Association of Television Viewing Behaviors with Energy Intake and Dietary Quality in US Adolescents (NEXT Generation Health Study)

Background/objective: Television (TV) viewing has shown both acute and sustained influences on eating behaviors, though previous studies have examined limited aspects of these exposures and outcomes. The objective of this study was to examine cross-sectional associations of multiple indicators of dietary quality and TV viewing behaviors in a national sample of US adolescents.

Methods: Subjects completed self-administered surveys and the Automated Self-Administered 24-hour dietary recall (n=479, mean±SD age=16.2±0.52, 47% non-Hispanic White, 55% female, 48% overweight/obese). Dietary outcomes included mean daily energy intake (kcal, EI), EI from total fat , EI from saturated fat, EI from added sugars, nutrient density (Nutrient Rich Foods 9.3 score, NRF), and whole plant food intake (WPF, daily cup or oz equivalents per 1000 kcal). TV behaviors included total TV/DVD viewing time (hours/day), frequency of TV viewing during meals (TV meals, days/week), and during snacks (TV snacks, days/week). Associations were examined using multiple linear regression models using log-transformed dietary outcomes, and adjusted for weight status, race/ethnicity, sex, family affluence, and parental education, and total EI where appropriate. Interactions of TV behaviors with weight status (normal weight v. overweight/obese) were evaluated using a multiplicative interaction term and stratified analyses where applicable.

Results: Subjects reported watching TV/DVDs 2.3 ± 1.8 hours/day. TV meals and TV snacks occurred at least 1 day/week for 70% and 71% of the sample, respectively. Each hour of TV/DVD viewing was associated with a 3.6% increase in EI (p=0.001), a 5% decrease in WPF (p=0.003), a 2% increase in total fat EI (p=0.02), a 3% increase in saturated fat EI (p=0.006), and a 5% increase in added sugar EI (p=0.01). Each day/week increase in TV meals was associated with a 4% decrease in NRF (p=0.02), while each day/week increase in TV snacks was associated with a 4% increase in added sugar EI (0.005).  Weight status modified the relationships of TV/DVD time (interaction term p=0.03) and TV meals (p=0.006) with EI; TV/DVD time was positively associated with EI in overweight/obese subjects only, whereas TV meal frequency was negatively associated with EI only in normal weight subjects.

Conclusions: TV/DVD viewing time was adversely associated with multiple indicators of dietary quality. Associations were less consistent for TV viewing specifically during meals and snacks, suggesting total TV exposure may be more relevant for eating behaviors than acute exposure. Little evidence of effect modification suggests that weight status has minimal bearing on associations of TV behaviors with dietary outcomes.