Cohen 2005.

Methods	Randomised trial of sexual abuse‐specific CBT vs non‐directive supportive therapy
Participants	Included (n = 82) Females 8 to 15 years of age who had experienced validated contact sexual abuse by someone ≥ 5 years older within the previous 6 months, had significant symptoms related to the abuse and had a non‐offending parent or primary caregiver able to participate in treatment. Participants were referred from a variety of sources. Mean age: 11 years. Female: 69%. Caucasian 60%, African American 37%, bi‐racial 2%, Hispanic 1%. Trauma exposure: 54% of participants experienced anal and/or vaginal intercourse, 33% were abused more than 10 times, and, for most, the abuse was accompanied by use of threatened or actual force Excluded Children with mental retardation, pervasive developmental disorder, active psychosis or substance abuse or serious medical illness, or whose caretaker was not long‐term (≥ 12 months) or had active psychosis or substance abuse Settting Urban outpatient child psychiatric programme specialising in the treatment of traumatic stress in children, USA
Interventions	Sexual abuse‐specific CBT (n = 41) CBT methods with the child focused on depression, anxiety and behavioural difficulties and with the parent focused on parental emotional distress, enhanced support for the child and management of behavioural difficulties. Major components for the child included identification of feelings, stress inoculation techniques, gradual exposure exercises, cognitive processing of the abuse, education about healthy sexuality and safety skill building. Parental components paralleled those for the child, with the addition of building parenting management skills. Each of the 12 weekly sessions consisted of 45 minutes with the child and 45 minutes with the parent Non‐directive supportive therapy (n = 41) Therapists did not provide specific suggestions or directive advice but encouraged exploration of alternative attributions, behaviours and feelings. Issues were addressed as they were raised by child or parent. Interventions provided a high degree of non‐judgmental empathy and support; encouraged identification, clarification and acceptance of upsetting feelings; and re‐established trust and positive interpersonal expectations. Each of the 12 weekly sessions consisted of 45 minutes with the child and 45 minutes with the parent Both therapies Appropriate referrals to non‐therapeutic ancillary care were made when indicated, and mothers in either group with a DSM‐III‐R diagnosis were also offered a referral for individual therapy Therapists Both therapists had received a master's level degree in clinical social work and were trained and experienced in both models. Halfway through the programme, therapists swapped to the other treatment model. Both therapies were manualised. Therapists received weekly supervision, and sessions were audiotaped and audited weekly to ensure adherence, which was rated at > 90%
Outcomes	Trauma‐related symptoms Scale:Trauma Symptom Checklist for Children (54‐item) Rater: child/adolescent Depression Scale: Children’s Depression Inventory (27‐item) Rater: child/adolescent Anxiety Scale: State‐Trait Anxiety Inventory for Children TSCC Rater: child/adolescent Behaviour Scale: Child Behavior Checklist Rater: parent When Post therapy and at 6‐ and 12‐month follow‐up
Notes	Because loss to follow‐up was greater than 54% in the supportive therapy group, only loss to follow‐up data were used in this review
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random number series generated by computer
Allocation concealment (selection bias)	Unclear risk	Not reported
Blinding of participants (performance bias	Unclear risk	Participants could not be blinded, but both groups received a psychological therapy
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Outcome assessment could not be blinded, as all measures were child reported or parent reported
Incomplete outcome data (attrition bias) All outcomes	High risk	Last observation carried forward was used to account for missing data, but loss to follow‐up was greater than 54% for the NST group at all intervals and was high overall: post therapy 40%, 6 months 44%, 12 months 52%
Selective reporting (reporting bias)	Low risk	All outcomes appear to have been reported
Other bias	Low risk	No other bias was apparent