Chapter 8 196 peer groups in addition to starting many IMR groups was not feasible for participating institutions. Few problems in addiction, medication adherence, and insight at baseline contributed to finding no effects in these domains No significant effects were found in the domains of the three illness management components: addiction, medication adherence, and insight. Moreover, no statistically significant improvement over time was observed in either the experimental or control group. This may be because most participants had few problems in these three domains. Consequently, there is little scope for improvement. No multiple adjustment methodologies were applied As we aimed to comprehensively explore the effects of the IMR, we examined several secondary outcome measures that could raise questions regarding the need to correct for this. However, multiple adjustment methodologies have been criticized within the statistical, epidemiological, and medical literature because they would trivialize interpretative problems (31-33). In addition, multiplicity adjustments are not considered appropriate for exploratory secondary analyses (31, 32). Therefore, we suggest that the interpretation of our findings may be relevant. Possible IMR contamination of the control group After the pilot study, IMR was provided as an optional care modality to all outpatient teams of the participating institutions except for the ACT teams. During the initial years of data collection at the largest participating institution, these outpatient teams were called Illness Management and Recovery (IMR) teams. All clinicians in these teams received a two-day IMR basic training. Therefore, clinicians treating patients in the control group may have applied IMR principles or interventions. This may have contributed to reducing differences in treatment characteristics between the experimental and control conditions, thereby reducing the impact of IMR in this study. However, a possible remedy as using a cluster-randomized design, randomized by location, was no longer possible in this context. This may have led to an underestimation of the effects of IMR. However, only the experimental group received IMR in the group format. Future directions In future research, it would be interesting to examine the effects of IMR on illness self-management and all three types of recovery when all design and
RkJQdWJsaXNoZXIy MjY0ODMw