Teens may be less likely to drink and smoke if their friends participate in substance abuse prevention programs, a recent U.S. study suggests.

Researchers focused on sixth graders in rural communities in Pennsylvania and Iowa who attended schools that offered seven weeks of counseling to both students and parents to see if kids who didn’t participate in this program might still benefit from it.

Three years later, in ninth grade, the students who didn’t attend the counseling sessions were 40 percent more likely to get drunk and more than twice as likely to smoke if none of their friends went through the program either, compared with those who skipped the program but had at least three friends who did participate.

“Adolescents are often influenced by their friends,” said lead study author Kelly Rulison, a researcher in public health and education at the University of North Carolina at Greensboro.

“Teens whose friends participated in a family-based substance use prevention program benefited from the program even though their own families did not participate,” Rulison said by email.

Roughly one third of U.S. high school students drink, and one in five of them admits to doing so excessively or getting in a car with a drunk driver, according to a recent survey of teens by the Centers for Disease Control and Prevention (CDC).

Nearly nine in ten smokers tried their first cigarette by age 18, according to the CDC. Each day in the U.S. alone, more than 3,800 youth aged 18 or younger smoke their first cigarette.

To see if teens cautioned against drinking and smoking might influence their friends to abstain, Rulison and colleagues followed 5,449 students who didn’t participate in counseling that was offered at their schools.

During sixth grade, students and their parents who did join the program went through seven weeks of counseling, meeting separately for an hour and then together for another hour.

Parent-only sessions focused on establishing rules, discipline and communication, while teens concentrated on resisting peer pressure and building social skills. Together, parents and children explored ways to improve communication and cohesiveness within their families.

Researchers also surveyed teens who didn’t participate in these sessions, questioning them about drug and tobacco use as well as asking them to name up to two best friends and up to five other close friends.

At the start of the study, there were no differences in tobacco or alcohol use among the teens who didn’t participate in the counseling program.

But as time passed, the teens who had more friends in the program were much less likely to get drunk or smoke than their peers who didn’t have any.

In this real-world study, researchers didn’t control which teens participated in the program or whether nonparticipants had friends who went through the counseling sessions, the researchers note in the Journal of Adolescent Health. That makes it impossible to say for sure whether having friends in the program, rather than other factors, may have encouraged the teens outside the program to drink and smoke less.

But it’s not surprising that some teens may be able to sway their friends to avoid these behaviors, said Ken Winters, founder of the Center for Adolescent Substance Abuse Research at the University of Minnesota in Minneapolis.

“It is likely that the influence is greatest when the teen receiving the program is a leader among his peers – a peer influencer rather than a peer follower,” Winters, who wasn’t involved in the study, said by email.

The study also offers more evidence to parents of the benefit of keeping tabs on their children as well as their kids’ friends, said Bradley Boekeloo, a researcher in community health at the University of Maryland in College Park, who wasn’t involved in the study.

“The study does suggest that parent involvement in their kids’ free time through monitoring of unstructured activity is very important to preventing youth drunkenness and smoking,” he said by email.