Green 2011.

Study characteristics
Methods	Allocation: randomised using a minimisation procedure controlling for: i) frequency of SH; ii) diagnosis of conduct disorder; iii) diagnosis of major depression; iv) psychosocial distress. Follow‐up period: 6 and 12 months. N lost to follow up: "Loss to follow‐up was low (<4%)" (p.1).
Participants	Inclusion criteria: i) aged 12 years to 16 years 11 months; ii) presenting to child and adolescent services with a history of at least two episodes of SH in the previous 12 months. Exclusion criteria: i) unable to communicate in English; ii) diagnosed with severe low weight anorexia nervosa; iii) diagnosed with psychosis; iv) attends a special learning disability school; v) currently in secure care. Numbers: of the 366 participants, 183 were allocated to the intervention arm and 183 to the control arm. Profile: 88.5% (n = 324) were female, 100% (n = 366) were multiple repeaters, 62.0% (n = 227) were diagnosed with a depressive disorder; 33.3% (n = 122) were diagnosed with a behavioural disorder. Source of participants: adolescents presenting to local child and adolescent services with a history of at least two episodes of SH within the previous 12 months. Location: Manchester and Chester, UK.
Interventions	Experimental: manualised developmental group psychotherapy involving elements of cognitive behavioural therapy, dialectical behavioural therapy, and group psychotherapy. Control: treatment as usual according to the clinical judgement of the adolescents' child and adolescent mental health service team. Treatment as usual excluded any type of group‐based intervention. Therapists: clinicians with a minimum of thee years post‐qualifying experience and who also received training in delivering developmental group psychotherapy. Type of therapy offered: group psychotherapy Length of treatment: Acute treatment phase: 6 weeks. Weekly booster sessions continued for as long as required (maximum theoretical length of treatment unclear in trial report).
Outcomes	Included: i) repetition of SH according to self‐ and/or collateral report; ii) suicide; iii) suicidal ideation; iv) depression. Excluded: i) global functioning; ii) health economics information.
Notes	Source of funding: "This study was funded by the Health Foundation and sponsored by the University of Manchester" (p.11). Declaration of author interests: none stated.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Allocation was by minimisation controlling for: high or low self harm...presence of behavioural disorder...or depressive disorder...and presence or absence of high psychosocial risk..." (p.4). Comment: use of a minimisation algorithm is likely to have minimised the role of bias in the generatioU of the randomisation sequence.
Allocation concealment (selection bias)	Low risk	Quote: "Randomisation was by remote telephone..." (p.4). Comment: Randomisation by remote telephone would have enabled the allocation sequence to have remained adequately concealed.
Blinding (performance bias and detection bias) Of participants	High risk	Quote: "It was not possible to blind...the participants themselves to treatment allocation" (p.4).
Blinding (performance bias and detection bias) Of personnel	High risk	Quote: "It was not possible to blind clinicians...to treatment allocation" (p.4).
Blinding (performance bias and detection bias) Of outcome assessors	Low risk	Quote: "The main outcomes were recorded by outcome assessors blinded to treatment allocation...Assessor guesses of allocation following end point were no better than chance" (p.4).
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Quote: "Analysis of primary and secondary outcomes was by intention‐to‐treat subject to availability of data" (p.4). Comment: Although the authors describe using an intention‐to‐treat approach to analyses for both primary and secondary outcomes, not all participants allocated to the intervention and control groups are included in subsequent analyses suggesting that outcome reporting may be incomplete.
Selective reporting (reporting bias)	Unclear risk	Comment: No reason to suspect that all outcomes were not measured, however, in the absence of the trial protocol, this cannot be ascertained.
Other bias	Low risk	Comment: No apparent other sources of bias.