5 111 PERSONALIZED MONITORING AND FEEDBACK draft ESM questionnaire based on these reports. At the same time, participant and therapist (JWvdB) worked on the case formulation using the same file information and the participant’s insights in and explanations for the perpetuation of his former sexual offense(s). With the criteria for formulating ESM items in mind (supplemental materials in Riese et al., 2021), the questionnaire was further developed by each participant in collaboration with the therapist (JWvdB). While compiling ESM questions, they considered patients’ preferred word use, self-identified risk-relevant features and their assumed interrelationships, and their association with the sexual offenses. After consulting experts on the use of ESM in clinical samples (HR and DCvdV), the personalized questionnaire was finalized. 5.2.2.2 Feedback report Participant feedback reports contained bar graphs presenting the mean scores and 95% confidence intervals for each risk-relevant feature. Information about interrelations was provided using time series plots showing the trajectories of each variable over time, and by computing two networks for each participant (i.e., a temporal and a contemporaneous network; Epskamp, van Borkulo et al., 2018). The temporal network indicated whether the risk-relevant features predicted each other from one assessment to the next, whereas the contemporaneous network indicated to what extent risk-relevant features interacted with each other within the same measurement window. In both networks, risk-relevant features are represented by nodes, while arrows and edges indicate respectively temporal or contemporaneous associations between these features. Temporal and contemporaneous networks were estimated using the graphicalVAR package in R (Epskamp, 2020), and visualized with the qgraph package in R (version 1.9; Epskamp et al., 2012). The feedback report of participant #3 is shown in Figure 5.1. 5.2.2.3 Instruments Four instruments were used to answer research questions on the extent to which personalized information on interrelated risk-relevant features can be deployed to inform forensic case formulation in men with a history of sexual offenses and to evaluate effects of collecting personalized information. 5.2.2.3.1 Post-study interview The semi-structured interview designed for this study contains open-ended questions regarding: the understandability and perceived importance of graphs in the feedback report, the extent to which this report and registering risk-relevant features on smartphone adds value for the case formulation, and the effect of the ESM measurement. The recorded interviews were transcribed for all but two participants, who did not consent to record the interview. For these two cases, the interviewer notes were used. Participant responses were partitioned into quotes, assigned to an initial code, and combined for
RkJQdWJsaXNoZXIy MjY0ODMw