*2.3. Dietary Measures*

Diet was assessed using repeated 24-h dietary recalls on non-consecutive days during two in-person visits as part of the BrainChild Study. In a subset of 35 participants, we also obtained 3-day dietary records on one of the two in-person visits, and dietary intake was estimated using a total of four days of dietary assessments for these participants. The mean interval between the two in-person visits was 34 ± 57 days. We used the multi-pass method for dietary recall, in which a trained staff member asked the participant to recall what food and beverages they had consumed over a 24-h time period with the input of both the child and the child's parent. The trained staff member then went through three other "passes" to complete quantity of food/beverages consumed as well as to include missing or forgotten food/beverages. Use of the multi-pass 24-h dietary recall method is a valid method to assess energy intake in children [36]. Once the dietary recalls were collected, the recalls were analyzed using the Nutritional Data System for Research software v.2018 developed by the Nutrition Coordinating Center (NCC), University of Minnesota, Minneapolis, MS [37]. The variables used were percent calories from fructose, percent calories from glucose, and percent calories from added sugar by available carbohydrate. Data from dietary recalls were manually checked for quality. To determine outliers, we performed linear regression analysis, using body weight to predict total energy intake. Residuals were standardized and examined for any values that were >3 or <−3 standard deviations from the mean. Records containing data that exceeded these values were not included in the analysis. Using this method, 267 dietary recalls were included in the analysis, and one dietary recall was excluded.
