Introduction: The current epidemic of childhood obesity is of great concern as research continues to show that children are becoming more obese. According to the Institute of Medicine, today's children are the most sedentary and inactive generation thus far. The purpose of the study was to examine how a representative sample of kindergarten children in the United States of normal weight differ from kindergarten children who are either overweight or obese
Methods: The data were from the National Early Childhood Longitudinal Study (ECLS) kindergarten cohort and were collected on public and private school Kindergarten children around the nation during the 1998-1999 school year. The dependent variable was body mass index score and the predicting variables were fine motor score, gross motor score, physical activity, weekday television viewing, weekend television viewing, gender, age, race, and socioeconomic status.
The data were analyzed using multinomial logistic regression with the subjects falling into one of three categories, normal weight, overweight or obese. The sampling weights were used in order to make inferences to the national kindergarten population. The Taylor series method was used to take into account the clustered and multistage characteristics of the sampling.
Results: After accounting for gender, age, race and socioeconomic factors in the model, the variables found to be significant to overweight children compared to normal weight children were gross motor skills (p=.001, 95% CI. -.09, -.03) and weekend television time (p=.007, 95% CI .01, .07). The variables found to be significant for obese children compared to normal weight children were gross motor skills (p=.001, 95% CI -.22,-.14), weekday television time (p=.010, 95% CI .02, .14) and weekend television time (p=.003, 95% CI .06, .08).
Discussion: In summary for every one unit decrease in score on the gross motor test children were .94 times more likely to be categorized as overweight and .83 times more likely to be categorized as obese. For every one hour of television watched on the weekend children were 1.04 times likely to be categorized as overweight, 1.05 times for obese. For every one hour of television during the weekday children were 1.08 times more likely to be categorized as obese. The results of this study suggest that development of gross motor skills should be emphasized and that television viewing should be limited in early childhood.