What is a Good Outcome of Treatment for Adolescent Depression? A Mixed-Methods Exploration of Measurement, Concepts, and Priorities

Doctoral Thesis


Background and aim: Evidence-based care places outcome measurement at the core of mental health practice and research. But there is no consensus on what constitutes a ‘good outcome’. This thesis aimed to advance the debate specifically for adolescent depression, through a mixed-methods exploration of outcome concepts, priorities, and measurement.

Methodology: (1) A narrative review of the outcomes literature in mental health identified an initial taxonomy of outcomes. (2) A systematic review considered outcomes reported in recent treatment effectiveness studies. (3) Qualitative content analysis explored outcome perspectives amongst youth, parents, and clinicians following therapy. (4) A Q-methodological study assessed the relative importance given to different outcomes by youth and practitioners. (5) Quantitative analysis examined the convergence of reliable change ratings across selected outcome domains and measures in a naturalistic dataset.

Results: (1) based on all five studies an outcome taxonomy for adolescent depression was developed, consisting of 32 outcomes across seven domains. (2) Treatment effectiveness studies primarily reported on change in symptoms (94%), followed by global functioning (55%). (3) Symptom change was a key focus for youth, parents, and clinicians post treatment, but closely followed by coping, family functioning, and academic functioning. (4) Four distinctive viewpoints emerged on what outcomes matter the most: Symptoms – feeling better; Self-management – resilience through coping skills; Parental support – resilience through family support; and Functioning – less interference with daily life. (5) Symptom change appears to be an imperfect proxy for change in functioning and progress towards personal goals.

Conclusion: Change in depressive symptoms emerged as a core outcome. However, youth, parents, and clinicians endorsed additional outcomes, conveying a multidimensional picture that is inadequately captured by measuring symptoms alone. To promote outcome assessment that is streamlined, person-centred, and can illuminate treatment mechanisms, the consideration of additional outcomes, beyond symptom change, may be beneficial.

University College London