Qian 2017.

Methods	Design: RCT, parallel group
Participants	Country: China Setting: clinical Sample size calculation: not reported Sample size: 86 children Sex: 54 = boys, 14 = girls (group one: 32 (84.2%) = boys, 6 (15.8%) = girls; group two: 22 (73.3%) = boys, 8 (26.7%) = girls) Age: 6‐12 years (group one: mean = 8.3 years (SD = 1.3); group two: mean = 7.8 years (SD = 1.2)) Ethnicity: not reported Socioeconomic status: not reported IQ: group one: mean = 105.7 (SD = 13.9); group two: mean = 101.8 (SD = 10.4) ADHD diagnosis: group one: inattentive = 17 (44.7%), hyperactivity‐impulsivity = 0 (0%), combined = 21 (55.33%); group two: inattentive = 16 (53.3%), hyperactivity‐impulsivity = 1 (3.3%), combined = 13 (43.3%) ADHD medication: group one: 10 participants had maintained steady dosage of medications for more than a half year and remained unchanged during the entire study. New medications could not be initiated during the study. Comorbidity: group one: ODD = 7 (18.4%), learning disorder = 8 (21.1%), special phobia = 5 (13.2%); group two: ODD = 7 (23.3%), learning disorder = 4 (13.3%), special phobia = 2 (6.7%) Medications for comorbid disorders: not reported Inclusion criteria: diagnosis of ADHD: meeting DSM‐IV criteria based on parent ratings on ADHD‐RS‐IV confirmed by semi‐structured interview by experienced paediatric psychiatrist using the clinical diagnostic interview scale six to 12 years of age Exclusion criteria: history of head injury diagnosis of other congenital or acquired neurological conditions estimated full‐scale IQ < 80 diagnosis of autism spectrum disorder, psychosis, or an emergent psychiatric condition that needed immediate medication new medications could not be initiated during the study. Baseline characteristics: no significant difference found in reported baseline characteristics
Interventions	86 participants allocated to one of two groups Group one (n = 44): executive skills training based on Dawson Guare's (Dawson 2010) training with the content adapted culturally to ensure acceptability to Chinese children. Groups of six to eight families received 12 weekly × one‐hour sessions in clinical setting. First and last sessions included parents; it was not clear whether the children participated. The first session focused on setting behavioural goals, action plans, environmental modifications and reward systems and how to help with homework. The last session was on how to continue to use the learned skills. Only children participated in sessions two to 11 and handbook‐specified homework between sessions. Group two (n = 42): waiting list *Analysis, including baseline characteristics, was based on the following sample sizes: intervention = 38 participants, waiting‐list control = 30 participants Attendance: 86.4% (38/44) of children in intervention group complied with the training, completing 10 or more sessions in the 12‐session period. All missed group sessions were administered to the trainee individually. The percentage of the number of sessions administered individually was not specified.
Outcomes	Primary outcomes Social skills: Weiss Functional Impairment Rating Scale, social activities subscale, parent‐rated Emotional competencies: BRIEF, emotional control subscale, parent‐rated General behaviour: BRIEF, total score, parent‐rated Secondary outcomes Core ADHD symptoms: ADHD‐RS, fourth version, parent‐rated Outcome assessment: post‐intervention assessment
Notes	Study ID:NCT02327585 Sponsorship source: the study was supported by grants from the Beijing Municipal Science and Technology Commission (No. Z151100004015103), the Major State Basic Research Development Program of China (973 Program, No. 2014CB846100), National Key Research Plan of Ministry of Science and Technology of China (No. 2016YFC1306103), and the Capital Health Development Research Fund (No.2011‑4024‑04). Year conducted: 2017 Duration of the study: 12 weeks Comments: the study was conducted in accordance with the Declaration of Helsinki and was approved by the Institutional Review Board of Peking University Sixth Hospital. Lead author's name: Ying Qian Institution: Child Psychiatric Research Center, Peking University Sixth Hospital (Institute of Mental Health), National Clinical Research Center for Mental Disorders, Key Laboratory of Mental Health, Ministry of Health (Peing University) Email: yangli_pkuimh@bjmu.edu.cn, corresponding author Dr Li Yang Address: Beijing 100191, China
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Comment: randomised block design with participants randomised to a block that comprised a permutation of four participants, two for each group separately. The design was used to balance the individuals between the intervention and waiting‐list groups.
Allocation concealment (selection bias)	Low risk	Comment: randomisation grouping concealed in envelopes and recruited participant notified of his or her group sequentially
Blinding of participants and personnel (performance bias) All outcomes	High risk	Comment: blinding not possible
Blinding of outcome assessment (detection bias) All outcomes	High risk	Comment: primarily used parent rating scales and parents were not blinded to group status. Unclear, however, if assessors of executive functioning performance tests were blinded to group status
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Comment: 6/44 children in intervention group and 12/42 children in control (waiting‐list) group dropped out. No reasons given for dropout. Baseline data for the 18 children who dropped out not provided and no analysis of attrition
Selective reporting (reporting bias)	High risk	Comment: The study registration specified the following secondary outcomes, which were not mentioned or reported in the paper: Conners; and Cambridge Neuropsychological Test Automatic Battery. The following measures were not mentioned in the study registration but were reported in the paper: BRIEF; WEISS Functional Impairment Scale‐Parents.
Vested interest bias	Unclear risk	Comment: funding source not reported
Other sources of bias?	Low risk	Comment: no other sources of bias identified