Rowe 2010.

Study characteristics
Methods	Substitutive intervention versus compensative intervention versus control, placebo or no intervention Design: "Randomized controlled, multicentre pilot trial" Stratification: "Randomization lists were generated using block randomization stratified by centre and degree of hemianopia (partial or complete) with treatment allocation ratio of 1:1:1." Randomisation sequence: "Participants were individually randomized to one of three treatment groups using a secure (24‐hour) web‐based randomization programme." Comparisons: 3 groups: Group 1 ‐ Fresnel prisms; Group 2 ‐ visual search training; Group 3 ‐ control (standard care/information only) Allocation concealment: treatment allocation was disclosed to the patient by the treating clinician: "The local PI (orthoptist) obtained the treatment allocation and subsequently assigned the participant to the treatment arm." Blinding: outcome assessors for visual field assessment and reading speed were blind to treatment allocation. "due to the nature of the intervention, it is not possible to blind other study personnel or the participant." Power calculation: no ‐ pilot trial; designed to enable sample size calculation for future trials Intention‐to‐treat analysis: "Outcome data were analysed according to the intention‐to‐treat principle." Other recruitment details: "Participants were recruited from stroke units based in 15 United Kingdom (UK) National Health Service (NHS) trusts. Potentially eligible participants were identified by stroke research nurses, and screened for inclusion by a local principal investigator (a qualified orthoptist registered with the Health and Care Professions Council, UK). Participants eligible for inclusion, and providing consent, attended for a baseline assessment, which included assessment and documentation of patient demographics, visual signs and symptoms, visual acuity measures, any additional ocular problems, comorbidity, severity of stroke and level of disability". Patient and public involvement: protocol stated: "This trial has involved a stroke survivor directly in the development of this protocol (JR) and will liase closely with her for advice and direction throughout the conduct of the trial and in the dissemination process. Involvement of stroke survivors in oversight committees is also planned for this trial".
Participants	Total study population: 87 participants randomised. 71 participants at 26‐week follow‐up Withdrawals: 2 "complete withdrawal" ("patients withdrawn from all data analysis and follow‐up"). 9 "partial withdrawal" ("patients withdrawn from follow‐up"). 5 loss to follow‐up Method of diagnosing VFD: "Stable homonymous hemianopia (partial or complete) induced by recent stroke, defined following WHO guidelines". "The visual field assessment will be conducted by a qualified Orthoptist at baseline and at the 6‐week, 12‐week and 26‐week follow‐up visits. An Esterman strategy is to be used for quantitative visual field assessment. This can be performed using either: The Esterman programme on Humphrey or Octopus perimetry; The III4e target on Goldmann with additional checks of static points in the central visual field." Characteristics of population: participant details are listed in Table 6. Type and severity of visual problems: participant details are listed in Table 7. Inclusion criteria: "a. 18 years of age or older; b. best corrected visual acuity of 0.5 or better in each eye at distance; c. stable homonymous hemianopia (partial or complete) induced by recent stroke, defined following WHO guidelines, present over 2 weeks (to exclude rapid recovery cases) but less than 26 weeks prior to randomization; d. refractive error within ± 5 dioptres; e. willing and able to give consent for the study; f. prior to stroke able to read and understand English." Exclusion criteria: "a. unable to consent due to severe cognitive impairment; b. assessed to have ocular motility impairment and/or visual inattention in addition to the visual field impairment; or c. had pre‐existent visual field impairment due to previous stroke." Baseline comparison of treatment groups: "There were no notable differences at baseline between three arms." This study included patients with visual field defects only (no visual neglect). Method of diagnosing visual perceptual problems: "as assessed by the orthoptist"
Interventions	Group 1: prisms (n = 27) Intervention: Fresnel prisms (40 prism dioptre strength) Intervention type: substitution Materials: "Participants were assessed and given sector Fresnel prisms of 40 prism dioptre strength on their glasses (or plain glasses if not already worn). Separate prism segments were used as a mechanical displacement to expand the upper and lower quadrants." Where can materials be accessed? In the UK: NHS supplies for Fresnel prisms. Procedures: "The participant will be instructed to maintain central fixation through their glasses. They will then be instructed to use head movements to explore their field to the affected side when they become aware of an object of interest through the prism. The first prism will be placed at the participant’s first visit; if possible the second prism will also be fitted at this time. However, if the participant is not comfortable with both prisms being fitted at once, the second prism can be placed at a second visit (2 weeks later, ± 1 week) if no adaptation difficulties to the first prism have occurred. If adaption difficulties have occurred the patient can continue with only the first prism and this will be captured on the case report forms." Provided by: orthoptist. Delivery: face‐to‐face, clinic location (patients could be in or outpatients). (Table 5) Regimen: "The prisms should be worn for a minimum of 2 h daily from prism affixation until 6‐week follow‐up visit as a minimum, after this the patient can elect to continue treatment if they wish." Tailoring: no. Modification: no. Adherence: "There were 73 protocol deviations in 58 patients (68.2% overall: 77% in the Fresnel prisms arm, 93% in the visual search arm and 34.5% in the standard care arm). The majority of deviations (n = 41, 56.2%) related to lack of compliance in the intervention arms (e.g. prism not worn a minimum of 2 hours daily for 6 weeks or visual exercises not carried out for 30 minutes daily for 6 weeks). Compliance level was similar across the intervention arms. Patients in the Fresnel prisms arm wore the prisms during 27 days on average." Group 2. scanning training (n = 30) Intervention: "visual search training" Intervention type: compensatory (scanning) training Materials: "Comprised an A4 landscape card with horizontal and diagonal numbered circles radiating out from a central fixation target." Where can materials be accessed? From author. Procedures: "The participant will be instructed to hold this at a distance of 8 inches from their eyes (to ensure a wide field of vision is utilised), glasses can be worn as required. Participants will be asked to transfer gaze quickly between printed targets on the A4 card. The targets are printed off centre to the right and left sides along the horizontal as well as oblique planes to ensure stimulation of a wide area in the blind and seeing parts of the visual field." Provided by: instructions provided by orthoptist. Training carried out at home by patients. "Participants will be instructed on the scanning exercises following randomisation to ensure their understanding of the procedure of doing this training. In addition, printed instructions will be provided with the visual training target card". Delivery: home (self‐adminstered) (Table 5). Regimen: "Participants will be instructed to continually scan between the various targets for 30 min daily from baseline until their 6‐week follow‐up, after which they can elect to continue treatment if they wish." Tailoring: no. Modification: no. Adherence: "There were 73 protocol deviations in 58 patients (68.2% overall: 77% in the Fresnel prisms arm, 93% in the visual search arm and 34.5% in the standard care arm). The majority of deviations (n = 41, 56.2%) related to lack of compliance in the intervention arms (e.g. prism not worn a minimum of 2 hours daily for 6 weeks or visual exercises not carried out for 30 minutes daily for 6 weeks). Compliance level was similar across the intervention arms ... patients in the visual search training arm followed the visual search exercises 28 days on average." Group 3: standard care (n = 30) Intervention: advice only ("all three arms will receive the same information leaflets") Intervention type: control Materials: "Participants were given information leaflets from the UK Stroke Association and the UK Royal National Institute for the Blind about visual impairment following stroke." Where can materials be accessed? UK Stroke Association and the UK Royal National Institute for the Blind. Procedures: standard care. Provided by: NA. Delivery: NA (Table 5). Regimen: NA. Tailoring: NA. Modification: NA. Adherence: NA
Outcomes	See Table 8 Visual field assessment Reading ability (Radner reading test) Visual function questionnaire (VFQ 25‐10) Rivermead mobility index Nottingham extended activities of daily living assessment ED‐5Q (a standardised instrument for measuring health outcome) Short Form‐12 (SF‐12) Adverse events Time points when outcomes were assessed: baseline, 6‐week, 12‐week, and 26‐week follow‐up
Notes	Review authors Fiona Rowe and Alex Pollock were involved in this study.
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	"Participants were individually randomized to one of three treatment groups using a secure (24‐hour) web‐based randomization programme. Randomization lists were generated using block randomization stratified by centre and degree of hemianopia (partial or complete) with treatment allocation ratio of 1:1:1. The local PI (orthoptist) obtained the treatment allocation and subsequently assigned the participant to the treatment arm."
Blinding (performance bias and detection bias) All outcomes	Low risk	Participants and clinical personnel unable to be blinded due to the nature of the intervention. Blinded outcome assessors
Incomplete outcome data (attrition bias) All outcomes	Low risk	Incomplete data all accounted for. Intention‐to‐treat analysis (where possible)
Other bias	Low risk	No other cause of bias noted