“There are two schools of thought about food tables. One tends to regard the figures in them as having the accuracy of atomic weight determinations; the other dismisses them as valueless on the ground that a foodstuff may be so modified by the soil, the season or its rate of growth that no figure can be a reliable guide to its composition. The truth, of course, lies somewhere between these two points of view.”  (McCance and Widdowson, 1943.) 

What are food composition data?

Food composition data are usually presented in the form of a database or table and contain information on the composition of foods. They provide information on macronutrients, i.e. protein, carbohydrates and fat, and micronutrients, i.e. vitamins and minerals. Some non-nutritive components, such as carotenoids and polyphenols, are also sometimes included (1). 

This information is used for multiple purposes, namely dietary intake assessments for patient care, education and research, product development and recipe analysis and management (2). 

The use of food composition data has increased over the years and, consequently, so has awareness around the limitations of the data. It is important to note that incorrect use of food composition information can lead to inappropriate food policies, misleading food labelling, inaccurate health claims and, ultimately, inadequate and unsafe consumer choices, e.g. where allergens are concerned (3).

How are food databases created?

In the early days, food composition tables were created from values obtained from laboratory analysis. This system then moved on to compiling data produced by other trustworthy laboratories — a practice first introduced when vitamin and amino acid values drawn from literature were included in the third edition of McCance and Widdowson. (4) 

Nowadays, there are a few other methods used to compile information onto a food composition table (4,5), which include:

  1. Chemical analysis of food samples. This is most often the preferred method. Food samples are chosen according to a defined sampling plan and are collected, transported to a laboratory and stored in an appropriate manner. If required, further preparation and/or cooking happens prior to analysing the food, using the right equipment and methods.
  2. Calculation of values using yield and nutrient retention factors. Due to limited resources, it is often not possible to perform a direct analysis of every food or nutrient, so compilers use different approaches to determine these values. For example, the values for a raw food or dish can be combined with information on likely weight and nutrient changes due to cooking in order to estimate values for the food once cooked. This also accounts for weight change (or yield) and changes in nutrient content with cooking.
  3. ‘Borrowing’ values from another food composition database. Sometimes, compilers will use nutrient values originally compiled by another organisation. These need to be checked to ensure they are compatible with their own database.
  4. Adopting values from other sources. It’s also common to utilise food composition databases from other countries or manufacturers’ data (e.g. from food labels). Before incorporating the information, compilers will evaluate it in terms of quality and applicability to their own database. 

Development of food composition data

Since the early 19th century, food composition data have evolved significantly. For a long time, these would be circulated in the form of printed copies. Although printed versions are still available in most countries, there has been a shift to electronic databases in recent years, as they are able to store more information and allow easier access and manipulation of data (5).

What to consider when using food composition databases

Despite their wide use in various fields, food composition data have several limitations, acknowledged by the scientific community but often not fully understood by users (3). Foods are biological materials and therefore have natural variability in their composition (4). These differences may be due to a number of factors (4,6), such as:

  • Differences in the animal or plant species assessed
  • Environmental factors, such as soil and climate
  • Variation in agricultural practices or the storage, processing and preparation of foods 

Other limitations of food composition data include the following:

  • The degree of variation in nutrient composition varies for different nutrients, and micronutrient values vary more widely than macronutrient values.
  • Partial or limited coverage of food items, with an increasing number of processed foods being consumed around the world nowadays.
  • Information on certain components may be limited, such as bioactive compounds that have sparked interest in recent in vitro and animal studies in recent years.
  • Incompatibilities exist between databases in different countries.
  • Results presented as averages are not representative of one single food item.
  • Users are not always aware of the difference in nutrient values between raw and cooked foods, and may wrongly use the values for raw foods in place of those for cooked ones. 

In addition, despite best efforts when it comes to updates, a certain degree of data ageing is inevitable due to limited resources and the increased complexity of the food supply (3,4,6). 

With all of this in mind, it’s important to consider these factors when making decisions about nutrient intakes that are derived from food composition data. Indeed, there are certain situations — such as creating a food label — in which using generic information from food databases may not be suitable, since allergens are brand specific, with variation in ingredients seen between brands. 

What is needed for the future?

Despite major efforts and progress during the past two decades on matching food descriptions, nutrient terminology, analytical methods, calculation and compilation methods, values from existing food composition tables and databases are not yet readily comparable across every country (4,6). 

Some current needs and directions for the future on food composition data have been highlighted (3,4,5,6):

  • Continuing work on international harmonisation of food nomenclature, description and on food composition programmes.
  • Identifying and providing new information on missing data for nutrients and bioactive compounds for all food groups, including traditional and culturally-specific foods.
  • Further research studies on sampling.
  • Ongoing improvement of analytical methods.
  • Training and education for scientists and users to ensure the appropriate use of information.

A lot of work continues to be done in this area, with several dedicated networks established around the world to improve this (7). You can read more about this on INFOODS and EuroFIR.

What can a good food data management software package do for you?

A significant proportion of users (including researchers, manufacturers, retailers, dietitians and nutritionists) will use the food composition database(s) integrated into the software package they have purchased (4). Software producers often incorporate additional foods or components into databases, or may select certain types of nutrient data. Therefore, users should be trained to evaluate software packages before purchase, especially if buying for a large number of end users (4). 

The range of functionalities required in today’s dietary analysis is huge. Amongst others, these include: creating client records; calculating the nutrient breakdown of recipes, meals, diet logs and menus; reporting on dietary intake and comparing against adequate dietary requirements; storing calculated records or exporting them for further statistical analysis; managing ingredients and allergens; costing products, meals and diets; printing labels for meals and clients; developing research, therapeutic or hospital diets; purchasing food according to different costs; and sharing nutritional information of menus with customers online (4). 

The Nutritics database is comprised of the official national databases for each country, augmented with additional information for improved accuracy. Our nutritionists have manually enhanced these global databases with photos, suggested portion sizes and detailed micro-nutrition for up to 258 nutrients. You won’t find data this complete anywhere else! 

Each database is available in the local language(s) and in English. In addition, we’ve created a large database of universal Arabic foods, and we maintain toggleable databases from brands, food suppliers, retail chains and clinical feeds, giving you complete control over where you source your data. 

References

  1. Marcus J. Culinary Nutrition. Waltham, MA: Elsevier/Academic Press; 2014.
  2. Pennington J. Applications of food composition data: Data sources and considerations for use. Journal of Food Composition and Analysis. 2008;21:S3-S12.
  3. Delgado A, Issaoui M, Vieira M, Saraiva de Carvalho I, Fardet A. Food Composition Databases: Does It Matter to Human Health?. Nutrients. 2021;13(8):2816.
  4. Food and Agriculture Organization of the United Nations. Food Composition Data – Production, Management and Use [Internet]. Rome; 2003. Available from: https://www.fao.org/3/y4705e/y4705e.pdf
  5. Food Data » EuroFIR [Internet]. Eurofir.org. 2021 [cited 24 November 2021]. Available from: https://www.eurofir.org/food-information/
  6. Williamson C. Synthesis report No 2: The Different Uses of Food Composition Databases. 2006.
  7. Church S. The history of food composition databases. Nutrition Bulletin. 2006;31(1):15-20.