Teenagers beware—drinking too many non-diet soft drinks may spark violent behavior.

A study published on Oct. 24 in the journal “Injury Prevention”—titled “The ‘Twinkie Defense’: The relationship between carbonated non-diet soft drinks and violence perpetration among Boston high school students”—suggests that adolescents who drink more than five cans of non-diet soft drinks per week are more likely to carry a weapon or act violently toward peers.

According to the study, regression analysis was used to determine the relationship between soft drink consumption and

teenagers’ behavior. The method was used because of the multiple variables in

the study.

“We used a technique called robust variance estimators to account for the fact that respondents were not all separate and individual, but they were grouped together in 22 schools,” said Sara Solnick, department chair of economics at the University of Vermont, in an email.

Logistic regression is a method for determining the relationship between predictor variables and a dependent variable, according to

TheMeasurementGroup.com.

Solnick and David Hemenway, director of the Harvard Injury Control Research Center, conducted the survey.

The study began in 2008. Boston high school students ranging from 14–18 years old participated. Of the 31 schools eligible, only 22 cooperated. Solnick and Hemenway used the Boston Youth Survey, a biennial paper-and-pencil survey to question the 1,878 9th – 12th grade students who participated in the study. Approximately 30 percent of the sample drank more than five cans of non-diet soda per week. Cans were measured at 12 ounces, or 355 mL, and respondents were told to count a 20-ounce bottle as two cans.

In the survey, students were asked, “In the past seven days, how often did you drink soda?” Students could not include diet soda.

According to the study, Solnick and Hemenway measured the effect of carbonated drink consumption on aggression and violent behavior in three ways: whether the respondent had been violent toward a peer, another child in his or her family and/or someone the respondent was in a dating relationship with.

They also examined whether the respondent carried a weapon during the past year.

The study found that 40.3 percent of the students who drank five or more cans per week had carried a gun or a knife, 56.7 percent had been violent to peers and 42 percent had been violent toward children in his or her family, while 26.2 percent were violent in dating relationships. Solnick said the violence seems to be

dose-related, meaning the more the teenager drank, the larger the possibility of violence. According to the study, 35.1 percent of students who drank one or less can per day were violent towards peers, while 46 percent of those who drank two to four cans were violent toward peers.

“My biggest concern with this study is that they don’t explain what they mean by violence,” said Jann Gumbine, licensed psychologist and professor at the University of California, Irvine College of Medicine.

Solnick believes the study is accurate. It had various limitations, however. The data was based on self-report, meaning participants evaluated themselves, but Hemenway and Solnick have no reason to believe the respondents who drank more soft drinks are more likely to exaggerate their violent behaviors than those who consumed fewer carbonated beverages, according to the study.

“[Solnick and Hemenway] admit they didn’t even ask what type of soda the teens were drinking, so we don’t know if the drink was caffeinated or not, or what ingredient is believed to cause the violence,” Gumbine said. “I would interpret this study cautiously until it is replicated and more research is conducted.”