Duke 2008 | ASHE | Equity and Efficiency in Health and Healthcare

Background: Sedentary activities are frequently taxed with being a major contributor to the nation's obesity epidemic, and chief among the sedentary villains is television viewing. Yet the relationship between television viewing and obesity is in fact surprisingly ambiguous, and there is little evidence that television viewing displaces physical activity. Television food advertising may be a more important pathways through which television viewing affects obesity. If so, one would expect that viewing of non-commercial television would have no impact on obesity, while viewing of commercial television would have a large impact. These distinctions are important because parents and their children may be more willing to substitute non-commercial television viewing for commercial television viewing than they are to reduce television and video viewing overall. Objectives: To test whether, among children ages 5-18, (a) the association between television viewing and obesity differs depending on the content viewed; and (b) the association between television viewing and obesity is mediated by physical activity and diet. Methods: The 1997 and 2002 waves of the Child Development Supplement of the Panel Study of Income Dynamics (PSID) were used. The PSID is a nationally representative survey, with over-sampling of African-American populations. Detailed data on time use (including the titles of all of television shows and videos viewed) along with measured body mass index (BMI) were available for 1,914 children ages 2-10 in 1997 and followed up in 2002. Television and video (or DVD) viewing were coded into one of three categories: (a) children's entertainment television shows, (b) children's educational television shows, and (c) any content type in video or DVD format. In a multivariate linear regression, BMI in 2002 was regressed on time spent viewing each of the three content types in 1997. Controls were included for parental income, work status, and education, and for the child's age, gender, race/ethnicity, region of country, and urbanicity. Survey weights were used to adjust for the racial/ethnic non-random sampling, and the standard errors of the estimates were adjusted to reflect the clustering of many of the respondents in families. Results: In the fully adjusted regression, viewing of non-educational children's television was significantly associated with higher BMI 5 years later. Viewing of educational television and of any content on DVD or video were not significantly associated with BMI. These results were significantly but not fully mediated by the inclusion of diet and physical activity variables. Significant effect modification by race/ethnicity was observed. In stratified analyses, these results were significant only for non-Hispanic White children. For African-American children, television viewing of any type was not significantly associated with obesity 5 years later. Conclusions: The effect of television viewing on obesity may be specific to the content viewed. Public health interventions may be able to substantially improve both effectiveness and cost-effectiveness by asking children and their parents to replace viewing of commercial entertainment television with either commercial-free videos and DVDs or with commercial-free educational television.