We used two complementary approaches to test whether the prevalence of alcohol/drug use changed from before to during the pandemic, seeking convergent findings. Older adolescents are more likely to use substances [17], so the goal of both designs was to separate the developmental, age-related increases in substance use from increases attributable to the pandemic. Our first approach was a pretest/posttest design: We tested whether the prevalence of substance use changed among a subsample of youth (n = 1079) who completed a main ABCD Study assessment in the months leading up to the pandemic (September 2019-January 2020) and completed the first survey during the pandemic. This window was intentionally brief to minimize the time elapsed between the prepandemic and during-pandemic surveys (and thus any expected developmental increases in substance use). Owing to study design, these 1,079 youth tended to be younger than the full sample (mean age = 11.8 vs. 12.4 years). While the pretest/posttest strength has the advantage of making a within-subject comparison of alcohol/drug use, it has the disadvantage of limiting sample size to only those youth who completed a prepandemic assessment within the acceptable interval.
