White teenagers who are the most avid watchers of R-rated movies or who have television sets in their bedrooms are more than twice as likely to take up smoking compared with white teens who don't, according to a report published today.
Experts said the study confirmed Hollywood's pervasive influence by showing that even when other risk factors -- such as peer smoking -- were taken into account, media exposure remained a powerful force on white children.
The report also found that although African American teens watched more R-rated movies than their white classmates and were more likely to have their own TVs, their rate of smoking wasn't linked to their viewing habits.
"Why is it that whites are responsive to this and blacks aren't? What's going on?" asked Dr. James Sargent, a professor of pediatrics at Dartmouth Medical School who studies teen smoking but wasn't involved in the report.
Other researchers have documented a broad link between media exposure and teen smoking, but the new study, published in the Archives of Pediatrics and Adolescent Medicine, is the first to demonstrate that the effect isn't universal.
Christine Jackson, a social ecologist at the Pacific Institute for Research and Evaluation in Chapel Hill, N.C., and colleagues from the University of North Carolina at Chapel Hill interviewed 382 white students and 353 black students from the central part of the state in spring 2002.
At the time, the students were 12 to 14 years old, and none had tried a cigarette.
Two years later, 34% of the black teens had started smoking, as had 27% of the white teens.
As Jackson had anticipated, the biggest factors were having best friends who smoked, being a thrill-seeker and having inattentive parents.
But she found that for white teens, high exposure to R-rated movies made them 2.7 times as likely to start smoking. Having a TV in their bedroom made them 2.2 times as likely to take up the habit.
"As a researcher and a parent, I believe that teens are influenced by a variety of things they view in movies, including smoking," said Madeline Dalton, director of the Hood Center for Children and Families at Dartmouth Medical School, who was not involved in the research.
Virtually every R-rated movie released from 1988 to 1997 portrayed favorable images of smoking, according to a 2002 study by Dalton, Sargent and other researchers.
Jackson said she wasn't sure why the media held less sway for African American teens. One possibility cited in the study was that black teens didn't relate as well to the movie characters because most were white.
Behavioral scientist Robert Wellman, in his own research on smoking and the media, found that programs that engaged children psychologically were the most effective in encouraging them to smoke.
"If black adolescents don't identify with white actors, then the smoking behavior of white actors would be less engaging and therefore less effective in influencing black adolescents' behavior," said Wellman, a professor at Fitchburg State College in Massachusetts.
But Sargent said he didn't think that was the whole story, because black teens watched movies starring black actors who smoked.
Perhaps black teenagers, who watch significantly more hours of movies and TV shows than white teens, become desensitized to the images of smoking.
Sargent has found a similar discrepancy between white and black teens in nationwide data that have not been published. Latinos appear to fall somewhere between blacks and whites in terms of media influence.
He said the sample didn't include enough Asians to draw any conclusions.