BackgroundIn posttraumatic stress disorder (PTSD), physiological reactions during trauma scripts have been anchored to the diagnosis. According to the construct of research domain criteria, physiological reactions and self-rating scales could be used to evaluate treatment effects. ObjectiveIn the present study, self-rated PTSD symptoms combined with physiological reactions during trauma script were used as outcome measurements in the domain of arousal/regulatory systems in a controlled randomised study of Group Music and Imagery (GrpMI) treatment for females with PTSD or complex posttraumatic stress disorder (CPTSD) related to violence and/or sexual abuse. Methods45 traumatised women were randomised to 12 weeks of active treatment or waiting. Before and after the intervention, an assessment was done using physiological measurements during script-driven imagery (SDI) procedures. Subjective Units of Distress (SUD) were collected immediately after the trauma script (TS). Reactions during the SDI procedure were reported using the Responses to Script Driven Imagery (RSDI) scale, measuring re-experiencing, avoidance, and dissociation. Self-reported PTSD symptoms were accessed using the PTSD checklist for DSM-5 (PCL-5). Heart rate (HR), heart rate variability (HRV), and electrodermal activity were sampled during the baseline (BL) and TS conditions of the SDI procedure. As a measure of trauma-related reactivity, the difference between TS and BL was used for statistical calculations. HRV measures included high (HF; 0.15-0.4 Hz) and low (LF; 0.03-0.15 Hz) frequency band power, the LF/HF ratio, and the root mean square of successive inter-beat differences (RMSSD). Measures of electrodermal activity included skin conductance levels (SCL) and frequencies of non-specific skin conductance responses (NS-SCR). Further, correlations between self-rated PTSD symptoms and physiological reactivity measures were analysed. ResultsDuring the TS, the absolute levels of HR, LF/HF-ratio, and NS-SCR, as well as the trauma-related reactivity of HR, RMSSD, HF, LF/HF-ratio, SCL, and NS-SCR, showed significant changes indicating decreased arousal during trauma script after treatment. Compared to the waitlist control, an interaction analysis showed significant treatment effects in the BL level of HR, the absolute TS level of HR and HF, and the trauma-related reactivity of RMSSD and HF, suggesting an improvement of vagal function in the treatment group. Significant treatment-related reductions were found in symptoms of PTSD, re-experiencing and avoidance, as well as SUD. The changes pre- to post-treatment in HR reactivity and self-rated PTSD symptoms correlated significantly. Furthermore, the initial HR reactivity predicted treatment outcome as measured with PCL-5. ConclusionIn the evaluation of treatment methods for PTSD, a combination of self-report and physiological measures seems to be feasible. The physiological measures, in combination with a robust decrease in self-rated PTSD symptoms, indicate that trauma-focused GrpMI is a promising treatment for PTSD or CPTSD. More studies are needed to confirm the results, and further research comparing with other active treatments is necessary to establish the precise role of the treatment.