Overbeek 2013.

Methods	Multi‐centre cluster‐randomised controlled trial of a trauma‐focused psychoeducational intervention compared with supportive therapy in children exposed to interparental violence
Participants	Included (n = 164) Children 6 to 12 years of age who experienced psychological and/or physical interparental violence, who were referred to the programme by agencies such as the police and child protection agencies. Mean age: 9.22 years. Female: 44.5% (69). Mean length of abusive relationship: 10.87 years Excluded Children and parents with intellectual, psychiatric or behavioural problems that prevent participation in groups Setting Eight organisations in 7 cities in The Netherlands, 2009 to 2012
Interventions	En nu ik! (’It’s my turn now!’) (n = 108) Focused on how to differentiate and express emotions; increasing feelings of emotional security, learning how to cope with feelings and problems in a non‐violent way, developing a trauma narrative, improving parent/child interaction and psychoeducation, processing interparental violence experiences. Nine sessions of 90 minutes were provided for a maximum of 8 children, and 9 parallel sessions for the non‐violent custodial parent Therapists Therapists for parallel sessions usually included a mental health professional and a social worker who had received 1 day of training in the manualised intervention and participated in ≥ 3 peer supervision meetings. 69% of sessions were rated as having no or few deviations Jij hoort erbij ("You belong") (n = 56) Comparable with intervention on non‐specific factors by offering positive attention, positive expectations, recreation, distraction, warmth and empathy of the therapist, as well as social support among group participants. Also 9 sessions of 90 minutes for a maximum of 8 children, and 9 parallel sessions for the custodial parent Therapists Parallel sessions were provided by a mental health professional along with a mental health professional in training or a social worker. Therapists followed a manual for every session and participated in ≥ 3 peer supervision meetings. 78% of sessions were rated as having no or few deviations
Outcomes	PTSD symptoms Scale: Trauma Symptom Checklist for Young Children (TSCYC; 90‐item) Rater: parent Scale: Trauma Symptom Checklist for Children (TSCC; 54‐item) Rater: children ≥ 7.5 years of age Behaviour Scale: CBCL (119‐item) Rater: parent Scale: Teacher Report Form (133‐item) Rater: teacher Depression Scale: CDI (27‐item) Rater: children ≥ 7.5 years of age When At 1 week and 6 months after completion of the programme
Notes	Netherlands Trial register: NTR 3064
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	“An independent researcher will make the allocation schedule with a computerized random number generator”. Therapists were randomised in blocks of 3 (2:1 intervention:control)
Allocation concealment (selection bias)	Low risk	“Concealed random allocation”
Blinding of participants (performance bias	Low risk	Parents and children were blind to group allocation until 2 weeks before the start of the programme; both programmes were presented as useful
Blinding of outcome assessment (detection bias) All outcomes	Low risk	‘The researchers coding the observation tasks and analyzing the data will be blind to the group condition of parents and children, as well as assessment’
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	The ‘last value carried forward will be applied’ for missing data. Loss to follow‐up: 20.1%
Selective reporting (reporting bias)	Unclear risk	All outcomes listed in protocol were reported
Other bias	Low risk	No other bias was identified