Objectives. Not all patients with post-traumatic stress disorder (PTSD) respond to cognitive behavioural therapy (CBT). Literature suggests group music therapy might be beneficial in treating PTSD. However, feasibility and effectiveness have not been assessed. The study objectives were to assess whether group music therapy was feasible for patients who did not respond to CBT, and whether it has an effect on PTSD symptoms and depression. Design. The study employed mixed methods comprising of an exploratory randomized controlled trial, qualitative content analysis of therapy, and patient interviews. Method. Patients with significant PTSD symptoms (n = 17) following completion of CBT were randomly assigned to treatment (n = 9) or control groups (n = 8). The treatment group received 10 weeks of group music therapy after which exit interviews were conducted. Control group patients were offered the intervention at the end of the study. Symptoms were assessed on the Impact of Events Scale-Revised and Beck Depression Inventory II at the beginning and end of treatment. Results. Treatment-group patients experienced a significant reduction in severity of PTSD symptoms (−20.18; 95% confidence interval [CI]: [−31.23, −9.12]) and a marginally significant reduction in depression (−11.92; 95%CI: [−24.05, 0.21]) at 10 weeks from baseline compared to the control. Patients viewed music therapy as helpful and reported experiences concur with current literature. Conclusions. Group music therapy appears feasible and effective for PTSD patients who have not sufficiently responded to CBT. Limitations include the small sample size and lack of blinding. Further research should address these limitations, test sustainability, and identify specific factors that address symptoms in treatment.