Forrester 2014.

Methods	RCT Method of randomisation: not described Blinding of outcome assessors: stated as 'no' Adverse events: no Deaths: not mentioned Dropouts: 5 (3 in treatment group, 2 in control group) ITT: no
Participants	Country: USA 39 participants (21 in treatment group, 18 in control group) Not ambulatory at start of study Mean age: 63 years in experimental group and 60 years in control group Inclusion criteria: first stroke; residual lower extremity hemiparesis involving the ankle (1/5 to 4/5 MRC); capable of generating at least trace muscle activation in PF‐DF; adequate language and neurocognitive function; sit in the chair for 30 to 60 minutes per session of ankle training Exclusion criteria: total plegia (0/5) at paretic ankle; fixed or painful contractures; dementia; orthopaedic, arthritic, or inflammatory condition limiting ankle movement; deep venous thrombosis or pulmonary thromboembolism; vision impairment; severe receptive or global aphasia
Interventions	2 arms: Experimental group: robot‐assisted ankle training, daily for 60 minutes/day, 10 sessions until discharge Control group: passive manual moving and stretching ankle, for the same time
Outcomes	Outcomes were recorded at baseline and at discharge. Outcome measures: walking ability (Functional Independence Measure walking), balance (Berg Balance Scale), walking velocity, active range of motion, muscle strength, spatiotemporal gait parameters (step time, step length, step symmetry), motor control variables (angular velocity, co‐ordination)
Notes	Unclear amount of therapy
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not clearly described; authors only state "blocked randomisation"
Allocation concealment (selection bias)	High risk	Not described
Blinding of outcome assessment (detection bias) All outcomes	High risk	Authors state "not blinded".
Incomplete outcome data (attrition bias) All outcomes	High risk	5 participants were excluded from analysis after randomisation.