ACTRN12614000106639.

Methods	Single‐blind RCT.
Participants	Number screened: 1814. Number included: 93. Number followed up: 92. Number of withdrawals: 1. Diagnosis of DCD: DSM‐IV‐TR. Presence and absence of comorbid conditions: children were excluded if they had global developmental delay, an intellectual disability or other medical condition that is known to cause motor impairment. Regarding participants completing the study Age: in school with a school assistant (mean 6 years 3 months, SD 11 months), in school with a physiotherapist (mean 6 years 7 months, SD 11 months), in a clinic with a physiotherapist (mean 6 years 1 months, SD 13 months). Sex: in school with a school assistant (25 boys, 12 girls), in school with a physiotherapist (23 boys, 7 girls), in a clinic with a physiotherapist (18 boys, 8 girls). Ethnicity: majority were white.* Country: Australia. Setting: school and clinic. Sociodemographics: a mixture of ranks for the School Index for Disadvantage in each mode which is based on socioeconomic status. Inclusion criteria Aged 5 to 9 years. Parental concerns in the DCDQ. Exclusion criteria Other medical conditions.
Interventions	Intervention: in school with a school assistant vs in school with a physiotherapist vs in a clinic with a physiotherapist. Intervention schedule: 45 min/wk. Duration of intervention: 8 wk. Mode of delivery: face‐to‐face small group (4 to 6 children). Intervention material: all groups used predominantly task‐oriented therapy consisting of age‐appropriate activities and the principles of skill mastery. The intervention used meta‐themes (repetition of skills, heightened feedback about performance and experience at success). The warm‐up activities for each session addressed issues, such as strength, motor planning, and proprioception that are more consistent with the perceptual‐motor approach to intervention. Intervention procedure: warm‐up activities followed by functional activities. Each session had a theme (e.g. going to the beach, on the farm). Intervention provider: physiotherapist with paediatric experience (2 to 6 years) or school assistant with 3 hr DCD/intervention training and previous experience running gross motor groups, supervised by a physiotherapist. Place of intervention: school or outpatient clinic. Intervention compliance: overall attendance rates for participants at intervention sessions were 92% for group 1 (school/school assistant), 93% group 2 (school/physiotherapist) and 88% group 3 (clinic/physiotherapist).
Outcomes	Primary MABC test. TGMD‐2. Secondary Pictorial Scale of Perceived Competence and Social Acceptance. School Function Assessment. Adverse effects or events: none recorded.*
Notes	Study start date: 14 July 2006 ‐ actual date of first participant enrolment. Study completion date: 5 March 2007 ‐ actual date last participant enrolled. Sample calculation: yes. Ethics approval: yes. Comments from study authors Limitations: specific age group of children with DCD (5 to 9 years); public schools in southern metropolitan Adelaide, South Australia; school and clinic environments are not the only possible environments for intervention to take place and others could be considered such as sporting clubs or gyms; other personnel could also be considered including teachers or family; participation was not able to be analysed due to a poor response rate by teachers. Key conclusions of study authors Group intervention programmes for DCD can be run by either a health professional or school assistant (supported by physiotherapist) in either the school or clinic environment and provide successful outcomes. Comment from review authors Active control only. * Email correspondence with study authors: February 2016. We contacted the first study author and obtained raw data and supplementary information.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation generated from a computer random numbers table.
Allocation concealment (selection bias)	Low risk	Allocated by a third party independent of the study and stratified according to school socioeconomic status.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Impossible.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Assessor blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Only 1 dropped out (moved overseas).
Selective reporting (reporting bias)	Low risk	Protocol identified (ACTRN12614000106639).
Other bias	Low risk	Funding: authors acknowledged the Volunteer Service for Flinders Medical Centre Inc. for funding some equipment in this study. Conflicts of interest: study authors, who worked for School of Health Sciences and School of Exercise and Health Sciences at universities, indicated no conflicts of interest.