Ten‐Brink 2017.

Study characteristics
Methods	RCT Setting: rehabilitation centre, Netherlands
Participants	Stroke patients with clinically diagnosed symptomatic stroke (first or recurrent, ischaemic, or intracerebral haemorrhagic lesion) admitted consecutively. Patients had to be aged between 18 and 85 years and had to have sufficient comprehension and communication skills. Patients were not included in cases of interfering psychiatric disorders or substance abuse, when they were physically or mentally unable to participate, or when expected discharge was < 3 weeks Age, median, years (IQR): intervention = 59.31 (14.45), control = 61.48 (13.37) Sex (men): intervention = 74%, control = 69% Hemisphere damaged: lesion side, %: intervention = left 21, right 77, bilateral 3; control = left 21, right 73, bilateral 6 Time since stroke, median days: intervention = 41.50 (39.00), control = 37.00 (37.00)
Interventions	Prism adaptation or plain lenses. PA procedure was adapted from Rossetti et al. Patients wore a pair of goggles fitted with wide‐field point‐to‐point prismatic lenses, inducing an ipsilesional optical shift of 10° (PA) or goggles with plain lenses (SA). Exposure consisted of ±100 fast‐pointing movements to 3 stimuli (red, yellow, blue) presented on a horizontal axis at a distance of ±65 cm. Left and right stimuli were located 10° away from the body midline. The investigator indicated which stimulus was the target. A board was held under the chin to prevent viewing of the hand at its starting position but allowing an unobstructed view of targets and terminal errors. The co‐ordinates of the touch responses were recorded
Outcomes	CBS Dynamic navigation task Static cancellation task All taken immediately post intervention with no longer‐term follow‐up
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	"Before the start of the study, the investigator put 70 printed cards with the treatment condition (35 PA and 35 SA) in envelopes. After completion of the baseline assessment, the investigator opened an envelope and allocated the patient based on the treatment written on the card." Insufficient detail on randomisation and allocation
Allocation concealment (selection bias)	Unclear risk	See above
Blinding of participants	Low risk	Patients could not be (completely) blinded to treatment because they had to wear the goggles. However, patients were not explicitly told which treatment they had received, and none of them expressed any awareness of assigned condition (after informal enquiry)
Blinding of personnel	High risk	Nurses, physical therapists, and occupational therapists who filled in the CBS were blinded to treatment conditions (low risk of bias) The investigator (AFTB) who treated and tested patients regarding secondary outcomes was not blinded to treatment because she had to put on the goggles (high risk of bias)
Blinding of outcome assessment (detection bias) All outcomes	High risk	As above
Incomplete outcome data (attrition bias) All outcomes	High risk	All 4 dropouts in intervention arm
Selective reporting (reporting bias)	Low risk	Outcomes reported equally
Other bias	Low risk	Nothing obvious