Abstract
OBJECTIVE: To examine how the associations between meal consumption and BMI over 8 years differ by weight status in a sample of adolescents. DESIGN: Longitudinal, population-based study. Breakfast, lunch and dinner consumption and BMI were self-reported. Linear regressions were used to examine how the associations between meal consumption and BMI differed by weight status. SETTING: Adolescents in the Minneapolis/St. Paul metropolitan area. PARTICIPANTS: Adolescents (n 1,471) were surveyed as part of the EAT 2010-2018 in 2009-2010 (M(age) = 14·3 years) and 2017-2018 (M(age) = 22·0 years). RESULTS: The prevalence of regular breakfast, lunch and dinner consumption (≥ 5 times/week) ranged from 45 to 65 %, 75 to 89 % and 76 to 94 %, respectively, depending on weight status category. Among adolescents with a sex- and age-specific BMI < 15th percentile, regular consumptions of breakfast, lunch and dinner during adolescence were positively associated with BMI in emerging adulthood compared with irregular consumption of breakfast, lunch and dinner (<5 times/week) after adjustment for socio-demographic characteristics (β = 5·43, β = 5·39 and β = 6·46, respectively; all P-values <0·01). Among adolescents in the BMI 15-85th and 85-95th percentiles, regular consumptions of breakfast, lunch and dinner were positively associated with BMI but to a lesser extent (P-values <0·01). For participants with a BMI ≥ 95th percentile, regular consumptions of breakfast, lunch and dinner were positively associated with BMI, but the associations were not statistically significant (P-values > 0·05). CONCLUSIONS: The relationship between meal consumption during adolescence and BMI in emerging adulthood differs by adolescent weight status. Future studies should investigate underlying factors related to meal consumption routines and BMI.