Background Group Support Psychotherapy (GSP) is an evidence-based treatment for mild to moderate depression among people living with HIV (PLWH). However, little is known about which psychosocial processes mediate its effects on reduction in depression symptoms and how these processes emerge and evolve over time. Methods We conducted a secondary analysis of a cluster-randomised trial in Uganda, comparing GSP to Group HIV Education among PLWH with mild to moderate depression. Using generalised structural equation modelling, we assessed whether changes in psychosocial factors mediated intervention effects on reduction of depression symptoms at 2 and 6 months. Single-mediator and sequential mediation models were used to evaluate early, late, and sequential psychosocial pathways. Indirect effects were estimated using bias-corrected bootstrap confidence intervals. Analyses accounted for clustering at the health-center level and district stratification, with additional sensitivity analyses accounting for health workers nested within health centers. Results In single mediator analyses, psychosocial changes occurring during therapy showed evidence of contributing to reduction in depression symptoms both at the end of treatment and six months later, although these effects were generally modest and sensitive to clustering assumptions. Improvements in active coping and decreased denial, behavioral disengagement, alcohol use and self-blame during therapy mediated symptom reduction with modest indirect effects. In contrast, post-treatment improvements in self-esteem, help-seeking, positive reframing, acceptance, and decreased stigma demonstrated larger and more robust indirect effects. Sequential mediation analyses suggest that mediation effects of emotional support seeking initiated during therapy, continued and strengthened over time, whereas mediation effects of self-esteem and stigma reduction emerged only after treatment. Conclusion Reduction in depression symptoms following GSP appears to be a dynamic process in which early psychosocial changes initiate symptom improvement, while later psychosocial changes sustain and strengthen these effects over time. Emotional support seeking may function as an upstream process that facilitates broader psychosocial change.