A laboratory-based test of the relation between adolescent alcohol use and panic-relevant responding.

A burgeoning literature supports a link between alcohol use and panic-spectrum problems (e.g., panic attacks, disorder) among adolescents, but the direction of influence has yet to be properly examined. From a theoretical perspective, panic-spectrum problems may increase risk for problematic drinking via affect regulation efforts (e.g., self-medication), and problematic consumption also may increase or initiate panic-relevant responding (e.g., learning or kindling models). The objective of the current investigation was to examine the role of prior alcohol use in predicting panic-relevant responding, as well as panic symptom history in predicting the desire to consume alcohol, in the context of either a voluntary hyperventilation or a low-arousal task. Participants were community-recruited adolescents aged 12-17 years (n = 92, Mage = 15.42, SD = 1.51; 39.1% girls). Results indicated that prior alcohol use predicted panic-relevant responding among those undergoing the hyperventilation task (but not the low-arousal task), and that this finding was robust to the inclusion of theoretically relevant covariates (i.e., age, sex, negative affectivity). However, panic symptom history did not predict the desire to consume alcohol as a function of either the hyperventilation or low-arousal condition. This work sheds further light on the nature of the relation between panic-spectrum problems and problematic alcohol use in adolescence. Specifically, the current findings suggest that frequent alcohol use may increase panic vulnerability among adolescents, whereas acute panic symptoms may not elicit the immediate (self-reported) desire to drink.