Choi 2015.

Methods	Design: RCT, parallel group
Participants	Country: Korea Setting: outpatient Sample size calculation: not reported Sample size: 80 children Sex: 32 (44%) = boys, 40 (56%) = girls Age: mean = 11.2 years (SD = 0.93, range = 9‐13) Ethnicity: not reported Socioeconomic status: not reported IQ: all IQ > 90 ADHD diagnosis: subtypes not reported ADHD medication: all participants were under medication at the time of intervention, type of medication not reported Comorbidity: not reported Medications for comorbid disorders: not reported Inclusion criteria meeting ADHD DSM‐IV criteria based on a structured interview by psychiatrist total Wechsler Intelligence Scale for Children (WISC) IQ score above 90, based on the full‐scale IQ (WISC, Revised Korean Version, Kwak, Park, Kim, 2001) (Kwak 2002) Behavior Problem Scale score on Child Behavior Checklist (CBCL) within clinical range Exclusion criteria: not reported Baseline characteristics: no significant differences on study background variables or pretest measures
Interventions	80 participants allocated to one of three groups Group one (n = 25): Emotion Management Training (EMT), which is an emotion identification and expression treatment, and consists of four major components: (1) identification and labelling of emotional words; (2) emotional recognition and expression; (3) emotional understanding; and (4) emotional regulation in social situations. Each session began by discussing any problems or issues related to homework from the previous one, followed by exercises, and ending with an evaluation of the session. 50‐minute × once‐weekly sessions lasting 16 weeks. Group two (n = 28): social skills training (SST) programme (based on studies conducted by Elliot 1991 and Pfiffner 1997). SST is a form of behavioural training focused on teaching various social skills to children with ADHD to improve their interaction with peers and teachers. It uses various behavioural techniques such as prompts, role play, and reinforcement. Each session was focused on teaching a particular social skill such as listening skills, conversation skills, joining in, and reacting to rejection, negotiating, and reacting to being teased and criticised. Each session started with discussing homework, followed by exercises, and ending with a new homework assignment and an evaluation of the session. Group three (n = 27): waiting list. Children were later randomised into one of the two programs. Attendance: at least 12 of 16 sessions of either EMT or SST. Mean number of sessions attended by the 75 programme completers was 14.9 (SD = 1.3), with an overall attendance rate of 90.5%. No group differences in the number of sessions attended
Outcomes	Primary outcomes Social skills: Peer Relational Skills Scale, child‐rated Emotional competencies: Emotion Expression Scale for Children, child‐rated Outcome assessment: post‐intervention, one week after end of intervention
Notes	Study ID: not reported Sponsorship source: the author(s) received no financial support for the research, authorship, and/or publication of this article. Year conducted: 2015 Duration of the study: 16 weeks Comments: ethics approval. The study was reviewed and approved by the Research Ethics Commitee of the university at which the experiment was conducted. Lead author's name: Eun Sil Choi Woo Kyeong Lee Institution: Kyungil University, Korea and Seoul Cyber University, Korea Email: wisemind96@iscu.ac.kr Address: 193‐15, Miadong Kangbuk‐gu, Seoul, 142‐700, Korea
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Comment: from descriptions, it was unclear if all children had been included before randomisation. However, the sentence describing how WL children after 16 weeks were pooled with newly selected children indicated that randomisation was made progressively. It was unclear who did the sampling in blocks, how these blocks were generated, and if this process was to be considered random.
Allocation concealment (selection bias)	Unclear risk	Comment: it was unclear exactly how allocation concealment was done.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Comment: parents, children and trainers were aware of group's status.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Comment: the PICO measures of emotion expression and peer relational skills are both self‐report questionnaires.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Comment: two children in the EMT group and three children in the SST group did not complete the study. The numbers of non‐completion were small and balanced. However, while it was stated that the reason for non‐completion was dropout during treatment, the reason for this dropout was not specified.
Selective reporting (reporting bias)	Unclear risk	Comment: as we did not locate a study registration, it was unclear whether all planned measures had been reported accordingly. However, the measures presented in the paper were all described in the results section.
Vested interest bias	High risk	Comment: it was not specified if the therapist delivering the interventions was also one of the authors. The first author is the author of the manual used in one of the intervention arms. There may have been a bias given the first author's investment in the first study arm programme.
Other sources of bias?	High risk	Comment: no information on comorbid disorders. It was mentioned that all participants were under medication at the time of the intervention but it was not clear if this referred to ADHD medication or medication for comorbid disorders. It was not specified whether, for example, autism would be a reason for exclusion.