# Dynamics of depression symptoms in adolescents during three types of psychotherapy and post‐treatment follow‐up

## TL;DR

This study explores how depression symptoms in adolescents change during and after three types of psychotherapy, revealing that insomnia and fatigue are key predictors of other symptoms.

## Contribution

The study applies longitudinal network analysis to identify symptom dynamics and treatment mechanisms in adolescent depression.

## Key findings

- Insomnia and fatigue showed the highest outstrength centrality during treatment, predicting other symptoms.
- Worthlessness was the central symptom during post-treatment follow-up.
- Different therapies may lead to similar outcomes through distinct mechanisms.

## Abstract

According to the network theory of mental disorders, psychopathology emerges from symptoms that causally influence one another and create interconnections and feedback loops that maintain atypical mental states. Analysis of symptom networks during and following psychotherapy may provide clues to some of the mechanisms through which change occurs. Youth with depression are an important population in which to better understand psychotherapy mechanisms because current evidence‐based interventions for this population show only modest effects.

Participants were adolescents with major depressive disorder (N = 465; ages 11–17; 75% female) in a randomized controlled trial comparing cognitive behavioral therapy, short‐term psychoanalytical psychotherapy, and brief psychosocial intervention (IMPACT, ISRCTN83033550). Eleven self‐reported depression symptoms were used to compute two longitudinal networks: (1) treatment phase, using baseline, 6 and 12 weeks data; and (2) follow‐up phase, using 36, 52, and 86 weeks data.

During the treatment phase, all depression symptoms were interconnected. Symptoms of insomnia and fatigue showed the highest outstrength centrality (ability to predict other symptoms over time). In contrast, few symptoms were interconnected during the post‐treatment phase except worthlessness, which had the highest outstrength centrality. Allowing network parameters to differ across the three treatment types improved model fit during the treatment phase and revealed that symptoms with the highest outstrength centrality varied by treatment type.

Individual symptoms may make key contributions to subsequent depressive psychopathology in adolescents. Longitudinal network analysis reveals that insomnia and fatigue predict other symptoms, allowing for consideration of specific mechanisms associated with depression treatment. The findings further suggest that negative cognitions about the self may emerge as a central putative cognitive vulnerability in those with a history of depression. Our exploratory findings also suggest that the three therapies (cognitive behavioral therapy, short‐term psychoanalytical psychotherapy, and brief psychosocial intervention) may have achieved equifinality in part through different mechanisms.