Trauma-related memories in PTSD after interpersonal violence: an ambulatory assessment study.

Background: Ambulatory assessment (AA) is increasingly recommended for assessing symptoms of posttraumatic stress disorder (PTSD). Previous AA studies provided new insights into the phenomenology of trauma-related memories, but also divergent findings. Notably, the range of trauma-related memories (a major target of psychotherapeutic interventions) reported in AA studies was as wide as 7.3 to 74.5 per week which might result from different methods used in these studies. Objective: We aimed at assessing the frequency of trauma-related memories in PTSD related to interpersonal violence and investigated whether this frequency is dependent upon the method. Method: For each patient trauma-related memories were assessed using two variants of smartphone-based AA: (1) Event-based sampling (EBS), i.e. participants entered data on each intrusive memory as it occurred; (2) Time-based sampling (TBS), i.e. participants reported the number of trauma-related memories they had experienced during the last two hours after they had been alerted by the smartphone. The numbers reported during the TBS-block were either analysed as reported by the participants or restricted to one per hour (rTBS). The impact of smartphone-assessments on trauma-related memories was assessed during a post-monitoring questionnaire. Results: While trauma-related memories were frequent across assessments, the methodology had a huge impact on the numbers: EBS (median = 7) and rTBS (median = 6) yielded significantly lower weekly numbers of intrusive trauma-related memories than TBS (median = 49). Accordingly, the possibility to report unrestricted numbers of trauma-related memories clearly impacted the results. The post-monitoring questionnaire identified another source for the divergent findings: while feeling disrupted by the smartphoneassessments was unrelated to the numbers reported during EBS, feeling disrupted was related to an increase of trauma-related memories during TBS and rTBS. Conclusions: The method clearly impacts the recorded number of trauma-related memories. Future research should clarify whether other variables (e.g. the subjective stress related to intrusive memories) are less dependent on the methodology.