Scientist Discovers Neural Factors Useful for Predicting Alcohol Use in Youth

Alcohol Use in Youth

According to the current issue of the American Journal of Psychiatry, researchers at the University of California, San Diego School of Medicine have discovered 34 neural factors that predict adolescent alcohol consumption.

The Study

The study’s findings provide evidence that it’s possible to identify at-risk youth before they engage in heavy alcohol use.

Underage drinking is a frequent occurrence in the United States—two-thirds of 18-year-olds report regular alcohol use. The Centers for Disease Control say drinkers between the ages of 12 and 21 consume 11% of the alcohol ingested in the United States.

Adolescent drinking is known to cause adverse behavior, namely drunk driving, risky sexual behavior, marijuana use, suicide, and a poor academic record.

The study consisted of 137 adolescents, aged 12–14, who had never tried alcohol. They had neuropsychological tests and fMRIs taken of their brains, and were assessed annually. By the time the adolescents reached 18, 70 of them had become moderate to heavy users of alcohol, while 67 continued to be nonusers.

The scientists used a machine learning algorithm known as “random forests” to make a predictive model. Random forest classifications can accommodate large sets of variables while using smaller study samples.

It was found that 12- to 14-year-olds were more susceptible to early age drinking if:

  • Their neuroimaging results indicated thinner cortices (the outer layer of neural tissue covering the brain)
  • They were male and/or came from a better socioeconomic background
  • They reported dating, had more externalized behaviors, such as lying and cheating, and believed alcohol would help them socially
  • They produced poor results on executive function tests

The study didn’t include the early use of marijuana, because only 15% of the sample recorded the eventual use of marijuana at least 30 times. The authors believe that it’s possible that the noted risk factors also apply to marijuana and other drug use. Studies with larger sample sizes are recommended.