Martijn van Teffelen

Chapter 6 132 Assessed for eligibility (n = 951) - Inpatient (n = 343) - Outpatient (n = 420) - Community (n = 188) Excluded (n = 816) - Not meeting inclusion criteria (n = 681) - Declined to participate (n = 127) - Other reasons (n = 10) Analysed (n=61) - 1 missing session (n = 11) - 2 missing sessions (n = 7) - 3 missing sessions (n = 2) - 5 missing sessions (n = 2) - 6 missing sessions (n = 3) - 7 missing sessions (n = 3) - Lost to follow-up (hack, early therapy start) (n=8) - Discontinued intervention (burden, non-response) (n= 9) Allocated to CBM-I (n = 74) - Received allocated intervention (n = 61) - Withdrew before signing informed consent (n = 13) - Lost to follow-up (hack, technical problems, early therapy start) (n = 14) - Discontinued intervention (burden, non-response) (n = 6) Allocated to active control (n = 77) - Received allocated intervention (n = 74) - Withdrew before signing informed consent (n = 3) Analysed (n= 74) - 1 missing session (n = 10) - 2 missing sessions (n = 9) - 3 missing sessions (n = 5) - 4 missing sessions (n = 4) - 5 missing sessions (n = 1) - 6 missing sessions (n = 1) - 7 missing sessions (n = 4) - 8 missing sessions (n = 1) Allocation Analysis Follow-Up Randomized (n = 151) Enrollment Figure 1 Study 2 - Participant flow diagram Note . Unfortunately, some participants were lost to follow-up due to a successful cyberattack on Maastricht University on 23 December 2019; these participants are labeled as ‘hack’. State measures. To measure different aspects of state hostility and general psychiatric symptoms in the past three days participants completed four measures. First, they were administered an adapted state-version of the eight-item hostility subscale of the Aggression Questionnaire (AQ-HS) (Buss & Perry, 1992), measuring self-reported hostile thoughts in the past three days, which is scored on a 5-point Likert scale ranging from 1 (totally disagree) to 5 (totally agree). Second, we administered the 15-item State-Trait Anger Expression Inventory-2