The Athlete Diet Index: an electronic diet quality assessment tool for elite Australian athletes.

Background: To date, there have been few studies that have reported the nutritional adequacy of athletes’ dietary intake. Diet quality indices provide a cost-effective, efficient measure of nutritional adequacy and dietary patterns compared to population guidelines. However, a valid and reliable diet quality index that considers the unique requirements of athletes does not currently exist. An athlete-specific diet quality assessment tool could offer sports nutrition practitioners an innovative and practical method to rapidly assess the dietary intake of athletes engaged in high performance sport. Aims: The aims of this thesis were to: 1) ascertain the existing research on validity of dietary assessment methods in athlete populations; 2) develop an athlete-specific diet quality tool (Athlete Diet Index; ADI) informed by experienced sports dietitians, and then pilot the ADI with athletes; 3) determine the relative validity and reliability of the ADI; and 4) evaluate the diet quality of elite Australian athletes using the ADI, as well as determine suitability for use in sports nutrition practice. Methods and Results: A systematic search of the literature involving studies published between 1980 until present that compared two or more methods of dietary assessment in athlete populations, yielded 18 included articles. Most studies (n=11) compared self-reported energy intake of athletes to energy expenditure assessed via doubly labelled water; while the remaining studies (n=7) compared new dietary assessment tools to a reference method, such as the food record (FR). The review confirmed there are limited studies involving validation of existing or emerging dietary assessment methods in athlete populations. Next, a mixed-methods approach involving a review of existing diet quality indices, relevant literature, and focus groups with sports dietitians (n=20) was undertaken to develop athlete-relevant items. Draft ADI items were categorised into three sections. The first section evaluated usual dietary intake, special diets, and intolerances, as well as culinary skills; the second section captured current training load, nutrition practices supporting training, and sports supplement use, while the last section included demographic items. Draft items were subsequently confirmed using a modified Delphi technique in a sub-group of sports dietitians (n=9). The final ADI (n=68 items) was then transferred to an electronic platform for pilot testing in athletes (n=15) to assess instrument readability and inform face validity, supporting its relevance and utility for use by athletes. Following on from this, a scoring matrix was developed using a selection of ADI items that assess usual dietary intake (i.e. intake of core and discretionary foods, plus alcohol), as well as markers of healthy dietary habits, particular nutrients of concern for athletes (i.e. calcium and iron), and preferred dietary behaviours. The total ADI score (out of a possible 125) was comprised from three sub-scores (i.e. Core Nutrition, out of a possible 80; Special Nutrients, out of a possible 35; and Dietary Habits, out of a possible 10); with higher overall scores demonstrating better adherence to Australian dietary guidelines and international sports nutrition recommendations. Next, assessment of the relative validity (n=50) and test-retest reliability (n=68) of the ADI was conducted in athletes. Spearman’s rank correlation coefficient demonstrated the ADI to have moderate validity (rs=0.69; p 12 h/week; p