Geroin 2011.

Methods	RCT Method of randomisation: software‐generated randomisation scheme Blinding of outcome assessors: no Adverse events: none Deaths: none Dropouts: none ITT: yes
Participants	Country: Italy 30 participants (10 in treatment group 1, 10 in treatment group 2, and 10 in control group) 5 in both treatment groups and 7 in control group were ambulatory at start of study Mean age: not stated Inclusion criteria: at least 12 months from their first unilateral ischaemic stroke; age < 75 years; European Stroke Scale score between 75 and 85; MMSE score ≧ 24; ability to maintain standing position without aids for at least 5 minutes; ability to walk independently for at least 15 metres with the use of walking aids (cane and orthoses) Exclusion criteria: preceding epileptic fits; an electroencephalography suspect of elevated cortical excitability; metallic implants within the brain and previous brain surgery; medications altering cortical excitability or with a presumed effect on brain plasticity; posterior circulation stroke; deficits of somatic sensation involving the paretic lower limb; presence of vestibular disorders or paroxysmal vertigo; presence of severe cognitive or communicative disorders; presence of other neurological or orthopaedic conditions involving the lower limbs; presence of cardiovascular comorbidity; performance of any type of rehabilitation treatment in the 3 months before start of study
Interventions	3 arms: Robot‐assisted gait training (Gait Trainer GT 1) combined with transcranial direct current stimulation Robot‐assisted gait training (Gait Trainer GT 1) combined with sham transcranial direct current stimulation Walking overground All participants received 10 x 50‐minute treatment sessions, 5 days a week, for 2 consecutive weeks
Outcomes	Outcomes were recorded at baseline and after 2 weeks: Primary outcomes were the 6‐minute walk test and the 10‐metre walking test Secondary outcomes were spatiotemporal gait parameters, FAC, Rivermead Mobility Index, Motricity Index leg subscore, and Modified Ashworth Scale
Notes	We combined the results of both robotic‐assisted groups (arms 1 and 2) into a single group, which we compared with the results of the control group (arm 3).
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Software‐generated list
Allocation concealment (selection bias)	Low risk	Central allocation
Blinding of outcome assessment (detection bias) All outcomes	High risk	Not done
Incomplete outcome data (attrition bias) All outcomes	Low risk	No missing outcome data