Predictors of Treatment Outcome in Adolescent Depression

Abstract

Major depressive disorder is a global public health concern that is common in adolescents. Targeting this illness at the early stages of development is critical and could lead to better long-term outcomes because the adolescent brain is highly plastic and, hence, neural systems are likely to be more malleable to interventions. Although a variety of treatments are available, there are currently no guidelines to inform clinicians which intervention might be most suitable for a given youth. Here, we discuss current knowledge of prognostic and prescriptive markers of treatment outcome in adolescent depression, highlight two major limitations of the extant literature, and suggest future directions for this important area of research. Despite significant effort, none of the potential demographic (gender, age, race), environmental (parental depression, family functioning), and clinical (severity of depression, comorbid diagnoses, suicidality, hopelessness) predictors have been robustly replicated to warrant implementation in clinical care. Studies on biomarkers that truly reflect pathophysiology are scarce and difficult to draw conclusions from. More efforts should be directed towards potential neurobiological predictors of treatment outcome. Moreover, rather than evaluating potential predictors in isolation, modern machine learning methods could be used to build models that combine information across a large array of features and predict treatment outcome for individual patients. These strategies hold promise for advancing personalized healthcare in adolescent depression, which remains a high clinical priority.