Chen 2016a.

Methods	Computer‐generated random numbers by independent research staff Assessors blinded
Participants	Multi‐centre trial in China 250 participants; 148 male 100% stroke within 2 to 7 days Dysphagia identified by bedside swallowing assessment and videofluoroscopic swallowing study Baseline characteristics and prognostic values similar between both groups
Interventions	Rx: acupuncture and conventional stroke rehabilitation care C: conventional stroke rehabilitation care only Duration: 3 weeks Follow‐up: 7 weeks
Outcomes	Primary outcome: NIHSS index Secondary outcomes: FMA for motor function, rate of recovery based on BSA, VFSS, MMSE, and MoCA
Notes	Exclusions: serious heart, liver, and kidney‐related diseases; blood coagulation dysfunction; inability to complete the MMSE test or bedside swallowing assessment; congenital disabilities; posterior circulation infarcts; receiving thrombolytic; participated in other clinical trials within previous 3 months; pregnant or breastfeeding
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated random numbers provided by independent research staff
Allocation concealment (selection bias)	Low risk	Random numbers placed into sequentially numbered, opaque, sealed envelopes
Blinding (performance bias and detection bias) All outcomes	High risk	Participants and acupuncturist aware of treatment allocations. All allopathic medical staff and rehabilitation therapists blinded
Blinding of participants and personnel (performance bias) All outcomes	High risk	Participants and acupuncturist not blinded
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Outcome assessors blinded
Incomplete outcome data (attrition bias) All outcomes	High risk	5 participants lost to follow‐up; 4 discontinued intervention. Not all participants given VFSS examination
Selective reporting (reporting bias)	Low risk	All outcomes reported
Other bias	Low risk	None identified