Methodology of antidepressant drug trials for treating depression in adolescents.

ABSTRACT Despite the sophistication of the currently available studies on antidepressant treatment of adolescent depression, various methodological problems limit the conclusions that can be drawn. It is difficult to evaluate trends across the available studies because of differences in definitions of treatment response, therapeutic response and placebo response rates, symptomatic subtypes of depression, comorbidity, drugs studied, and study designs that limit the statistical power to detect treatment effects. Some fundamental issues concern the sample selection and reliability of the measurement instruments, as well as the assumption that efficacy of antidepressants in adults is easy to demonstrate. None of the studies now available carefully examine behavioral, social, or academic variables that clinicians commonly use to assess antidepressant effects in adolescents. Before concluding that antidepressants are less effective in treating depression in adolescents than in adults, these methodological issues need further study. The main problem is not merely the small number of studies or low number of subjects. Until the validity of the diagnostic construct of depression in adolescents is better developed, or until more homogeneous samples of patients can be defined, it is pointless to encourage further attempts to assess antidepressant efficacy using the past methodologies.