5 122 CHAPTER 5 offenses might also reduce the time needed to prepare the feedback report. An example of a web-based application for personalized ecological momentary assessment in psychiatric care that enables both the standard set of items and generating a feedback report (bar graphs and time series plots) at relatively short notice is PETRA (Bos et al., 2022). Because participants indicated that especially the network graph is valuable in understanding interrelationships of risk-relevant features, it is recommended to investigate to what extend network graphs can be conducted automatically. However, temporal networks could not be conducted by graphicalVAR for all participants in this study. This is not surprising, as recent simulation studies show that sensitivity (i.e., the true-positive rate, indicating the proportion of edges that had nonzero values in both the true and estimated networks) of graphicalVAR is low with sample sizes feasible for clinical practice (Epskamp, Waldorp et al., 2018; Hoekstra et al., 2022; Mansueto et al., 2022). According to Hoekstra and colleagues (2022), sensitivity can be increased by reducing the number of nodes (to 6 variables), or by increasing the timeseries (500 timepoints). Since personalized ESM measurements often include more than 6 items and 500 items of measurement is virtually unfeasible within clinical practice, therapists and men with a history of sex offenses should rely on contemporaneous networks to examine interrelationships of risk-relevant features. Results on the added value of the ESM measurement and the resulting feedback for the case formulation, as well as the (low) burden participants experienced are consistent with findings from studies in psychiatric patients (e.g., Kroeze at al., 2017; van den Bergh et al., 2022; Von Klipstein et al., 2020). The same applies to the high degree of adherence and (cognitive and behavioral) engagement with ESM in forensic populations (Habets et al., 2022; Hillbrand & Waite, 1994). 5.4.1 LIMITATIONS Some limitations of this study should be mentioned. The first concerns the sample size. Although selecting a suitable sample size in qualitative research is an area of conceptual debate and practical uncertainty, a sample of five participants is unlikely to allow for the attainment of saturation (Vasileiou et al., 2018). Second, all patients were adult white male patients admitted for outpatient forensic treatment. As a result, generalization of the study outcomes to other populations is not possible (e.g., juveniles, women, men with different cultural or ethnic backgrounds, incarcerated people convicted of sexual offenses). Third, this study exclusively focused on self-reported perceptions and experiences of the participants. Future studies could focus more on the experiences and opinions of therapists. Fourth, in the absence of a golden standard for ESM study duration (Dejonckheere & Erbas, 2022), the duration of ESM data collection was two weeks. As found in a recent meta-analysis of Vachon and colleagues (2019), this is a common ESM study duration (N = 72, range 1 – 150 days, mean 11.2 days). Nevertheless, the impact of using longer (or shorter) study durations could be examined in future research.
RkJQdWJsaXNoZXIy MjY0ODMw