PREPRINT - Predictors, Moderators and Mediators of Treatment Outcome in Randomized Controlled Trials for Adolescents with Depression: A scoping review

Background: Advancing the field of precision medicine in the treatment of depression in adolescents requires an examination of predictors, moderators and mediators (PMMs) of depression symptom outcome in randomized controlled trials (RCTs). Aims: This scoping review describes findings of PMM analyses in RCTs for the treatment of depression in adolescents. Methods: Databases searched included: MEDLINE, Embase, APA PsycInfo and CINAHL. Included publications tested PMMs of depression symptom outcome (e.g. symptom reduction, remission, etc.) in RCTs pertaining to the treatment of adolescents, ages 13-17. Results were extracted and coded for statistical significance, analytic approach, a priori model development, and adjustment for multiple comparisons. Aggregated results were summarized by PMM variable domain and RCT sample.Results: Eighty-one papers reporting on PMM results across 33 RCTs were identified. Fifty-three variable domains were tested as baseline predictors of depression severity outcome, 41 as moderators, 19 as post-baseline predictors, and 5 as mediators. Most reported results were nonsignificant. For variable domains tested as a PMM in three or more RCTs, only improvements in sleep throughout treatment and early response predicted outcome with complete agreement across relevant studies. The two publications that reported a priori hypotheses and adjustment for multiple comparisons both found that baseline depression symptom severity and family conflict predicted poorer outcomes. Conclusions: Clinicians need to recognize the current limited evidence from RCTs to practice precision medicine principles when treating adolescents with depression. Our findings inform future study design for researchers who wish to incorporate questions about specific PMM variables into their study methods.