Shirk 2014.

Methods	RCT of modified CBT vs usual care for referred adolescents with a depressive disorder and a history of interpersonal trauma
Participants	Included (n = 43) Adolescents referred for outpatient treatment who had been exposed to interpersonal trauma and met diagnostic criteria for a depressive disorder. Mean age: 15.48 years. Female: 36. Type of trauma: reported at least 1 incident of physical abuse (49%), witnessing family violence (58%), sexual abuse (67%) and verbal/emotional abuse (47%) in response to a highly structured screening interview. Diagnosis of PTSD: modified CBT group 30%, usual care group 61% Excluded Receiving concurrent psychological treatment for (1) depression, (2) attempted suicide within 3 months before intake, (3) engaged in self injurious behaviour that required hospitalisation or emergency room treatment within the previous 3 months, (4) met diagnostic criteria for bipolar disorder and/or substance dependence disorder, (5) presented with psychotic symptoms or intellectual deficit. None were receiving antidepressants Setting Two outpatient clinics of a large urban community mental health centre in an ethnically diverse urban city in the Rocky Mountains, USA
Interventions	Modified CBT (n = 20) A CBT protocol for adolescent depression was modified to address cognitive deficits and distortions associated with interpersonal trauma. Core elements of the manualised therapy included mood monitoring, cognitive restructuring, relaxation training and interpersonal problem solving and mindfulness and explicitly addressed cognitions related to the experience of interpersonal trauma Therapy consisted of 12 weekly individual sessions, which were not time‐limited; however, the mean number of sessions attended was 5.33 Therapists Therapy was delivered by 1 male doctoral level psychologist with 28 years of clinical experience and 1 female master's level therapist with 10 years of experience. Therapists completed a 1‐day workshop and a practice case under supervision before the start of the trial Graduate‐level coders rated 86% of treatment elements delivered as specified in audio recordings of sessions Usual care (n = 23) Treatment strategies and procedures that therapists used regularly and believed to be effective for particular case formulations. Therapists described these as eclectic, although favouring client‐centred, psychodynamic and family approaches Usual care also consisted of weekly individual sessions, which were not time‐limited. The mean number of sessions attended was 6.22 Therapists Usual care therapists were 2 female doctoral level psychologists with 3 and 4 years of clinical experience. They were supervised by the clinic team leader; this included weekly group supervision and individual consultation on an as needed basis
Outcomes	PTSD symptoms Scale: Disorders and Schizophrenia ‐ Present and Lifetime Version (K‐SADS‐LS) Rater: adolescent Depression Scale: Beck Depression Inventory (BDI‐II; 21‐item) Rater: adolescent Scale: K‐SADS‐LS (BDI‐II; 21‐item) Rater: adolescent Behaviour Scale: CBCL (118‐item) Rater: caregiver When BDI‐II was used during sessions 1, 4, 8 and 12 and at 16 weeks post treatment. K‐SADS‐PL and CBCL were rated post treatment
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Clinicians were assigned on the basis of clinic location, but randomisation was not described
Allocation concealment (selection bias)	Unclear risk	Not described
Blinding of participants (performance bias	Low risk	Both groups received individual therapy
Blinding of outcome assessment (detection bias) All outcomes	Low risk	PTSD symptoms, behaviour and depression scales were completed by participants and caregivers
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Last observation carried forward (LOCF) was used for missing post‐treatment BDI‐II scores. Loss to follow‐up: 18.6%
Selective reporting (reporting bias)	High risk	Only BDI‐II depression scores for females were reported
Other bias	High risk	Percentage of participants with a baseline diagnosis of PTSD was twice as high (61%) in the control group as in the intervention group (30%). Therapists in the intervention group had 38 years of clinical experience between them as compared with 7 years in the usual care group. Also, intervention therapists received 'more targeted and frequent' case supervision