Chou 2009.

Methods	RCT Method of randomisation: not stated Blinding: not stated Adverse effects: not stated ITT analysis: not stated Losses to FU: 5
Participants	Country: UK Number of participants included: 33 (17/16) Demographics: aged 59‐90 years, 48% male Type of stroke: both ischaemic and haemorrhagic strokes Diagnosis: CT/MRI Severity on entry: unclear Time from stroke onset: 13‐33 months Setting: inpatient Comparability: comorbidity and past history similar
Interventions	Comparison: real acupuncture + PT versus sham acupuncture + PT Acupuncture treatment Acupuncture rationale: stated Needling details Points used: body acupoints Numbers of points used: 2 acupoints Depths of insertion: 0.03 inch Deqi elicited: yes Needle stimulation: electrical Needle retention time: 20 minutes Needle type: the needles were 1 inch long Treatment regimen Number of treatment sessions: 16 sessions Frequency of treatment: 2/week Total course: 8 weeks Practitioner background: has been practicing Traditional Chinese Medicine for many years after graduating from medical school (China Medical University, Taichung, Taiwan) Co‐intervention: PT Control interventions: sham acupuncture + PT
Outcomes	Improvement of cognitive function (LOTCA‐G) Quality of life (SF‐36, SS‐QOL) FU: 8 weeks
Notes	‐
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	The method of random sequence generation was not reported
Allocation concealment (selection bias)	Unclear risk	Information on allocation concealment was not reported
Blinding (performance bias and detection bias) All outcomes	Unclear risk	Information on blinding was not reported
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Information on blinding was not reported
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Information on blinding was not reported
Incomplete outcome data (attrition bias) All outcomes	Low risk	Treatment: 3 participants could not finish the treatment protocol Control: 2 participants decided not to receive rehabilitation
Selective reporting (reporting bias)	Unclear risk	Free of selective reporting bias was assessed as 'unclear' due to some clinically important outcomes unstated, such as quality of life, mortality and adverse events
Other bias	Unclear risk	No information provided