Design and validation of web and computer-based food frequency questionnaire: a scoping review

• Main Authors: Shahril, M.R., Laila Ruwaida, M.H.
• Format: Conference or Workshop Item
• Language: English
• Published: 2019
• Subjects: Q Science (General); QA75 Electronic computers. Computer science
• Online Access: http://eprints.unisza.edu.my/2010/1/FH03-FSK-19-34666.pdf; http://eprints.unisza.edu.my/2010/

Web and computer-based food frequency questionnaire (FFQ) provide a lot of advantages including saving cost, reducing participants burden and improving response quality. However, designing and validating a web or computer-based FFQ for a study can be a challenge. Therefore, this study aims to conduct a scoping review on the design and validation of web and computer-based FFQ among adult subjects. An electronic search was conducted on the PubMed database to identify studies published from 2009 to 2018 that included a web or computer-based FFQ for assessment of dietary intake among adult population. A pool of 27 articles were examined in-depth to assess the development, comparison or validation of a web or computer-based FFQ. Most of the studies (55%) were done in European countries. The number of food/drink items listed in the FFQ ranges from 12 to 270 items and portion size of food was estimated using images, ranging from three to seven images per food item or a standard portion size using household measurement. The reference period was usually over a period of one month up to one year. Majority of the study (78%) focus on intake of all food groups except for several studies that focuses on specific nutrient or food group namely sodium, fatty acids, vitamin D, calcium, fruits and vegetables, and beverage intake. There were only three studies that recorded supplement intake during web or computer-based dietary assessment. Out of 27 studies, a total of 22 carried out validation studies using a variety of reference method such as 24hours diet recall (24%), food diary/ records (27%), weighed food record (27%), paper version of FFQ (22%) and biomarker (1%). As a conclusion, a variety of web and computer-based FFQ were used in research worldwide. Strategies to improve the application of current evidence on best practices in designing and validating a web or computer-based FFQ have the potential to improve nutritional epidemiology studies.