Adequate nutrition is one of the pillars of public health. Before developing and implementing effective intervention programmes to improve nutrition at the population level, it is important to know the nutritional situation of the target group.
Assessment of energy and nutrient intake:
The estimation of nutrient intake from food consumption requires reliable data on food composition. These data are also the fundamentals of food-based dietary guidelines for healthy nutrition, containing the necessary information on food sources for different nutrients. Furthermore, food composition tables can provide information on chemical forms of nutrients and the presence and amounts of interacting components, and thus provide information on their bioavailability. For some nutrients such as vitamin A, vitamin E and niacin, the concept of equivalence has been introduced to account for differences in the availability and biological activity of different chemical forms.
Non-nutritive food components:
Although most food composition tables focus on energy, macro- and micronutrients, interest in non-nutritive components is increasing. Considering the beneficial effects of biologically active secondary plant cell compounds such as polyphenols and carotenoids, more data on these are needed. On the other hand, there are a number of naturally occurring or ‘man-made’ non-nutritive substances with negative effects, and to control exposure, the main dietary sources must be known. Another aspect is contaminants, which could have detrimental effects on consumers’ health. Among these are agrochemicals, industrial pollutants reaching the food chain and substances formed during food preparation. A valid risk assessment requires data on exposure, and thus on the contents of contaminants in foods. However, these data are highly variable and may significantly differ even within narrowly confined regions.
Current food composition databases are far from complete:
The fact that composition tables generally do not provide information about the origin of substances found in food can also influence their usability. For example, the German Nutrient Data base does not discriminate between naturally occurring and added sucrose impeding the estimation of added sucrose intake that should be limited.
Points of focus: Considering the increasing number of persons relying on community nutrition and catering, healthy menu lines can improve the consumers’ diets and contribute to nutrient supply. The development and implementation of appropriate guidelines also need food composition databases (FCBs) to compose meals. The ever-increasing number of new food preparations and manufactured products has resulted in a need for procedures for regularly updated data. Moreover, there is a lack of data particularly for essential trace elements such as copper, chromium or molybdenum and also vitamin K, as well as the already mentioned non-nutritive components.
Limited comparability between countries is another issue. Regional differences arise especially from the use of local varieties, different soil quality or meteorological aspects. This variability is further increased with composite meals because of variation in recipes.
Information about food composition is necessary for the assessment of diet quality and the development and application of food-based dietary guidelines, providing a useful tool for the field of public health nutrition. In this regard, more attention should be paid to the preparation, extension and maintenance of FCBs.
Introduction: a brief history of food composition databases
Food composition databases (FCDBs) are not a recent attainment. With the emergence of chemistry, interest in the components of food also arose and the first FCDBs date back to the second half of the nineteenth century. In those times, sufficiency and adequacy of the diet, particularly with regard to macronutrients and minerals, were the main foci of food analysis, as most vitamins were only discovered later. Studies on prisoners to determine their nutritional requirements and search for possible alternative food sources are other examples of this pursuit (Carpenter, 2006). Interestingly, natural variation in food composition was already being taken into account by the Dutch scientist Jacob Moleschott (Moleschott, 1859). However, the influence of food on health was also recognised (Carpenter, 2003) and it is noteworthy that the major British FCDB published in 1940 by McCance and Widdowson (1940), and regularly updated since then, was preceded by earlier work aimed at providing data for the composition of therapeutic diets for diabetics.
The first FCDB for the United States of America, compiled by Atwater and Woods (1896), also encompassed data on processed foods but not for dairy products. Another two very early FCDBs come from Sweden and Denmark, having been published in 1879 and 1888, respectively. Systems from other countries followed later, mostly after World War II. Moreover, the problem of the lack of comparability between the various national reports has been recognised for some time. Accordingly, a first attempt to create a FCDB for international use was made as long ago as 1949 on the Food and Agriculture Organization's initiative (Church, 2006). Today, many countries have their national databases that are regularly updated. However, in light of the increasing globalisation of the food market, regional and international databases are becoming ever more important.
Food composition data in nutritional assessment
Providing the necessary data for the estimation of the intake of nutrients and other food components is still one of the major applications of FCDB. Today, the nutrient content of a given diet is generally calculated with special software programmes that are based on available FCDB. Although it is theoretically possible to analyse directly the nutrient content of the consumed food, this is rarely carried out and is not applicable to large-scale assessments required for representative national nutrition surveys. For these, calculating the composition of the diet using FCDB is the method of choice. However, different FCDBs may differ strongly from each other with regard to the nutrients they include. Thus, their utility depends on the purpose of the analysis. Although energy and total amounts of macronutrients constitute the basic components, data on some compounds are less readily available. This is true for some fatty acids and trace elements, folate, as well as for particularly bioactive compounds and contaminants (Greenfield and Southgate, 2003; Palmer Zimmerman et al., 2008). Furthermore, it is not always possible to know about the provenance of a certain nutrient in composite foods as in the case of sugar, for instance, wherein the distinction between naturally occurring and added mono- and disaccharides is rather difficult in a mixed diet. However, the content of certain critical nutrients such as added sugars, saturated fatty acids or salt is often used as an indicator of the healthiness of a diet (Livingstone and Rennie, 2009).
Factors influencing food composition and nutrient intake
Food composition can be influenced by a number of factors, many of which can barely be controlled. Indeed, seasonal variation can be quite large for the content of micronutrients and bioactive substances in plant foods. Regional disparities also occur. An example is the composition of fatty acids in milk depending on the diet received. Milk from cows kept on high altitude pastures in the Alps is richer in n-3 polyunsaturated fatty acids and conjugated linoleic acid than milk from those in the lowlands or kept indoors. Season also has an effect (Collomb et al., 2008).
When considering a certain food as a source for a given nutrient, bioavailability has an important role but may be difficult to estimate if information about factors reducing or enhancing it and the nutrient's chemical and binding form is not available. In addition, some nutrients can be derived in the body from precursors, such as vitamin A from provitamin A-carotenoids or niacin from tryptophan, so that the total amount provided by a food can be much higher than that for the mere preformed nutrient. To take this into account, equivalents are used on the basis of the metabolism of the different forms. Vitamin A serves as an example with 6 mg of β-carotene and 12 mg of other vitamin A-active carotenoids, yielding 1 mg retinol (FAO/WHO, 2002). However, more recent findings suggested higher factors for carotenoids with regard to its lower bioavailability from fruit and vegetables (West et al., 2002). Thus, there is still some discrepancy about the optimal conversion factors to use for a mixed diet and they can also be altered by physiological conditions. Indeed, the conversion of tryptophan to niacin shows individual variations and is much more efficient during pregnancy (Horwitt et al., 1981). Information about the contents of different metabolites may therefore enable a more accurate estimation of nutrient intake and its contribution to an individual's status.
Food composition data as a basis of nutrition counselling
FCDBs are not only needed for the assessment of nutritional status but they also provide the basis for the formulation of food-based dietary guidelines (FBDGs) to improve a population's diet. FBDGs translate nutrient-based recommendations into a form that is more applicable by consumers. As such, they have an important role when it comes to behavioural modifications that are considered a major means for the prevention of overweight and lifestyle-associated non-communicable diseases. Although recommendations generally refer to food groups as a whole, information on the nutritional quality of individual foods is needed in order to compose a diet designed to meet specific requirements. Accordingly, the concept of nutrient profiling aims to classify foods according to defined criteria, typically fat content and fatty acid pattern, amounts of sugars, fibre, salt and certain micronutrients. This approach is especially required to determine whether nutrition and health claims should appear on food (Lobstein and Davies, 2009; for more details on nutrition and health claims see Buttriss and Benelam, 2010).
Besides indicating the nutrient contents by weight unit or portion on food packages, rating systems allowing an easy and rapid appraisal of the product, thus facilitating consumer choices, have been proposed. Among these, the traffic light system has been intensely debated, being particularly popular in the United Kingdom (Lobstein and Davies, 2009). Labelling is increasingly considered as a tool to educate consumers to make healthier food choices. Simple front-of-package labels appear better suited to this purpose by helping consumers in their decision taking (Grunert and Wills, 2007). Indeed, despite some controversies, use of labels can have a beneficial influence on food choices (Pietinen et al., 2007; Variyam, 2008).
Recently, a number of nutrient profiling tools have been developed to enable a rapid evaluation of foods. As examples, the Nutrimap system (Bio Intelligence Service, Paris, France; Labouze et al., 2007) and the HANCP tool created as part of the European Union-funded project Food-Profit may be mentioned. The latter particularly targets small- and medium-sized enterprises of the food sector to help producers and caterers improve the nutritional quality of their products (Colom, 2009).
With the increasing importance of catering and community nutrition, menus offered in canteens and restaurants have become a focus of public health nutritionists as a means to improve diet patterns. For example, the German Nutrition Society (Deutsche Gesellschaft für Ernährung, DGE) and the Austrian Nutrition Society (Österreichische Gesellschaft für Ernährung, ÖGE) offer a certificate for healthful menu lines prepared according to their guidelines (DGE, 2009) and, again, FCDBs provide the basis for composing the meals and choosing food items.
FCDB in the planning of clinical and therapeutic nutrition
A special application of FCDB to diet composition is the planning of nutrition as part of a therapy. Well-known examples of diseases that require avoidance or accurate monitoring of certain nutrients include diabetes mellitus and dyslipidaemia. Although data on total carbohydrate and fat contents and major fatty acid classes are widely available, there is less information on other components with pathological relevance such as fructose, lactose or various amino acids, gluten or antigens. The importance of reliable data on food composition for therapeutic diets was recognised long ago and, for McCance, it was the main reason for compiling his database which, in its updated form, is still in use in the United Kingdom (Church, 2006). For diabetics, exchange lists have been commonly used classifying foods according to their carbohydrate, energy and fat content. Foods can be substituted within groups, facilitating diet composition and adjustment of medication for patients. Newer approaches have focussed on glycaemic index and carbohydrate counting (Gillespie et al., 1998).
The need for data on other food components
In recent times, new functions beyond mere supply of energy and nutrients have been assigned to nutrition, as food components with biological activities have been discovered. They have emerged as substantial contributors to the health-promoting effects of fruits, vegetables, nuts, seeds and wholegrain cereals, and encouraging the consumption of these foods is therefore an integral part of nutrition campaigns. The knowledge about optimal food sources requires data about the distribution and amounts of bioactive compounds in foods (Ritchie et al., 2006; for more details on the EuroFir eBasis, providing information on bioactive compounds, see Kiely et al., 2010). However, so far, most FCDBs do not encompass these data or only for some foods (Palmer Zimmerman et al., 2008). In addition, their concentration in foods is subject to a high variability depending on plant variety, cultivation and geographical conditions (Hecke et al., 2006; Raffo et al., 2006). On the other hand, data on potentially harmful food components such as pesticides, contaminants or by-products of food processing, such as trans-fatty acids or acrylamide, are also important when making nutritional recommendations, providing the basis for the estimation of exposure and risk assessment (Jakszyn et al., 2004). They also provide the basis for recipe reformulation and modifications of production methods aimed at reducing the amount present in food (Birlouez-Aragon et al., 2010). In addition, completing missing values for specific compounds is important to reduce the risk of underestimating exposure that can arise from designating a missing value as null (Merchant and Dehghan, 2006).
Information on food composition is of great importance for scientists and practitioners working in the fields of nutrition and public health. The most apparent role is to provide the basis for dietary assessment and the formulation of healthier diets. Ready meals and food served in canteens are increasingly included in this approach considering their contribution to daily nutrition. For the general population, food-based dietary guidelines are considered the best means to convey more understandable recommendations for food choice. Nutrition and health claims have to be supported by sound scientific evidence, including data on the food's nutrient content. The latter is even more important when one or a group of specific nutrients have to be avoided or controlled for a pathological reason. Therapeutic nutrition may require additional data that are so far not widely available. This is also the case for bioactive compounds mainly found in plant foods that, in light of their varied health effects, have recently become the focus of scientific interest. Finally, data on contaminants and other potentially harmful compounds are also of public relevance to enable risk assessment. Against this background, there exists a strong need to harmonise existing data and collect new data on food composition.
Atwater WO, Woods CD (1896). The chemical composition of American food materials. US Department of Agriculture. Office of experiment stations. Washington Bull 28, 1–47.
Birlouez-Aragon I, Morales F, Fogliano V, Pain J-P (2010). The health and technological implications of a better control of neoformed contaminants by the food industry. Pathol Biol 58, 232–238.
Buttriss JL, Benelam B (2010). Nutrition and health claims: the role of food composition data. Eur J Clin Nutr 64 (Suppl 3), S8–S13.
Carpenter KJ (2003). A Short History of Nutritional Science: Part 1 (1785–1885). J Nutr 133, 638–645.
Carpenter KJ (2006). Nutritional studies in Victorian prisons. J Nutr 136, 1–8.
Church SM (2006). The history of food composition databases. Nutr Bull 31, 15–20.
Collomb M, Bisig W, Bütikofer U, Sieber R, Bregy M, Etter L (2008). Seasonal variation in the fatty acid composition of milk supplied to dairies in the mountain regions of Switzerland. Dairy Sci Technol 88, 631–647.
Colom A (2009). HANCP—a new tool for small and mediumsized companies to reformulate processed foods and meals (FOOD PRO-FIT). Abstract of a symposium presented at the 19th International Congress of Nutrition 2009, Bangkok. Ann Nutr Metab 55 (Suppl 1), 65.
Deutsche Gesellschaft für Ernährung (DGE) (2009) Qualitätsstandards für die Betriebsverpflegung [Quality standards for canteen meals] 2nd edn, leaflet, Bonn. Available (in German) at: http://www.jobundfit.de/index.php?id=32&L=0&C=0&G=0.
FAO/WHO (Food and Agriculture Organization of the United Nations/World Health Organization) (2002) Human Vitamin and Mineral Requirements. Report of a Joint FAO/WHO Expert Consultation, Bangkok, Thailand. FAO, Rome, Italy.
Gillespie SJ, Kulkarni KD, Daly AE (1998). Using carbohydrate counting in diabetes clinical practice. J Am Diet Assoc 98, 897–905.
Greenfield H, Southgate DAT (2003). Food Composition Data: Production, Management and Use, 2nd edn. FAO: Rome.
Grunert KG, Wills J (2007). A review of European research on consumer response to nutrition information on food labels. J Publ Health 15, 385–399.
Hecke K, Herbinger K, Veberic K, Trobec M, Toplak H, Stamper F et al. (2006). Sugar-, acid- and phenol contents in apple cultivars from organic and integrated fruit cultivation. Eur J Clin Nutr 60, 1136–1140.
Horwitt MK, Harper AE, Henderson LM (1981). Niacin-tryptophan relationships for evaluating niacin equivalents. Am J Clin Nutr 34, 423–427.
Jakszyn P, Agudo A, Ibáñez R, García-Closas R, Pera G, Amiano P et al. (2004). Development of a food database of nitrosamines, heterocyclic amines, and polycyclic aromatic hydrocarbons. J Nutr 134, 2011–2014.
Kiely M, Black LJ, Plumb J, Kroon PA, Hollman PC, Larsen JC et al., EuroFIR consortium (2010). EuroFIR eBASIS: application for health claims submissions and evaluations. Eur J Clin Nutr 64 (Suppl 3), S101–S107.
Labouze E, Goffi C, Moulay L, Azaïs-Braesco V (2007). A multipurpose tool to evaluate the nutritional quality of individual foods: Nutrimap. Public Health Nutr 10, 690–700.
Livingstone MB, Rennie KL (2009). Added sugars and micronutrient dilution. Obes Rev 10 (Suppl 1), 34–40.
Lobstein T, Davies S (2009). Defining and labelling ‘healthy’ and ‘unhealthy’ food. Public Health Nutr 12, 331–340.
McCance RA, Widdowson EM (1940). The chemical composition of foods. Medical Research Council Special Report Series No. 235. His Majesty's Stationery Office: London.
Merchant AT, Dehghan M (2006). Food composition database development for between country comparisons. J Nutr 5, 2.
Moleschott J (1859). Die Physiologie der Nahrungsmittel. Ein Handbuch der Diätetik. [The physiology of foods. A manual of dietetics] 2nd revised edn, Ferber′sche Universitätsbuchhandlung (Emil Roth): Giessen.
Palmer Zimmerman T, Stumbo P, Chenard C, Braithwaite E, Selley B, Databank Directory Committee of the National Nutrient Databank Conference USA (2008). 2008 International Nutrient Databank Directory. Available at: http://www.healthcare.uiowa.edu/gcrc/nndc/survey.html.
Pietinen P, Valsta LM, Hirvonen T, Sinkko H (2007). Labelling the salt content in foods: a useful tool in reducing sodium intake in Finland. Public Health Nutr 11, 335–340.
Raffo A, La Malfa G, Fogliano V, Maiani G, Quaglia G (2006). Seasonal variations in antioxidant components of cherry tomatoes (Lycopersicon esculentum cv. Naomi F1). J Food Compos Anal 19, 11–19.
Ritchie MR., Cummings JH, Morton MS, Steel CM, Bolton-Smith C, Riches AC (2006). A newly constructed and validated isoflavone database for the assessment of total genistein and daidzein intake. Br J Nutr 95, 204–213.
Variyam JN (2008). Do nutrition labels improve dietary outcomes? Health Econ 17, 695–708.
West CE, Eilander A, van Lieshout M (2002). Consequences of revised estimates of carotenoid bioefficacy for dietary control of vitamin A deficiency in developing countries. J Nutr 132, 2920S–2926S.
The authors declare no conflict of interest.
About this article
Cite this article
Elmadfa, I., Meyer, A. Importance of food composition data to nutrition and public health. Eur J Clin Nutr 64, S4–S7 (2010). https://doi.org/10.1038/ejcn.2010.202
- food composition
- public health
- non-nutritive compounds
Iron, iodine and vitamin D deficiencies during pregnancy: epidemiology, risk factors and developmental impacts
Proceedings of the Nutrition Society (2021)
Development of a nutrient dataset based on a standardized approach for a nutrition survey conducted in the United Arab Emirates
Journal of Food Composition and Analysis (2021)
Revista Paulista de Pediatria (2021)
Missing association between nutrient concentrations in leaves and edible parts of food crops – A neglected food security issue
Food Chemistry (2021)