A naturalistic pilot study 43 (51, 55); in other earlier research on IMR, cut-off scores for good or high IMR fidelity were > 3.7 (6), > 3.8 (56). In their review, McGuire et al. reported a weighted mean of 4.05 on fidelity for “all studies” (SD= .93)(5). We also wished to identify the aspects of fidelity (as shown in item-scores) in which quality of implementation of IMR had to improve. Additionally, we tested whether the institute had successfully created a good infrastructure for trainers' supervision and training, our criteria for success being the institute's ability to fulfill its intention of completing a two-day training for all trainers, and of supervising trainers for two hours a week. The IMR review (5) refers to five studies in which training of trainers had taken two days (1, 51, 55, 57, 58). While three other studies reported that training took 40 hours (59), five days (8) and 48 hours (6) respectively, the latter had involved only 50% of the trainers. On the assumption that dropout from treatment should be minimized, we wanted to establish how much dropout from treatment we could expect in the main study. We set no prior targets for the number of completers. The review of IMR studies refers to a median dropout rate from IMR of 24% and a range of dropout rates from 18% to 30%; this review also refers to a median of 63% completers and a weighted mean of 36%, with a range of 15%-86% (5). Of the four earlier RCTs, one reported an IMR drop-out rate of 21 % (6), and a second reported that 46% of the participants assigned to the program attended fewer than half of the IMR sessions (7). A third reported 5% drop-out (8), but participants of this study had been selected “on the basis of consistent attendance of prior (non-IMR) services, and training and consultation focused heavily on consumer engagement” (5). The fourth RCT reported that drop-out from IMR—defined as participating in less than half the scheduled groups—was 72% (9). Our fourth secondary feasibility criterion that would enable us to plan the main study properly is whether the duration of IMR was predictable. According to the review of IMR studies (5), 9-12 months is a usual length for a program consisting of one session per week; in three earlier RCT studies in which IMR was also applied in a weekly group format, IMR lasted for 8-11 months (6) and 9 months (8, 9). Statistical Analysis We used a paired sample t-test to measure the effectiveness of IMR (one group pre- and post- measurement) on the IMR-scale client version, the IMR-scale clinician version, and the RMQ. Chi square tests and independent samples t-tests were used to test differences between completers’ and non-completers’ baseline characteristics.
RkJQdWJsaXNoZXIy MjY0ODMw