Television watching and frequency of family meals are predictive of overweight onset and persistence in a national sample of school-aged children

J Am Diet Assoc. 2007 Jan;107(1):53-61. doi: 10.1016/j.jada.2006.10.010.


Objective: To identify eating and activity factors associated with school-aged children's onset of overweight and persistent overweight.

Design: Data were gathered at four time points between kindergarten entry and spring of third grade. Children were directly weighed and measured and categorized as not overweight (<95th percentile of body mass index) or overweight (> or =95th percentile body mass index); parents were interviewed by telephone or in person.

Subjects/setting: Subjects were participants in the Early Childhood Longitudinal Study-Kindergarten Cohort, a nationally representative sample of children who entered kindergarten during 1998-1999. Children who weighed <2,000 g at birth, received therapeutic services before kindergarten, skipped or repeated a grade, or without complete height and weight data were excluded, resulting in 8,459 participants. Children with intermittent overweight were not examined (n=459); analyses addressed 8,000 children.

Main outcome measures: Three mutually exclusive groups of children were identified: never overweight, overweight onset, and persistent overweight.

Statistical analyses performed: Multilevel, multivariate logistic regression analyses estimated the effects of eating and activity factors on the odds of overweight onset and persistent overweight above child sex, race, and family socioeconomic status.

Results: Children who watched more television (odds ratio [OR] 1.02) and ate fewer family meals (OR 1.08) were more likely to be overweight for the first time at spring semester of third grade. Children who watched more television (OR 1.03), ate fewer family meals (OR 1.08), and lived in neighborhoods perceived by parents as less safe for outdoor play (OR 1.32) were more likely to be persistently overweight. Child aerobic exercise and opportunities for activity were not associated with a greater likelihood of weight problems.

Conclusions: This study supports theories regarding the contributions of television watching, family meals, and neighborhood safety to childhood weight status. When working with families to prevent and treat childhood weight problems, food and nutrition professionals should attend to children's time spent with screen media, the frequency of family mealtimes, and parents' perceptions of neighborhood safety for children's outdoor play.

Publication types

  • Research Support, U.S. Gov't, Non-P.H.S.

MeSH terms

  • Body Mass Index
  • Child
  • Child Development
  • Child, Preschool
  • Cohort Studies
  • Confidence Intervals
  • Eating*
  • Exercise / physiology*
  • Female
  • Humans
  • Logistic Models
  • Longitudinal Studies
  • Male
  • Multivariate Analysis
  • Obesity / epidemiology
  • Obesity / etiology*
  • Odds Ratio
  • Overweight
  • Play and Playthings*
  • Predictive Value of Tests
  • Risk Assessment
  • Social Class
  • Television*
  • Time Factors
  • United States / epidemiology
  • Weight Gain / physiology*