OBJECTIVE: Little is known about the impact of viewing depictions of alcohol in entertainment media on adolescent drinking behavior. Our aims were to assess drinking in a sample of popular contemporary movies and to examine the association of movie alcohol exposure with early-onset drinking in an adolescent sample. METHOD: We conducted a school-based cross-sectional survey (N=4655) with longitudinal follow-up of never-drinkers (N=2406) involving adolescents ages 10-14 years and recruited from 15 New Hampshire and Vermont schools. Screen depictions of alcohol use were timed for each of 601 popular contemporary movies. Each adolescent was asked if he/she had seen a unique list of 50 movie titles, randomly selected from the larger pool. Movie alcohol use was summed for movies the adolescent had seen, adjusted to reflect exposure to the larger pool and modeled as a continuous variable. RESULTS: Ninety-two percent of the movies in the sample depicted drinking; median screen time for movie alcohol use was 2.5 minutes (interquartile range [IQR]: 0.9-5.0 minutes). Median exposure to movie alcohol use from the 601 movies was 8.6 hours (IQR: 4.6-13.5 hours). Overall 23.1% of the cross-sectional sample had tried alcohol, and 14.8% of initial nondrinkers had tried alcohol at the follow-up assessment. We found statistical evidence to support a curvilinear association between higher exposure to movie alcohol use and increased risk of prevalent and incident alcohol use, with a statistically significant linear and quadratic effect, and suggesting a higher dose-effect relationship at lower movie alcohol exposure levels compared to higher levels. The linear and the quadratic associations remained strong and significant in cross-sectional and prospective models after controlling for sociodemographics (grade in school, school, gender, parent education), personality characteristics of the adolescent (sensation seeking, rebelliousness, self-esteem), school performance, parenting style, and smoking experimentation, suggesting that exposure to movie alcohol use is an independent risk factor for early-onset alcohol use. CONCLUSIONS: This study provides the first evidence for an association between exposure to movie alcohol use and early-onset teen drinking. The association is moderate in effect size and independent of a number of potential confounders. Because exposure to movie alcohol use is common, it may have important population implications as a potentially modifiable risk factor.
OBJECTIVE: Little is known about the impact of viewing depictions of alcohol in entertainment media on adolescent drinking behavior. Our aims were to assess drinking in a sample of popular contemporary movies and to examine the association of movie alcohol exposure with early-onset drinking in an adolescent sample. METHOD: We conducted a school-based cross-sectional survey (N=4655) with longitudinal follow-up of never-drinkers (N=2406) involving adolescents ages 10-14 years and recruited from 15 New Hampshire and Vermont schools. Screen depictions of alcohol use were timed for each of 601 popular contemporary movies. Each adolescent was asked if he/she had seen a unique list of 50 movie titles, randomly selected from the larger pool. Movie alcohol use was summed for movies the adolescent had seen, adjusted to reflect exposure to the larger pool and modeled as a continuous variable. RESULTS: Ninety-two percent of the movies in the sample depicted drinking; median screen time for movie alcohol use was 2.5 minutes (interquartile range [IQR]: 0.9-5.0 minutes). Median exposure to movie alcohol use from the 601 movies was 8.6 hours (IQR: 4.6-13.5 hours). Overall 23.1% of the cross-sectional sample had tried alcohol, and 14.8% of initial nondrinkers had tried alcohol at the follow-up assessment. We found statistical evidence to support a curvilinear association between higher exposure to movie alcohol use and increased risk of prevalent and incident alcohol use, with a statistically significant linear and quadratic effect, and suggesting a higher dose-effect relationship at lower movie alcohol exposure levels compared to higher levels. The linear and the quadratic associations remained strong and significant in cross-sectional and prospective models after controlling for sociodemographics (grade in school, school, gender, parent education), personality characteristics of the adolescent (sensation seeking, rebelliousness, self-esteem), school performance, parenting style, and smoking experimentation, suggesting that exposure to movie alcohol use is an independent risk factor for early-onset alcohol use. CONCLUSIONS: This study provides the first evidence for an association between exposure to movie alcohol use and early-onset teen drinking. The association is moderate in effect size and independent of a number of potential confounders. Because exposure to movie alcohol use is common, it may have important population implications as a potentially modifiable risk factor.