Concordance and predictors of concordance of children's dietary intake as reported via ecological momentary assessment and 24 h recall

Objective To provide preliminary evidence in support of using ecological momentary assessment (EMA), a real-Time data capture method involving repeated assessments, to measure dietary intake in children by examining the concordance of children's dietary reports through EMA and 24 h recall. Design Children completed eight days of EMA surveys, reporting on recent dietary intake of four pre-specified food categories ('Fruits or Vegetables', 'Chips or Fries', 'Pastries or Sweets', 'Soda or Energy Drinks'), and completed two 24 h recalls during the same period. Concordance of children's reports of intake during matched two-hour time windows from EMA and 24 h dietary recall was assessed using cross-Tabulation. Multilevel logistic regression examined potential person-level (i.e. sex, age, ethnicity and BMI category) predictors of concordance. Setting Children in Los Angeles County, USA, enrolled in the Mothers' and Their Children's Health (MATCH) study. Subjects One hundred and forty-four 144 children (53 % female; mean age 9·6 (sd 0·9) years; 34·0 % overweight/obese). Results Two-hour concordance varied by food category, ranging from 64·9 % for 'Fruits/Vegetables' to 89·9 % for 'Soda/Energy Drinks'. In multilevel models, overweight/obese (v. lean) was associated with greater odds (OR; 95 % CI) of concordant reporting for 'Soda/Energy Drinks' (2·01; 1·06, 4·04) and 'Pastries/Sweets' (1·61; 1·03, 2·52). Odds of concordant reporting were higher for Hispanic (v. non-Hispanic) children for 'Pastries/Sweets' (1·55; 1·02, 2·36) and for girls (v. boys) for 'Fruits/Vegetables' (1·36; 1·01, 1·83). Conclusions Concordance differed by food category as well as by person-level characteristics. Future research should continue to explore use of EMA to facilitate dietary assessment in children.