Keyes · The Journal of adolescent health : official publication of the Society for Adolescent Medicine 2020 · repeated cross-sectional survey study · n=58,444

Diverging Trends in the Relationship Between Binge Drinking and Depressive Symptoms Among Adolescents in the U.S. From 1991 Through 2018.

Cited 40 times in the scientific literature.

Level 4 - case-series / case-control

Repeated cross-sectional survey data (analogy for observational epidemiology)

PubMed 31676228 · doi:10.1016/j.jadohealth.2019.08.026 · record verified 2026-08-31

What was done

Data were analyzed from the US nationally representative cross-sectional Monitoring the Future surveys collected from 1991 to 2018 among 58,444 school-attending 12th-grade adolescents. Binge drinking was defined as any occasion of more than five drinks in the past 2 weeks. Depressive symptoms were assessed using a four-item scale (e.g., belief that life is meaningless or hopeless) dichotomized at the 75th percentile. Time-varying effect modeling was conducted to evaluate trends in the association across time by sex, race/ethnicity, and parental education.

What was found

In 1991, adolescents with high depressive symptoms had 1.74 times the odds of binge drinking (95% confidence interval 1.54-1.97). By 2018, the strength of the association between depressive symptoms and binge drinking declined 24% among girls and 25% among boys. There was no statistically significant relation between depressive symptoms and binge drinking among boys since 2009; among girls, the relationship remained positive across most of the study period, with no significant relationship in 2016-2017.

Why it matters

As adolescent binge drinking declines while depressive symptoms increase, their historical co-occurrence is uncoupling. Prevention and intervention programs cannot assume that adolescent depression and alcohol misuse share identical drivers or continue to track together in modern cohorts.

Limits

The repeated cross-sectional design cannot establish within-individual trajectories or causal relationships. The sample is restricted to school-attending 12th graders, omitting adolescents who dropped out or were absent. Depressive symptoms were measured using a brief four-item self-report screener rather than a clinical diagnostic assessment. All data were self-reported and subject to recall and social desirability biases.

Cited by