Premium
Cross‐lagged associations between behaviour problems and obesity in head start preschoolers
Author(s) -
Martoccio Tiffany L.,
Senehi Neda,
BrophyHerb Holly E.,
Miller Alison L.,
Horodynski Mildred A.,
Kaciroti Niko,
Contreras Dawn,
Peterson Karen E.,
Lumeng Julie C.
Publication year - 2020
Publication title -
pediatric obesity
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.226
H-Index - 69
eISSN - 2047-6310
pISSN - 2047-6302
DOI - 10.1111/ijpo.12627
Subject(s) - medicine , obesity , body mass index , percentile , childhood obesity , head start , prospective cohort study , demography , pediatrics , overweight , developmental psychology , psychology , statistics , mathematics , sociology
Summary Background Behaviour problems and obesity are related but research findings have been inconclusive regarding the direction of effects. Objectives This study examined the cross‐lagged associations between behaviour problems, body mass index (BMI) and obesity in preschoolers, and whether sex modified these associations. Methods Repeated measures of teacher‐reported externalizing (EXT) and internalizing behaviour problems (clinically significant T scores were >90th percentile), BMI z‐scores (BMI‐Z) and obesity status (BMI ≥95th for age and sex) were assessed in the fall (T1) and spring (T2) of the school year in Head Start preschoolers ( N = 423). Associations were examined with cross‐lagged modelling. Results Prospective paths from T1 clinically significant EXT to both T2 BMI‐Z ( β = .05) and obesity ( β = .18) were significant. There was no evidence that T1 BMI‐Z or obesity preceded T2 behaviour problems. However, sex‐specific models indicated that T1 BMI‐Z was prospectively associated with higher T2 EXT for boys ( β = .13), but not girls. T1 EXT was predictive of subsequent BMI‐Z ( β = .09) and obesity ( β = .33) at T2 for girls only. Conclusion Findings suggest that behaviour problems, particularly externalizing behaviours, are prospectively related to childhood obesity, and early prevention methods should reflect sex‐specific modifications.