Study designs of evaluations included in the review
Studies with four or more participants that used some form of control (whether another group of patients or patients used as their own control) were eligible for inclusion. The review included randomised controlled trials (RCTs) and pre-test post-test studies.
Specific interventions included in the review
Studies evaluating CIT in which the unaffected upper limb was restrained and the affected limb was treated with intensive therapy were eligible for inclusion. In the included studies, the unaffected limb was constrained using one or more of a padded mitt, hand splint or sling. The interventions evaluated included at least one of the following: CIT circuit training, occupational therapy (OT), behavioural training, shaping, repetitive tasks, physiotherapy (PT), structured therapy, intensive force use and task-orientated therapy. The control interventions, where these existed, included the following, alone or in various combinations: traditional OT, PT, bilateral circuit training, standard care, less intensive therapy and no intervention. The duration of CIT interventions ranged from 12 days to 10 weeks.
Participants included in the review
Studies of adults (aged over 18 years) with a diagnosis of ischaemic or haemorrhagic stroke resulting in hemiparesis were eligible for inclusion. Studies were included regardless of the severity of stroke or the length of time since stroke onset. Overall, 63% of the patients were male, the age ranged from 33 to 83 years, the time since stroke onset ranged from 2 days to 17 years (representing acute, sub-acute and chronic stroke) and, where reported, 71% had a right-sided hemiparesis.
Outcomes assessed in the review
Studies that assessed the following functional outcomes were eligible for inclusion: Action Research Arm Test, Fugl-Meyer Assessment of Motor Recovery, Motor Assessment Log (MAL), Wolf Motor Function Test, Functional Independence Measure, Barthel Index, Actual Amount of Use Test, Arm Motor Ability Test and Functional Test of the Hemiparetic Upper Extremity. Studies had to report sufficient data to enable the calculation of an effect size (d).
How were decisions on the relevance of primary studies made?
Two researched independently selected studies and resolved any disagreements by discussion.