Katz 2005.

Study characteristics
Methods	Parallel control trial Setting: rehabilitation hospital, Israel.
Participants	19 patients with first right hemispheric stroke and persistent USN VR group = 11, control group = 8 Mean age, years: VR group = 62, control = 63 Sex (men/women): VR = 7/4, control = 5/3 Hemisphere: right Mean days between stroke and treatment: VR = 48, control = 36
Interventions	A street crossing virtual environment was programmed via Superscape’s 3D Webmaster and run on a desktop computer, with successively graded levels of difficulty that provide users with an opportunity to decide when it is safe to cross a virtual street. The level of difficulty was graded by the number and velocity of cars that approach the pedestrian cross‐walk, as well as the side (right or left) from which they approach, thus increasing attentional demands on the user. In addition other destructors were included such as commercial signs, blinking lights, etc. Virtual reality training protocol continued for 4 weeks, with 3 sessions per week, each of 45 minutes' duration, for a total of 9 hours. Timing of the control group computer scanning training protocol was identical. ?? Is this a control group or another active intervention?
Outcomes	Star cancellation from the Behavioral Inattention Test (BIT) . Mesulam Symbol Cancellation test ADL checklist FIM total 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)	High risk	First 8 participants were randomly assigned to either VR training group (experimental) or computer visual scanning tasks group (control); remaining 3 participants were assigned to VR group to increase the number of participants who experienced the experimental condition
Allocation concealment (selection bias)	High risk	See above
Blinding of participants	High risk	No information about blinding provided
Blinding of personnel	High risk	No information about blinding provided
Blinding of outcome assessment (detection bias) All outcomes	High risk	No information about blinding provided
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Unclear
Selective reporting (reporting bias)	Unclear risk	Unclear
Other bias	High risk	Multiple outcomes, none primary; no multiple testing adjustment. 1‐tailed test not justified