Quality and Standards

When working with food composition data, it is important to be aware of the quality of the data available.

A high-quality food composition database should display the quality of the data in a manner that every user can easily judge how confident one can be in the data. To achieve this goal the quality of every single value stored in the database must be determined.

The following list shows the most important quality criteria for food composition databases.

Selected quality criteria for food composition data
The quality of food data can only be determined if a full documentation on the origin of the value is available. Therefore, the following information about every single data point must be known (according to Greenfield and Southgate. Food composition data. FAO, Rome, 2003):
•    Criteria for food selection
•    Number of analysed food items
•    Sample preparation
•    Type of analysis method
•    Quality control of analysis method

Further quality criteria are related to the food item as a whole (e.g. unambiguous food identification, complete nutrient data) and on the management of the food data (e.g. use of standard templates for data delivery, double entry of data, automatic error message when sum of nutrients is different from 100%, automatic error message when the data of a nutrient is higher or lower than theoretically possible).

Information about the quality of data is also important when comparing data from different countries. This allows those using the data to make a judgement on whether differences in food composition values are genuine or due to difference on quality standards.

EuroFIR AISBL has a well-developed Quality Assessment System in place, which can be used by Compiler Member Organisations and allow the detailed and accurate quality assessment of the food composition data in the national databases. The system is being implemented in the databases step by step and will lead to the availability of a quality index score for each data point in the database.