Martijn van Teffelen

Chapter 6 142 0,00 2,00 4,00 6,00 8,00 10,00 12,00 14,00 Baseline Post Pins Time CBMI AC Figure 4 Study 2 - VDT pins over time per condition Note. Error bars represent standard errors of the estimates. VDT = Voodoo Doll Task. Study 2 - Discussion In Study 2 we tested the efficacy of CBM-I intervention versus AC in a larger double-blind, sham-controlled clinical trial in people with clinical levels of hostility where the additional impact on hostility outcomes and general psychiatric symptoms was assessed. In line with our expectations, Study 2 revealed that CBM-I altered benign bias in a positive baseline VDT score, results showed a significant negative relationship between VDT score at post- intervention and working alliance (B = -0.04, SE = 0.02, Χ 2 = 6.32, p = .012, d = 0.42, 95%CI = -1.00 to -0.13). This suggests that less post-test aggressive behavior is related to a higher degree of working alliance. Perceived usefulness To explore if the perceived usefulness per condition was impacted by hostility levels we ran a binary logistic regression model with usefulness as dependent variable and condition, baseline hostility level (PID-5H score) and the condition by hostility interaction as independent variables. Within the CBM-I and AC conditions, respectively 72.5% and 40.0% of participants indicated they found the intervention useful. Moreover, results showed that for participants in the CBM-I condition the odds of indicating the intervention as useful are 29 times higher than in the AC condition ( OR = 29.46, p = .027, 95% CI = 1.48 to 585.65), and direction, and hostile bias and aggressive behavior in a negative direction compared to AC. Contrary to our