Park 2015a.

Study characteristics
Methods	Design: parallel RCT Duration of trial: 6 months (October 2013 to March 2014) Unit of randomisation: adult inpatients with cognitive impairment after stroke Recruitment and allocation: of 37 inpatients recruited and screened, 30 were eligible and randomised by random numbers table into IG (n = 15) or CG (n = 15).
Participants	Setting: rehabilitation hospital Country: Korea Sample size: 30 adults, 47% men; IG: 15; CG: 15 Inclusion criteria: history of 1 stroke; stroke with onset duration of < 3 months; score of ≤ 23 on the Korean version of MMSE; ability to understand instructions; ability to use the controller with the unaffected upper limb; and without unilateral hemispatial neglect and hemianopsia Age: mean: IG: 64.7 (SD 8.9) years; CG: 65.2 (SD 8.0) years Time since stroke onset: mean: IG: 1.5 (SD 0.5) months; CG: 1.8 (SD 0.6) months Types of stroke and lesion: not reported
Interventions	Intervention group Brief name: CoTras Recipients: adult inpatients with stroke undergoing rehabilitation within 3 months' onset with cognitive impairment Why: to provide a Korean computer‐based cognitive rehabilitation to enhance attention, concentration, implementation skills, and perception‐motor skills. What (materials): CoTras program (Netblue Co., Ltd, Korea) computerised cognitive rehabilitation training, made for Koreans; a joystick and a large button on the CoTras panel which make the training easier for people who are unfamiliar with computer use; see www.netblue.co.kr/eng/doc/product01-01.php What (procedures): during the sessions, all participants completed a standard rehabilitation programme according to a daily inpatient treatment schedule. In addition to standard rehabilitation, all participants received 30‐minute daily sessions of the computer‐based program treatment. Who provided: not reported; however, the research was conducted by occupational therapists, the software is commercially available, and the intervention is within the scope of practice defined in the review. How: unclear but appeared it was face‐to‐face and individually. Where: local inpatient rehabilitation hospital, no further details provided When and how much: 20 × 30‐minute sessions (i.e. 5 days per week for 4 weeks) (10 hours in total) Tailoring: training allowed adjusting to individual participant's abilities at all levels of the programme Modifications: none reported How well (planned): none reported How well (actual): none reported Comparator group Brief name: conventional cognitive rehabilitation Recipients: inpatient with stroke with cognitive impairment undergoing rehabilitation Why: to provide usual rehabilitation services. What: pencil and paper with emphasis on visual perception ability Who provided: not reported How: not reported, assumed face‐to‐face Where: inpatient rehabilitation hospital When and how much: matched that for the IG participants in terms of duration (in minutes) (i.e. 20 × 30‐minute sessions) Tailoring: not reported Modifications: not reported How well (planned): not reported How well (actual): not reported
Outcomes	Primary: none Secondary LOTCA Other MVPT‐3 Methods of data collection: all outcome measures were administered to the participants by the assistant researcher with 5 years' experience in using the measures. Data collection time points: baseline and 4 weeks postintervention
Notes	Funding: none stated Conflict of interest: none reported Published trial protocol: none located Trial registration: none reported Ethics approval: unclear, stated, "All subjects provided written informed consent before study inclusion according to the code of ethics of the World Medical Association (Declaration of Helsinki, version 2004)".
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random numbers table used to randomly assign participants to either CG or IG.
Allocation concealment (selection bias)	Unclear risk	Not reported.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not reported. Quote: "All subjects gave their voluntary consent to participate after receiving a detailed explanation of the purpose and methods of the study" and "During the sessions, all subjects participated in a standard rehabilitation program according to a daily inpatient treatment schedule. In addition to standard rehabilitation, all subjects received 30‐min daily sessions of either the CG or the EG [IG] treatment".  Comment: highly likely that participants and personnel knew which group they were in.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Quote: "All outcome measures were administered to the patients at baseline and at the end of treatment (after the 4‐week intervention) by the assistant researcher with 5 years' experience in using the measures". Comment: unclear whether this person was blinded to the allocation.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Study did not address this criterion.
Selective reporting (reporting bias)	Low risk	No protocol located but appeared to report all outcomes and data.
Other bias	Low risk	No other identifiable bias.