7 151 DISCUSSION risk of sexual offending. More specifically, change scores have additional value and should be taken into consideration in decision making on supervision and continuation of judicial measures. Second, the NBM-RSR states that key dynamic risk factor(s) can be elucidated that, when changed, are most likely to affect other dynamic risk factors and thereby are more likely to reduce the overall probability of sexual reoffending. From this model, the reverse also applies, in that dynamic risk factors with relatively fewer or weaker interrelationship with other dynamic risk factors in the network will have relatively little impact on the risk of sexual reoffending. Treatment providers and probation officers could either eliminate or markedly reduce their focus on such factors. Third, accepting the possibility that conditional dependencies in the network structures found in Chapters 3 and 4 indeed reflect mutualistic causal interactions between dynamic risk factors on a group level, risk management and treatment strategies to reduce recidivism would benefit from a stronger focus on dynamic risk factors with relatively strong strength centralities: social rejection/loneliness, cognitive problem-solving skills, impulsive behavior, and callousness. Fourth, the case studies presented in Chapter 5 might serve as a starting point for future use of ESM methodology in clinical work with adult men with a history of sexual offenses. Implementation and integration of ESM methodology in the treatment process by systematically collecting data on the impact and effectiveness of treatments in everyday clinical practice – for example, as part of routine outcome monitoring (ROM; de Beurs et al., 2011) – could provide new insights into patterns of the dynamic course of interrelated risk-relevant features on an individual level. According to participants, these patterns, most easily communicated through the use of network graphs, added value to their case formulation (for current practice see, for example, Craig & Rettenberger, 2018; Craig & Rettenberger, 2021; van Beek, 1999; and Vrinten et al., 2015). Increasing awareness of personal patterns in the dynamic course of riskrelevant features and of their possible associations with the risk of sexual reoffending obtained from this blended ESM methodology may enhance individual and group-based therapies, which in turn, may improve forensic outcomes and lower risk of reoffending. 7.4 LIMITATIONS 7.4.1 LIMITATIONS OF THE STUDIES IN THIS DISSERTATION Despite the theoretical and clinical relevance, some limitations of the studies presented in this dissertation should be acknowledged. A first limitation concerns the studies’ sample sizes. Regarding the meta-analysis, only a modest number of studies (52 studies; N = 13,446) could be included since research into the predictive propensities of dynamic risk assessment instruments developed for adult men was in an early stage at the time we conducted our analysis. This applies in spades to the studies on the predictive propensity of change scores. Regarding the sample size of studies on the
RkJQdWJsaXNoZXIy MjY0ODMw