The reproducibility of a diet using three different dietary standardisation techniques in athletes

Participants are often asked to replicate their diet before each trial to control dietary intake. However, little is known about the reproducibility of the diet using different approaches. The aim was to assess the reproducibility of a diet when a controlled diet (Cdiet), food record (Frecord) and 24-h dietary recall (Drecall) were used as dietary standardisation techniques.

Thirty athletes completed six visits to the laboratory. On the first occasion, a Drecall was performed and subjects were asked to replicate exactly the same diet on the day before visit 2, when another Drecall was performed. The day before the third visit, subjects completed a Frecord, which was presented at visit 3 and assessed using Drecall to ensure comparability between methods. Subjects were asked to replicate this Frecord before visit 4, which was assessed using Drecall. Finally, subjects were provided with a Cdiet of known composition, which they consumed for 24 h before visits 5 and 6. For each method, the difference in energy and macronutrient intakes between both occasions was measured.

Despite finding no differences in mean energy and macronutrient intakes between visits for any technique, important within-subject differences were apparent. The range of percentage coefficient of variation for all variables was between 2.7 and 5.8% for Cdiet, 10.1 and 18.6% for Frecord and 7.1 and 11.7% for Drecall.

This study has shown that Cdiet is the best approach to standardise dietary intake, especially when the expected effect of an intervention is small and an enhanced reliability is required.