associated diseases. The criteria for these guidelines were de®ned for each nutrient separately, which allows reassessment when new scienti®c knowledge emerges. NNR are to be used in several ways: as guidelines for planning of diets, as the basis for teaching and information, as the basis for food and nutrition policy and as the basis for evaluation of intake. Recommendations are given several levels: (1) `recommended intakes of nutrients' expressed as average daily intake over time, for use at planning of diets for groups; stating that the requirement is lower for almost all individuals; (2) `recommended nutrient density' to be used for planning of diets for groups of individuals over 7 y of age with a heterogeneous age composition, the values being adapted to the person with the highest requirement; (3) `lower limits' of average daily intake of nutrients, to be used only for the evaluation of results from dietary surveys. It is indicated that prolonged intake below these levels may induce a risk of de®ciency, while a higher intake is no guarantee against de®ciency symptoms in single individuals. Finally, upper limits for average daily intake of nutrients for adults were de®ned, stating that if the supply from the diet and vitamin and=or mineral supplements exceed these values there is a risk of undesirable effects. These values are to be used for the evaluation of intake of single individuals. In principle, the EU Scienti®c Committee on Food (SCF) used a very similar approach as the UK Panel on dietary reference values (moving away from single RDA values towards three values to indicate the spread of needs) with different names: average requirement (AR) corresponding to the UK EAR and population reference intake (PRI), corresponding to the UK RNI. The PRI (SCF) and the RNI (UK) correspond conceptually with the traditional RDA, and when diets are being designed or food supplied it should contain nutrients at this level over a period of time to ensure negligible risk of nutrient de®ciency in any person. Recommendations by the EU serve the same purposes as in all other countries, but besides have different objectives: they must be suitable for use in all Member States, regardless of the national nutritional policy or dietary habits, and they must enable free circulation of foodstuffs, which is an important principle. The use of these recommendations in labelling is therefore an important aspect. Risk management is becoming of more importance for the EU. In this respect, two steps are considered of importance. First, the establishment of upper safe levels and the assessment of risk. Second, evaluation of available risk management. The de®nition of upper safe levels of intake is a topic presently discussed by a sub-committee of the Scienti®c Committee for Food. To coordinate EU efforts in the ®eld of risk assessment, the Unit Evaluation of Health Risk was established in 1998. Although regulations should be based on scienti®c evidence it is clear that in trade disputes economic consideration also have a role. However, food safety and consumer protection have always been viewed as cornerstones of regulatory policy in the EU. Bene®cial effects of high intakes can be seen as the counterpart of risk and among some parts of the society there is a demand to allow health-related claims for (food) products with high levels of certain nutrients. However, at present, claims on medical issues are forbidden in the EU and this is unlikely to change in the foreseeable future. Conclusions (by Prof. Paul Walter) The discussion on nutrient recommendations or nutrient intake values relates to safety, scienti®c, regulatory and political issues. Several challenges can be identi®ed and it seems clear that the classical RDA approach is no longer satisfactory, because it does not take into account newly emerging science with regard to intakes beyond the RDA. In the USA, the attempt to include new science on risk reduction of chronic diseases in establishing new intake recommendations is courageous, but dif®cult to achieve. The results of the Panel on Dietary Antioxidants will be very important in this respect (expected by September l999). There is an urgent need for global harmonisation of criteria used for assessing adequacy of intake and for terminology. A recent development is the increasing importance of forti®ed foods and food supplements. This necessitates open and understandable information for the consumer. The consumption of a healthy, balanced diet may turn out to be dif®cult for several segments of the population, such as the aged, adolescents, or pregnant women, and selected consumption of food enriched with some nutrients may become a necessary tool. This may require permission to use health-related claims for forti®ed=functional foods. However, combined use of forti®ed food and food supplements might lead to safety considerations. Therefore, it is important to develop tolerable upper intake levels, in order to advise the consumer that beyond a certain intake of the respective nutrient the risk of an adverse effect will be increasing. There is a rather broad public interest for these topics and informing for the consumers is of paramount importance. The discussion of principal issues relating to criteria for nutrient adequacy should be pursued. The formation of a special working group led by the European Academy of Nutritional Sciences (EANS), preferably in close cooperation with the Food and Nutrition Board or the American Society for Nutritional Sciences (ASNS) in the USA, might deliver an important contribution to progress in this area. References Anonymous (1994): How should the recommended dietary allowances be revised? A concept paper from the Food and Nutrition Board. Nutr. Rev. 52: 216 ± 219. Anonymous (1997a): Dietary reference intakes, Nutr. Rev. 55: 319 ± 351. Anonymous (1997b): A model for the development of tolerable upper intake levels. Nutr. Rev. 55: 342 ± 351. DHSS (1991): Department of Health and Social Security. Dietary Reference Values for food energy and nutrients for the United Kingdom. Reports on Public Health and Medical Subjects, 41. London: HMSO. Lachance P (1994): The RDA concept: time for a change? Nutr. Rev, 52: 266 ± 270. Netherlands Food and Nutrition Council (1992): Recommended diet allowances 1989. Voorlichtingsbureau voor de Voeding, den Haag. Nordic Nutrition Recommendations (1996): Scand. J. Nutr. 40: 161 ± 165. NRC (1983a): Risk assessment in the federal government: managing the process. Washington, DC: National Academy Press. European risk assessment of nutrients JH Brussaard et al NRC (1983b) Science and judgment in risk assessment. Committee on risk assessment of hazardous air pollutants. Board on environmental studies and toxicology. Washington, DC: National Academy Press. Vitamin D, calcium and related minerals (1997): Prepublication copy, Institute of Medicine, Food and Nutrition Board. Dietary References Intakes for Calcium, Phosphorous, Magnesium, Vitamin D, and Fluoride. National Academy Press, Washington, DC. Folic acid and other B-vitamins (1998): Prepublication copy, Institute of Medicine, Food and Nutrition Board. Dietary References Intakes for Thiamin, Ribo¯avin, Niacin, Vitamin B6 Folate, Vitamin B12, Pantothenic Acid, Biotin and Choline. National Academy Press, Washington, DC. FSCT Report (1993) Reports of the Scienti®c Committee on Food (Thirty®rst series): Nutrient and energy intakes for the European Community. Commission of the European Communities, Directorate ± General Industry. Luxembourg: Of®ce for Of®cial Publications of the European Communities. Appendix European Academy of Nutritional Sciences (EANS) and TNO Nutrition and Food Research Institute Workshop, 11 December l998, Brussels. Approach of the US Food and Nutrition Board to daily nutrient requirements: `A useful basis for the European discussion on risk assessment of nutrients?' Scienti®c Programme Committee P Walter, President of EANS, University of Basle, Basle, Switzerland RJJ Hermus, Vice-president of EANS, TNO Nutrition and Food Research Institute, Zeist, the Netherlands H van den Berg, TNO Nutrition and Food Research Institute, Zeist, the Netherlands JH Brussaard, Scienti®c Secretary, TNO Nutrition and Food Research Institute, Zeist, the Netherlands Speakers RJJ Hermus, Vice-president of EANS, TNO Nutrition and Food Research Institute, Zeist, the Netherlands I Munro, Chair of Sub-committee on upper reference levels of nutrients, Mississauga, Ontario, Canada RM Russell, Member of B-vitamin Panel of FNB, Tufts University, Boston, USA A Somogyi, Head of Unit Evaluation of health risks, DG XXIV, European Commission, Brussels, Belgium VR Young, Chair DRI Standing Committee on the Scienti®c Evaluation of Dietary Reference Intakes, Massachusetts Institute of Technology, Cambridge, MA, USA P Walter, President of EANS, University of Basle, Basle, Switzerland 791