European Journal of Clinical Nutrition (2002) 56, Suppl 2, S8–S11 ß 2002 Nature Publishing Group All rights reserved 0954–3007/02 $25.00 www.nature.com/ejcn ORIGINAL COMMUNCIATION Selection of relevant dietary indicators for health

L Steingrı´msdo´ttir1*, L Ovesen2, O Moreiras3 and S Jacob4 for the EFCOSUM Group

1Icelandic Nutrition Council, Reykjavı´k, Iceland; 2Danish Veterinary and Food Administration, Institute of Food Research and Nutrition, Søborg, Denmark; 3Departemento de Nutricio´n, Facultad de Farmacia, Universidad Compluyense de Madrid, Madrid, Spain; and 4Institute of Food Science, Laboratory for Human Nutrition, Swiss Federal Institute of Technology Zurich, Zurich, Switzerland

Objective: To define a set of dietary components that are relevant determinants for health in Europe. The selected components are intended to serve as nutrition indicators for health in the European Health Monitoring Programme and, as such, must be limited in number, relevant to health in Europe and practical for all involved countries with respect to data gathering and comparability of data. Design: Major nutrition factors were determined by reviewing relevant epidemiological and clinical literature in nutrition and health as well as referring to reports from international expert groups, including the report from the project Nutrition and Diet for Healthy Lifestyles in Europe. The selection of factors was also based on the relative ease and cost involved for participating countries to obtain comparable and valid data. Results: The selected factors include foods or food groups as well as individual nutrients. Biomarkers are suggested for selected nutrients that pose the greatest difficulty in obtaining valid and comparable data from dietary studies. Conclusions: The following list of diet indicators for health monitoring in Europe was agreed upon by the EFCOSUM group in 2001, in order of priority: vegetables, fruit, bread, fish, saturated fatty acids as percentage of energy (%E), total fat as %E, and ethanol in grams per day. Biomarkers were suggested for the following nutrients: folate, vitamin D, iron, iodine and sodium. Energy has to be assessed in order to calculate %E from total fat and saturated fatty acids. Sponsorship: European Commission, DG SANCO F=3, Health Monitoring Programme. European Journal of Clinical Nutrition (2002) 56, Suppl 2, S8 – S11. DOI: 10.1038/sj/ejcn/1601423

Keywords: nutrition goals; dietary indicators; nutrition monitoring; food groups; nutrients; biomarkers

Introduction where major nutritional factors relevant to health in Europe One of the aims of the European Food Consumption Survey are defined. These include a report on Health and Nutrition Method (EFCOSUM) is to define a set of dietary components prepared by the French Presidency of the European Union that are relevant determinants of health in Europe. These (French Presidency, 2000), which refers to the report from dietary components are intended to serve as nutrition indi- the project Nutrition and Diet for Healthy Lifestyles in cators in the European Health Monitoring Programme and, Europe (Eurodiet Reports and Proceedings, 2001), and as such, should be limited in number, relevant to health and reports from the World Health Organization, Regional practical for all involved countries with respect to data Office for Europe (Development of the First Food and Nutri- gathering and comparability of data. tion Action Plan for the European Region, 2000; Health 21, Epidemiological and clinical research in the field of nutri- 2000). Other nutrition factors are still controversial or need tion has identified several important dietary factors relevant further study to be generally accepted as major determinants to the development of chronic diseases. For many major for health or disease aetiology. This paper gives some of the factors there is general consensus among scientists with background and rationale used by EFCOSUM participants for respect to their role in disease aetiology. This consensus is defining a limited number of nutrition indicators relevant to reflected in international reports and nutrition action plans, health in Europe.

Diet and health *Correspondence: L Steingrı´msdo´ttir, Icelandic Nutrition Council, Baro´nsstı´gur 47, 101 Reykjavı´k, Iceland. Cardiovascular disease (CVD) is the main cause of death in E-mail: [email protected] Europe, accounting for 43% of all deaths. However, there is a Relevant dietary indicators L Steingr´ımsdo´ttir et al S9 striking difference between European countries, the highest (excluding potatoes) for the general Dutch population. They rates being observed in Eastern Europe and the lowest in included all analytical studies published up to early 1998. France and the Mediterranean countries (Rayner & Peterson, There were 269 analyses on the association between intake of 2000). It is estimated that more than a third of cardiovas- fruit and vegetables and cancer (195 case – control; 74 pro- cular deaths of people under the age of 65 is attributable to spective), and 14 analyses on CVD (three case – control; 11 diet (Ferro-Luzzi & James, 1997). The most important dietary prospective studies). The preventable proportion of cancers factors are those that affect serum cholesterol levels, espe- and CVDs was calculated using three scenarios — best guess, cially the LDL cholesterol fraction (Adult Treatment Panel II, and optimistic and conservative guess — each representing a National Cholesterol Education Program, 1994). Thus, it is different set of assumptions or interpretations of the scien- widely accepted that diets high in saturated and trans unsa- tific evidence. The estimations were that an increase in the turated fatty acids increase the risk of CVD (Krauss et al, consumption of fruit and vegetables of 150 g per day (bring- 1996). However, some controversy has arisen regarding the ing the Dutch average intake to 400 g per day) would reduce beneficial effect of low total fat=high-carbohydrate diets on cancer deaths on average by 19% (best guess), ranging from CVD (Connors & Connors, 1997; Katan et al, 1997). Critics 6% (conservative guess) to 28% (optimistic guess). For CVD point out that low-fat diets may elevate triglyceride levels the proportion of preventable deaths was estimated at 16%, and lower not only LDL cholesterol but also HDL cholesterol. ranging from 6 to 22%. Thus, diets including moderate total fat, but low saturated The prevalence of obesity and overweight is rapidly fat, may result in an equally beneficial serum lipid profile increasing among all age groups in Europe, for both children and CVD risk to low fat diets. Epidemiological studies show and adults alike. Consequently, diseases resulting from that high consumption of fruit and vegetables is associated excess body fat, not the least diabetes type 2, are expected with a lowered risk of CVD, including both coronary heart to follow this development and become an ever-increasing disease and stroke (Klerk et al, 1998). The Dash study demon- burden on society and health care (WHO, 1998). Diabetes strated that dietary intervention with a combination diet prevalence is already rising rapidly in Europe, and it is that was high in fruit and vegetables and low-fat dairy estimated that at least 80% of diabetes type 2 is due to products decreased blood pressure in moderately hyperten- obesity and overweight. Lack of physical activity in daily sive subjects (Appel et al, 1997). Also diets including a weekly life combined with energy-dense, high-fat diets contributes fish meal are associated with lower cardiovascular risk to increased weight gain and obesity in most societies (Bray (Marckmann & Grønbæk, 1999), while sodium-rich diets & Popkin, 1998; Stookey, 2001). can contribute to hypertension and stroke (Sacks et al, Osteoporosis and the associated bone fractures among 2001). Finally, alcohol consumption is of significance to post-menopausal women and older men are predicted to ischaemic heart disease (IHD) and all-cause mortality as become an increasing burden on society as a result of the epidemiological studies have demonstrated that light to increasing age of European populations. Physical activity moderate drinkers of alcohol are at lower risk for IHD than and sufficient calcium and vitamin D from childhood to abstainers or heavy drinkers (Corrao et al, 2000; Sasaki, old age are preventive factors against osteoporosis (European 2000). Commission, 1997). Cancer accounts for 29% of all deaths in men and 22% in While the relative importance of nutritional deficiency women in the EU (French Presidency, 2000), and it is diseases has diminished in European populations in recent estimated that between 30 and 40% of these can be attrib- decades, certain nutrient deficiencies are still of concern. uted to dietary factors (Doll & Peto, 1981). Excess energy and This includes iron deficiency, which is prevalent among alcohol intake are risk factors for cancers of the mouth, young children and women of childbearing age in most pharynx, larynx, oesophagus and liver, while high intake European countries (UNICEF=WHO, 1999). Iodine deficiency of fruit and vegetables is associated with reduced risk of is also a health problem in many European countries, invol- cancers of the mouth, pharynx, oesophagus, stomach and ving 16% of the European region (Delange et al, 1993), and lungs (WCRF=AICR, 1997). Fruit and vegetables are rich only a few countries in Europe are free of iodine deficiency. sources of phytochemicals, with possible effects on slowing Lack of both of these micronutrients affects the health and or preventing chronic diseases, for instance, cancer and well-being of a large number of people. Iron deficiency CVD. However, the protective effects of vegetables and anaemia is associated with impaired immune function and fruit observed in epidemiological studies are not found diminished learning capacity in children and decreased using isolated constituents from plants (Lampe, 1999). physical fitness and work capacity in adults (WHO, 1989). Other factors still need clarification, such as the relationship Low, maternal, plasma free thyroxin due to dietary iodine between dietary fatty acids and cancers, as well as the deficiency is the most important preventable cause of mental possible protective role of fruit and vegetables in cancers of retardation in children (Monreale de Escobar et al, 2000); the colon, breast and prostate. however, the effect of mild iodine deficiency on mental A recent analysis by Veer et al (2000) aimed at quantifying function is debated (Laurberg et al, 2000). Iodized salt is an the public health benefit of increased consumption of fruit effective public health measure to introduce iodine in the and vegetables to the recommended level of 400 g per day general population, but few European countries require

European Journal of Clinical Nutrition Relevant dietary indicators L Steingr´ımsdo´ttir et al S10 iodination of salt. Folate is also a nutrient of special concern role of physical activity (this will be included as an indicator as a clear link has been established between the folate intake in the EUPASS project; European Physical Activity Surveil- of mothers and the occurrence of certain birth defects, lance System). In light of the need to keep the indicators especially defects in the development of the neural tube limited in number, calcium is not included in this limited set (Lumley et al, 2001). Biochemical indications of folate insuf- of relevant variables. ficiency are found in a large proportion of adults. Folate Foods or food groups may in many instances be easier to supplements have been shown not only to prevent birth compare between countries than nutrients. Some foods or defects but also to lower plasma homocysteine significantly, food groups may even give more relevant health information a major risk factor for CVD (Vermeulen et al, 2000). Green, than single nutrients. Fruit and vegetables are in this cate- leafy vegetables are a particularly rich source of folate as well gory, as more data are available concerning the importance as some fruits and bread, whereas in some countries cereals of fruit and vegetables in the diet than for single constituents are enriched and therefore are an important source. In the in these foods (Klerk et al, 1998). As consensus is lacking on United States folate fortification of flour has been reported to the beneficial effect of fruit juice on health, it was decided to result in a 50% reduction of hyperhomocysteinemia (Jacques exclude these beverages with the exception of freshly et al, 1999). A recent study has demonstrated a 19% decline squeezed fruit juice. To be specific, nuts, seeds and olives of the birth prevalence of neural tube defects after institution are not included in the fruit group, and pulses and potatoes of mandatory folic acid fortification in the US (Honein et al, are not included in the vegetable group. While certain other 2001). Some European countries — instead of food fortifica- food groups contain constituents positive for health, includ- tion — encourage all women of childbearing age to take folic ing vegetable oils, meat and meat products, dairy products, acid supplements as a preventive measure against neural etc, priority was given to fish (including shellfish) as well as tube defects in their offspring (Cornel & Erickson, 1997). bread, both being foods with great significance for health. Epidemiological studies have shown that populations con- suming fish on the average once or twice per week have a Diet indicators for health monitoring in Europe lower risk of CVD (Deckere et al, 1998), and consumption of The choice of dietary indicators must be governed not only bread is considered an indication of the proportion of by their relevance to health but also by the practicality of carbohydrate in the diet. obtaining reliable and comparable data in European coun- At the EFCOSUM meeting in Athens, 9 September 2000, tries. Such considerations limit, for example, the feasibility and in Brussels, 24 March 2001, the project group agreed of selecting many nutrients mentioned above in spite of upon the following list of diet indicators for health monitor- their importance for health. This includes intake data of ing in Europe, in order of priority. The selection is based on vitamin D, folate, sodium, iron and iodine, all of which are aspects of the report from the French presidency, the argu- of important health significance but are difficult to measure ments listed above as well as the need to keep their number in the diet in a comparable way between countries. Informa- in check. tion on vitamin D in food composition tables is incomplete; 1. Vegetables folate values of foods are not comparable in European 2. Fruit databanks due to different laboratory methods for determi- 3. Bread nation, and use of iodized salt and iron-fortified foods make 4. Fish intake estimates difficult for these nutrients. Finally, sodium 5. Saturated fatty acids, (percentage energy %E) intake is difficult to measure from food consumption data as 6. Total fat (%E) sodium content of otherwise similar foods varies greatly 7. Ethanol (g=day) according to the amount of salt used in preparation and at the table. Biomarkers should be considered for the following nutri- However, all of these nutrients can be estimated with the ents: use of biomarkers (Ovesen & Boeing et al, 2002). Considera- 8. Folate tion should therefore be given to including biomarkers of 9. Vitamin D these selected nutrients as dietary indicators instead of diet- 10. Iron ary intake data. Other important nutrients, while still posing 11. Iodine problems and needing harmonization and standardization 12. Sodium across countries, lend themselves better to comparison using intake data. These include total energy, total dietary fat, Energy has to be assessed in order to calculate %E from saturated fatty acids and ethanol. While total energy is not total fat and saturated fatty acids. a useful indicator by itself (body composition being pre- ferred), energy intake is needed to calculate the contribution References of saturated fatty acids as well as total fat to energy intake Adult Treatment Panel II, National Cholesterol Education Program (1994): Second report of the Expert Panel on Detection, Evaluation (total fat refers to total lipids). While dietary calcium has a and Treatment of High Blood Cholesterol in Adults. Circulation 89, role in the prevention of osteoporosis, priority is given to the 1333 – 1445.

European Journal of Clinical Nutrition Relevant dietary indicators L Steingr´ımsdo´ttir et al S11 Appel LJ, Moore TJ, Obarzanek E, Vollmer WM, Svetkey LP, Sacks FM, Lampe JW (1999): Health effects of vegetables and fruit: assessing Bray GA, Vogt TM, Cutler JA, Windhouser MM, Lin PH & Karanja mechanism of action in human experimental studies. Am. J. Clin. N (1997): A clinical trial on the effects of dietary patterns on blood Nutr. 70(Suppl), 475S – 490S. pressure. New Engl. J. Med. 336, 1117 – 1124. Laurberg P, Nøhr SB, Pedersen KM, Hreidarson AB, Andersen S, Bray GA & Popkin BM (1998): Dietary fat intake does affect obesity! Pedersen IB, Knudsen N, Perrild H, Jørgensen T & Ovesen L Am. J. Clin. Nutr. 68, 1157 – 1173. (2000): Thyroid disorders in mild iodine deficiency. Thyroid 10, Connors WE & Connors SL (1997): The case for a low-fat, high- 951 – 963. carbohydrate diet. New Engl. J. Med. 337, 562 – 563. Lumley J, Watson L, Watson M & Bower C (2001): Periconceptional Cornel CM & Erickson JD (1997): Comparison of national policies on supplementation with folate and=or multivitamins for preventing periconceptional use of folic acid to prevent spina bifida and neural tube defects (Cochrane Review). In The Cochrane Library, anencephaly (SBA). Terontology 55,134– 137. Issue 4. Oxford: Update Software. Corrao G, Rubbiati L, Bagnardi V, Zambon A & Poikolainen K (2000): Marckmann P & Grønbæk M (1999): Fish consumption and coronary Alcohol and coronary heart disease: a meta-analysis. Addiction 95, heart disease mortality: a systematic review of prospective cohort 1505 – 1523. studies. Eur. J. Clin. Nutr. 53, 5585 – 5590. Deckere EAM, Korver O, Verschuren PM & Katan MB (1998): Health Monreale de Escobar G, Obrego´n MJ & Escobar del Rey F (2000): Is aspects of fish and polyunsaturated fatty acids from plant and neuropsychological development related to maternal hypothyr- animal origin. Eur. J. Clin. Nutr. 52,749– 753. oidism or to maternal hypothyroxinemia? J. Clin. Endocrinol. Delange F, Dunn JT & Glinoer D (1993): Iodine Deficiency in Europe. A Metab. 85, 3975 – 3987. Continuing Concern. New York: Plenum Press. Ovesen L & Boeing H (2002): The use of biomarkers in multicentric Development of the First Food and Nutrition Action Plan for the studies with particular consideration of iodine, sodium, iron, European Region (2000): Report on a WHO Consultation, 8 – 10 folate and vitamin D. Eur. J. Clin. Nutr. 56(Suppl 2), S12 – S17. Malta November 1999. Copenhagen: WHO Regional Office for Rayner M & Peterson S (2000): European Cardiovascular Disease Europe. Statistics. British Heart Foundation, Oxford. Doll R & Peto R (1981): The cause of cancer: quantitative estimates of Sacks FM, Svetkey LP, Vollmer WM, Appel LJ, Bray GA, Harsha D, avoidable risks of cancer in the United States today. J. Natl. Cancer Obarzanek E, Conlin PR, Miller ER, Simons-Morton DG, Karanja N Inst. 66, 1191 – 1308. & Lin PH (2001): Effects on blood pressure of reduced dietary Eurodiet Reports and Proceedings (2001): European diet and public sodium and the dietary approaches to stop hypertension (DASH) health: the continuing challenge. Public Health Nutr. 4(2A), 275 – diet. New Engl. J. Med. 344,3– 10. 292 (Special issue, eds. AG Kafatos & CA Codrington). Sasaki S (2000): Alcohol and its relation to all-cause and cardiovas- European Commission (1997): Report on Osteoporosis in the European cular mortality. Acta Cardiol. 55, 151 – 156. Community. Preventive Actions. Brussels: EC. Stookey JD (2001): Energy density, energy intake and weight status in Ferro-Luzzi A & James WTP (1997): Diet and Health in Scientific and a large free-living sample of Chinese adults: exploring the under- Technological Options Assessment (STOA) Nutrition in Europe. Report lying roles of fat, protein, carbohydrate, fiber and water intakes. to the European Parliament, PE Number 166.481. European Eur. J. Clin. Nutr. 55, 349 – 359. Union, Directorate General for Research. UNICEF=WHO (1999): Regional Consultation Report: Prevention and French Presidency (2000): Health and Nutrition: Elements for a Eur- Control of Iron Deficiency Anaemia in Women and Children. opean Action. Paris, July. Geneva: WHO. Health21 (2000): An Introduction to the Health for all Policy Framework Veer P van’t, Jansen MCJF, Klerk M & Kok FJ (2000): Fruits and for the WHO European Region. Copenhagen: World Health Organi- vegetables in the prevention of cancer and cardiovascular disease. zation, Regional Office for Europe. Publ. Health Nutr. 3, 103 – 107. Honein MA, Paulozzi LJ, Mathews TJ, Erickson JD & Wong LY (2001): Vermeulen EGJ, Stehouwer CDA, Twisk JWR, van den Berg M, de Impact of folic acid fortification of the US food supply on the Jong SC, Mackaay AJC, van Campen CMC, Visser FC, Jacobs occurrence of neural tube defects. J.A.M.A. 285, 2981 – 2986. CACM, Bulterijs EJ & Rauwerda JA (2000): Effect of homocys- Jacques PF, Selhub J, Bostom AG, Wilson PWF & Rosenberg IH (1999): teine-lowering treatment with folic acid plus vitamin B6 on The effect of folic acid fortification on plasma folate and total progression of subclinical atherosclerosis: a randomized, placebo- homocysteine concentrations. New Engl. J. Med. 340, 1449 – 1454. controlled trial. Lancet 355, 517 – 522. Katan MB, Grundy SM & Willett WC (1997): Beyond low fat diets. WCRF=AICR (1997): Food, Nutrition and the Prevention of Cancer: a New Engl. J. Med. 337, 563 – 567. Global Perspective. Washington, DC: American Institute for Cancer Klerk M, Jansen MCJF, van’t Veer P & Kok FJ (1998): Fruits and Research. Vegetables in Chronic Disease Prevention. Wageningen: Division WHO (1998): Obesity — Preventing and Managing the Global Epidemic. of Human Nutrition & Epidemiology, Wageningen Agricultural Report of a WHO Consultation on Obesity. Geneva: WHO. University. WHO (1989): Preventing and Controlling Iron Deficiency Anemia through Krauss RM, Deckelbaum RJ, Ernst N, Fisher E, Howard B, Knopp RH, Primary Health Care: a Guide for Health Administrators and Pro- Kotchen T, Lichtenstein AH, McGill HC, Pearson TA, Prewitt E, gramme Managers. Geneva: WHO. Stone NJ, van Horn L & Weinberg R (1996): Dietary guidelines for healthy American adults. A statement for health professionals from the nutrition committee, American Heart Association. Cir- culation 94, 1795 – 1800.

European Journal of Clinical Nutrition