Burgar 2011.

Study characteristics
Methods	RCT Medical clearance to participate Eligibility Consent Baseline testing Stratification by FMA upper extremity score and randomised at each site into 3 groups. Hi dose Robot, Lo dose Robot, and control (all extra to existing therapy) Intervention Testing and 6‐month retention For this Cochrane Review, we only compared the 2 robot groups
Participants	54 participants, in 3 study arms, completed the preintervention testing and at least 5 hours of treatment Robot‐Lo group: n = 19; mean age 62.5 years; mean 17.3 days poststroke Robot‐Hi group: n = 17; mean age 58.6 years; mean 16.6 days poststroke Gender of participants: not reported No significant difference between the sites, other than in time since stroke There was a significant difference in age between the groups 3 inpatient settings in the USA
Interventions	The intervention was robot therapy.  Robot‐Lo group (and the control group): up to 15 × 1‐hour therapy sessions over 3‐weeks  Robot‐Hi group: 30 × 1‐hour therapy sessions over 3 weeks  Intervention was terminated when the participant received the maximum number of sessions, or when they were discharged from acute inpatient rehabilitation
Outcomes	FMA upper limb Upper limb portion of the  FIM Modified Ashworth Scale WMFT Measurements were taken before study initiation, after completion of training and at 6‐month follow‐up
Notes	Very few of the participants received the maximum number of additional input, as planned. Early discharges, scheduling conflicts, and patient tolerance were among the factors that reduced the total amount of therapy Conflict of interest: authors declared no competing financial interests Funding: supported by VA Rehabilitation and Service Development (B2695)