Abstract
Background: Accurate assessment of an athlete’s diet is important to ensure optimal nutrition for exercise performance and health. Limitations of conventional dietary assessment methods have generated interest in electronic technologies, such as Smartphone diet applications (apps), which may potentially improve the accuracy and efficiency of dietary assessment. Diet apps have typically been designed to facilitate weight loss through self-monitoring of dietary intake. More recently their potential as a dietary assessment tool has been investigated in the general population. However, their validity and feasibility of use in athletes is yet to be established.
Objective: To determine the relative validity and feasibility of the iPhone application Easy Diet Diary (EDD), both with and without sports dietetic consultation, for the assessment of energy and macronutrient intakes in high energy consuming male athletes.
Design: Fifty-three high energy consuming male athletes (endurance athletes and rugby players) recorded their dietary intake for four days using EDD and two weeks later using a weighed diet record (WDR). A follow-up interview was conducted after each method to clarify (“clean”) records with participants. Mean energy and macronutrient intakes estimated by EDD, before (EDD Uncleaned), and after cleaning (EDD Cleaned), and the cleaned WDR were compared. Paired t tests, intraclass correlation coefficients (ICCs) and Bland Altman plots were used to assess agreement between methods.
Results: Mean differences (with 95% confidence intervals) between EDD Uncleaned and the WDR were not significant for protein (4 g/d; -9, 17 g/d) and fat (-9 g/d; -21, 3 g/d), but were significant for energy (961 kJ/d; -1912, -12 kJ/d) and CHO (38 g/d; -68, -8 g/d). There were no significant differences between EDD Cleaned and the WDR for energy (-378 kJ/d; -1129, 373 kJ/d), fat (-3 g/d; -14, 9 g/d), protein (4 g/d; -7, 15 g/d), and CHO (-18 g/d; -42, 5 g/d). At the individual level, ICCs indicated agreement between EDD Uncleaned and the WDR was poor for fat (0.30; 0.04, 0.52), acceptable for energy (0.48; 0.25, 0.66) and CHO (0.65; 0.45, 0.77), and strong for protein (0.74; 0.59, 0.84). For EDD Cleaned and the WDR, ICCs indicated agreement between methods was acceptable for energy (0.62; 0.42, 0.76), strong for protein (0.82; 0.71, 0.89) and CHO (0.76; 0.62, 0.86), and poor for fat (0.31; 0.05, 0.54). The limits of agreement were wide for all nutrients between EDD Uncleaned and the WDR and for fat only between EDD Cleaned and the WDR. Agreement between methods remained relatively consistent across different levels of nutrient intake. The majority (96%) of participants preferred EDD to the WDR and 98% found it less time-consuming.
Conclusion: The EDD App, used in consultation with a dietitian, is a valid and feasible tool for assessing energy and macronutrient intakes in individuals or groups of high energy consuming male athletes, with the exception of fat at the individual level. However, EDD on its own (without dietetic consultation) was less comparable with the WDR for assessing energy and macronutrient intakes.