To Örebro University

Introduction: The Oxford WebQ is a web-based 24-hour dietary assessment tool used in UK Biobank and other large-scale prospective studies to assess dietary intake. The food composition table (FCT) used to calculate nutrient intakes, the McCance and Widdowson’s  FCT 6th edition (2002), has recently been replaced with the UK Nutrient Databank (2013), as this FCT is, on average, closer in time to when participants completed the questionnaire. Moreover, portion sizes, personalisation of fats used in cooking, food group disaggregation and the underlying code for nutrient calculation were revised and updated. In this study we aim to compare the two versions of obtained nutrient intakes in UK Biobank participants. 

Methods: A subsample of UK Biobank participants recruited towards the end of the recruitment period and/or those who provided email addresses, completed the Oxford WebQ questionnaire up to 5 times (2009/2012). After excluding participants with extreme energy intakes (men: >20,000 kJ/d, women: 18,000 kJ/d), a total of 210,109 participants, who completed the Oxford WebQ questionnaire at least once, were included. Means and standard deviations of macronutrient and fibre intakes were averaged for all completed WebQ questionnaires for each participant and Spearman correlations and paired t-tests were used to compare the mean nutrient intakes between the two versions.

Results: All nutrient intake differences were significantly different from zero after the WebQ update (P<0.001). Compared to the previous version , mean nutrient intakes in this updated version were slightly lower for energy (-144 kJ/d), protein (-1.2 g/d), total fat (-4.5 g/d) and saturated fat (-2.7 g/d), but slightly higher for carbohydrates (+1.8 g/d), total sugars (+5.3 g/d) and fibre (+1.4 g/d). High correlations were found between nutrients calculated using the two versions: energy (r=0.96), protein (r=0.97), total fat (r=0.95), carbohydrates (r=0.95), saturated fat (r=0.91), total sugars (r=0.96), and fibre (r=0.94).

Discussion: Absolute differences in nutrient intakes between the two versions were observed, although the ranking of individuals was minimally affected. Future work will assess the heterogeneous differences among individuals caused by the changes made to the WebQ, and this updated version will be available to all studies using the Oxford-WebQ.