Shigematsu 2013.

Methods	Participants randomised using code numbers issued by coauthor Outcomes blinded
Participants	1 centre in Japan 20 participants with stroke > 4 weeks Baseline characteristics similar Clinical, video endoscopic, and videofluoroscopic evidence of dysphagia
Interventions	Rx: 1‐mA anodal tDCS C: sham tDCS (n = 10) Treatment for 10 days
Outcomes	Dysphagia Outcome and Severity Scale, PAS, VFSS, video endoscopic evaluation of dysphagia
Notes	Exclusions: subarachnoid haemorrhage, history of epileptic seizures, severe consciousness disturbance, organic neck disease, history of surgery except for tracheotomy
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomised via code numbers issued by coauthor
Allocation concealment (selection bias)	Low risk	Allocation concealed by code numbers
Blinding (performance bias and detection bias) All outcomes	Unclear risk	Participant blinding unclear
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Unclear
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Outcomes blinded (rehabilitation doctor and speech‐language hearing therapists did not know participants' group allocation)
Incomplete outcome data (attrition bias) All outcomes	Low risk	None lost to follow‐up
Selective reporting (reporting bias)	Low risk	Results of the Dysphagia Outcome and Severity Scale reported pre‐, post‐, and at 1‐month follow‐up
Other bias	Low risk	None identified