Discrete Shifts in Symptom Networks During Psychotherapy for Chronic Depression – A Regime-Switching Network Analysis

Le résumé fourni par la source

The network approach to psychopathology conceptualizes mental disorders as systems of interacting symptoms rather than consequences of a latent factor. Recent developments have focused on modeling changes in symptom networks over time, where most approaches assume that these changes occur continuously. We propose an alternative by modelling discrete shifts between network states using hidden Markov models (HMMs). We conducted a secondary analysis of a multicenter randomized trial comparing disorder-specific and non-specific psychotherapy for chronic depression (n = 254). We identified two distinct states with the HMM: a low-symptom level state showing weaker symptom connectivity and a high-symptom level state with stronger positive connections. While participants typically remained in the same state across consecutive time points (80-88% of the time), transition probabilities differed between treatment groups. In particular, participants receiving disorder-specific psychotherapy showed higher probability of transitioning from the high-symptom to low-symptom state compared to the non-specific psychotherapy (19.5% vs. 12.0%). The regime-switching approach to analysis proved feasible and theoretically plausible, offering practical advantages for treatment evaluation within the network framework. By examining differences in network structure during healthy versus disordered states, this method provides insights into symptom dynamics that could inform both research and clinical applications.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.