Adaptation to poststroke visual field loss: A systematic review

Abstract Aim To provide a systematic overview of the factors that influence how a person adapts to visual field loss following stroke. Method A systematic review was undertaken (data search period 1861–2016) inclusive of systematic reviews, randomized controlled trials, controlled trials, cohort studies, observational studies, and case controlled studies. Studies including adult subjects with hemifield visual field loss, which occured as a direct consequence of stroke, were included. Search terms included a range of MESH terms as well as alternative terms relating to stroke, visual field loss, visual functions, visual perception, and adaptation. Articles were selected by two authors independently, and data were extracted by one author, being verified by the second. All included articles were assessed for risk of bias and quality using checklists appropriate to the study design. Results Forty‐seven articles (2,900 participants) were included in the overall review, categorized into two sections. Section one included seventeen studies where the reviewers were able to identify a factor they considered as likely to be important for the process of adaptation to poststroke visual field loss. Section two included thirty studies detailing interventions for visual field loss that the reviewers deemed likely to have an influence on the adaptation process. There were no studies identified which specifically investigated and summarized the factors that influence how a person adapts to visual field loss following stroke. Conclusion There is a substantial amount of evidence that patients can be supported to compensate and adapt to visual field loss following stroke using a range of strategies and methods. However, this systematic review highlights the fact that many unanswered questions in the area of adaptation to visual field loss remain. Further research is required on strategies and methods to improve adaptation to aid clinicians in supporting these patients along their rehabilitation journey.

Reported prevalence of visual field defects following stroke varies widely and, if present, can have negative implications on quality of life and activities of daily living. Hemianopic visual field defects are associated with a reduced prognosis for successful rehabilitation (Patel, Duncan, Lai, & Studenski, 2000;Han, Law-Gibson, & Reding, 2002), especially when combined with visual inattention (Cassidy, Bruce, Lewis, & Gray, 1999;Jehkonen et al., 2000). In addition, the extent of visual field loss will impact on the functional symptoms a patient experiences, hence, influencing the adaption process. For example, a patient with macular splitting hemianopia will experience more difficulty with reading tasks than those without this clinical sign (Trauzettel-Klosinski & Reinhard, 1998). Patients with hemianopic field defects cannot process images in the same way as those with a full visual field. They demonstrate numerous visual refixations and inaccurate saccades which result in impaired scanning, longer search times, and the visual omission of relevant objects (Zihl, 1995a). Visual inattention, otherwise known as visual neglect, can coexist with visual field loss, particularly in strokes located on the right side of the brain (Gottlieb & Miesner, 2004). If field loss is combined with visual inattention, a person typically does not automatically scan or track to the affected side, making adaptation more problematic and less likely to occur.
Treatment for visual field loss is inconsistent and not commonplace, even in stroke units where orthoptic services are provided.
There are three main approaches to rehabilitation of visual impairment: adaptation/compensatory, substitution, or restitution as discussed in a 2011 Cochrane review (Pollock et al., 2011). This review concluded that compensatory training was a more favorable option.
Such treatment may potentially increase speed of adaptation to the visual loss, but more research is needed in this area. Visual search training usually involves patients practicing identifying objects in their hemianopic and intact hemifields, improving their detection performance over a period of time. There is accumulating evidence that patients can improve their scanning performance with visual search training; however, it is unclear to what extent this training is transferable to everyday life skills, such as obstacle avoidance and increased hazard perception.
In real-life settings, some people adapt remarkably well to their visual field loss and within weeks of their stroke can read easily, negotiate new surroundings, and appear to have little detriment to their everyday activities, despite having no recovery of their visual field loss. A further group of people appear to be more affected by this loss of vision, struggling with everyday tasks such as reading, mobility, and location of objects around them. The authors have an interest in this specific area as it has been noticed in the clinical setting that there is a wide variation in the way people adapt to their visual field loss. We do not fully understand why some people adapt at a different rate to others. Those who adapt well have a noticeably improved quality of life over those who do not. If we can understand this process in more depth, this allows the potential for clinicians to influence this change in behavior and better support the patients' adaptation processes. This review aims to investigate current knowledge into the mechanism of adaptation to visual field loss, the factors that influence how a person adapts to visual field loss and the interventions that are available to aid the adaptation process specifically.
We aim to use the systematic review as a starting point for a clinical study to explore the factors that influence the adaptation process in more detail. The findings of the review and clinical study together will be related back to clinical practice, allowing clinicians to target interventions effectively to insure people adapt as quickly and efficiently as possible to visual field loss following a stroke. This review differs from others in the related topic area due to its specific focus on adaptation and the interventions that focus on assisting this process. This is not a full review of the interventions for visual field loss as this has been covered elsewhere (Pollock et al., 2011;Hanna & Rowe, 2017). Similarly, the review will not include restorative rehabilitation or recovery of visual field as this is outside the review objectives.

| ME THODS
We conducted a full systematic review of the literature dating from the start of recorded databases for each information source to April 2016, aiming to collect all evidence relating to adaptation to poststroke visual field loss. A detailed protocol was developed prior to the review and registered with PROSPERO (Shamseer et al., 2015).
By the term adaptation, we mean the process whereby people evolve and change behaviors, despite no change in their circumstances, in this instance, an unchanged defect in their visual field. This is different to recovery of visual field, whereby there is a physical change to the area of peripheral vision. We therefore define adaptation in this context to be a persons' behavioral and practical responses to the visual field loss over time. Adaptation may be a fully conscious reaction such as a person making attempts to move their head more frequently or increase their scanning eye movements or could indeed be factors out of conscious control such as a person's previous visual scanning experiences. This review does not specifically include the process of coping, or a person's emotional response to their visual field deficit. Coping is defined as a person's ability to effectively deal with something difficult, to minimize stress. Coping tends to be a short-term strategy that is prompted by a lack of alternatives, whereas adaptation involves more sustained planning and focuses on finding alternative ways of handling a task. The terms "adaptation" and "coping" are often used interchangeably, but for the context of this review, the focus is adaptation, making changes to deal with the situation, as oppose to coping or accepting things the way they are.
In general, people adapt to change by forming new expectations that lead to an ability to deal with the new conditions. To adapt to a change in visual status, a person needs to be able to accept the situation and then deal with the implications of this as well as make physical changes and develop strategies to allow them to adapt.

| Types of studies
The following types of studies were included: systematic reviews, randomized controlled trials, controlled trials, prospective and retrospective cohort studies, observational studies, and case controlled studies. Case reports, editorials, and letters were excluded. All languages were included, and translations obtained when necessary.

| Participants
We included studies reporting on subjects over the age of 18 years only, as children are likely to have different adaptation mechanisms.
Studies including subjects with hemifield visual field loss of any severity, which occured as a direct consequence of stroke, were included. Studies reporting on mixed populations were only included if 50% or more of subjects had a diagnosis of stroke and data were available within this category.

| Information sources and search strategy
We utilized systematic strategies to search key electronic databases and contacted known experts in the field. We used a range of search strategies as outlined below:

| Search terms
Search terms (Table 1) included a range of MESH terms as well as alternative terms relating to stroke, visual field loss, visual functions, visual perception, and adaptation. Due to the specific target area for this review, it was necessary to include search terms for factors that have the potential to influence the adaptation process. These search terms were identified and discussed by a group of stroke survivors who themselves had personal experience of adapting to visual field loss following stroke. The authors were aware that using the term "adaptation" alone would elicit few results, so search terms were included such as driving, reading, saccades, hazard perception, and visual tracking, to encompass the factors considered important for the adaptation process.

| Selection process
The titles and abstracts identified from the search were independently screened by the two authors (CH, FR) through each phase of the review (screening, eligibility, and inclusion) using the prestated inclusion criteria. Where further information was required for this process, the full paper was obtained and the selection criteria applied. A subsequent review of the full papers was undertaken to determine which studies should be included (CH, FR). In the case of disagreement between authors for inclusion, an option was available to seek the opinion of a third reviewer, however, this option was not required in practice as no disagreements occurred.

| Data extraction for included studies
A predesigned form was used for the data extraction process. The data extraction form encompassed all the factors identified by stroke survivors as having potential importance for the adaptation process: extent of visual field loss; site of brain lesion; age; gender; ethnicity; handedness; cognition; anxiety levels; social deprivation; preexisting ocular conditions; general signs and symptoms as well as ocular signs and symptoms. Data were extracted by one reviewer (CH) and verified for completeness and accuracy by another (FR).

| Quality assessment
One reviewer (CH) reviewed the quality of included studies using the following four checklists; this was subsequently verified by the second reviewer (FR). The term "quality" refers to: "the degree to which a study employs measures to minimize bias and errors in its design, conduct, and analysis" (Khan, Kunz, Kleijnen, & Antes, 2003 Checklists were adapted to insure they only included information considered important to appraise quality of the included studies.
Checklist items excluded were not considered by the reviewers as relevant to the appraisal process; for example, title, background, funding, and setting.

| RE SULTS
Results of the search are outlined in Figure 1. As expected, there were no identified studies which explored the factors that influence how a person adapts to visual field loss following stroke in a precise and systematic manner. In other words, no one article has explored and discussed all of the factors important for the adaptation process over time to answer this question fully. However, there were seventeen articles identified by the reviewers as containing a factor considered likely to be important for the process of adaptation to poststroke visual field loss. These were articles that contained information on the factors considered as potentially important for the adaptation process by the group of stroke survivors themselves. These articles, covering factors such as age, environment, compensation strategies, and awareness of symptoms, are discussed as a group. There were thirty additional studies identified that focused on the interventions for visual field loss that were deemed directly related to the factors above. Only articles that focused on adaptation factors or interventions likely to influence these were included, making this review distinct from other intervention reviews. In summary, a total of 47 articles (2,900 participants) were included in the overall review, divided into two sections for reporting: • Studies where the reviewers could identify a factor they consider is likely to be important for the process of adaptation to poststroke visual field loss.
• Studies that detailed interventions relating to the above factors.
Due to the variations and diversity across trials, with respect to reporting of outcomes as well as recruitment and selection of subjects, a meta-analysis of studies was not undertaken. A narrative summary of the data is presented in relation to included studies.

| Factors that have the potential to affect adaptation
There were seventeen articles (1,423 participants-with 809 of these having poststroke visual field loss) included in this section of the review. Table 2 summarizes the key data extracted from the studies including the proposed link to adaptation as deemed by the reviewers and stroke survivors. The seventeen articles included studies consisting of one randomized controlled trial, eleven cohort studies, and five prospective observational studies.

| Quality assessment
The quality of evidence was assessed for each of the 17 articles included in this section (Supporting Information Tables S1-S3 TA B L E 2 (Continued) Poland, & Stephenson, 2012), respectively. All articles were included in the review with quality of evidence taken into consideration in the discussion.
The factors extracted as likely to be important for the adaptation process covered five different areas: compensation strategies used by an individual; a person's awareness of their hemianopia; presence of inattention; effect of age; and environment surrounding a person during the poststroke period. The articles identified in each section will be discussed individually. There are likely to be a number of further factors important for the adaptation process which will need to be explored in more detail by further research, but we did not elicit any results within the remit of this review, relating specifically to adaptation. These factors include previous visual experiences, occupation of the patient prestroke, site of the brain lesion and perhaps most importantly, the extent of visual field loss. It is feasible to suggest that someone with a more extensive visual field loss will adapt in a different manner to someone with a field loss of a lesser extent. No articles reported on the direct relationship between extent of visual field loss and/or the presence of macular sparing and their importance in the process of adaptation.
This is a noted limitation of this review in that the authors cannot comment on an association between extent of field loss and its importance in the adaptation process; this factor warrants further exploration. Several of the included articles detailed the extent of visual field loss in their patients (Hardiess, Papageorgiou, Schiefer, & Mallot, 2010;Bergsma, Leenders, Verster, van der Wildt, & van den Berg, 2011;Kasneci et al., 2014), but this extent of loss was not related to adaptation in any way. One study by Trauzettel-Klosinski and Reinhard (1998) reported that the presence or absence of macular sparing influenced factors such as fixation behavior and reading performance. They found that the lesser the extent of macular sparing, the less stable the fixation. This finding is likely to influence the process of adaptation, in particular, when considering adaptive strategies such as eccentric fixation and predictive saccadic eye movements.

| Compensation strategies
Ten of the included studies discuss the impact of compensation strategies for poststroke visual field loss including use of eye movements, visual search, head movements, spatial localization, and scanning behavior. Compensation strategies may include the use of head and shoulder movements to aid tasks such as searching, obstacle avoidance and hazard perception, scanning the environment, and/ or saccadic eye movements/eccentric fixation for the purpose of improving close tasks such as reading.
Reading/saccadic adaptation-whereas visual acuity testing demands recognition of one optotype at a time, reading demands a more complex simultaneous overview of a group of letters. Patients with hemianopic visual field defects develop compensation strategies to aid reading ability using eccentric fixation and scanning eye movements. Eccentric fixation may help some patients with macular splitting and training in the strategies required for reading can help a patient to adapt to their loss. Eccentric fixation shifts the visual field deficit toward the affected side, creating a small useful visual field area along the vertical meridian. This adaptive strategy benefits a persons' ability to adapt, particularly for reading tasks (Trauzettel-Klosinski & Reinhard, 1998). A study by Meienberg, Zangemeister, Rosenberg, Hoyt, and Stark (1981) reported patients as developing compensatory search strategies to overcome difficulties with locating objects. To fixate targets in the seeing hemifield, subjects in this study were shown to undershoot the target to prevent losing it in the blind field, then hold it off the fovea on the seeing side of the macula. This is considered a useful strategy for improving reading ability in this group of patients. Meienberg et al. (1981) also discussed the difference between short-and long-term adaptation. In the short term, patients with hemianopia develop a staircase strategy to search for a target, whereas in the longer term, they employ a more efficient strategy of one large saccade to overshoot the target. In homonymous hemianopia with macular splitting, severe reading problems result from a loss of half of the reading visual field. This longer term adaptation was further highlighted in a study by Reinhard, Damm, Ivanov, and Trauzettel-Klosinski (2014) who found that patients with hemianopia showed significantly more dysmetric saccades to the blind side compared to the seeing side. The number of dysmetric saccades, however, did not correlate with duration of hemianopia, indicating insufficient spontaneous long-term adaptation in the patients.
Although a considerable amount of research has focused on hemianopic reading difficulties or hemianopic alexia and a persons' ability to compensate for this, we still do not fully understand why some people adapt to this reading difficultly more effectively than others.
Patients with hemianopia are reported to employ reading strategies that are inefficient and slower than those with a full visual field.
The severity of the reading problem is also influenced by the side of the defect, in relation to the direction of reading. In left to right readers (as in the English language), a right hemianopia significantly impairs reading as the person cannot see the oncoming groups of letters or words (Trauzettel-Klosinski & Reinhard, 1998). A left-sided hemianopia causes problems locating the start of a line of text, meaning they tend to skip lines or restart the same line twice. Those with right-sided hemianopia show prolonged search durations, prolonged fixation times, reduced amplitudes of saccades to the right, and multiple regression saccades (Machner et al., 2009;Zihl, 1995b).
Patients with a right sided visual field loss tend to lose the word they are fixing on, requiring a refixation toward the word in view. This refixation slows their reading time considerably (McDonald, Spitsyna, Shillcock, Wise, & Leff, 2006). How a person compensates for their reading difficulties and uses their visual scanning techniques is likely to impact on the overall process of adaptation, but again, this direct link has not been explored.
Search tasks-When a person experiences a loss of visual field they learn over time to compensate for their visual difficulties by improving the accuracy and speed of eye movements to the defective side. The development of adaptive eye movement strategies over time has been well documented, and the implication of these compensation strategies is that subjects develop a more effective visual search technique, for a variety of tasks such as obstacle avoidance and driving (Hardiess et al., 2010;Meienberg et al., 1981;Machner et al., 2009;McDonald et al., 2006;Pambakian et al., 2000;Papageorgiou, Hardiess, Mallot, & Schiefer, 2012;Zangemeister & Oechsner, 1996;Wood et al., 2011;Roth et al., 2009). This improved visual search is likely to be an important factor in the adaptation process, and this theory has been explored by Roth et al. (2009)  Some individual studies have focused on one specific aspect of everyday functionality concerning compensation strategies, but no one study has compiled the factors likely to be important for adaptation together into one study. A pilot study by Taylor et al. (2012) provides preliminary information regarding the development of head and shoulder movement strategies as a compensation mechanism following visual field loss. They suggest that head and shoulder movements could be an important factor for the compensation process. This theory needs investigation, with further research warranted in this area.

| Awareness of hemianopia/presence of symptoms
Three studies provided information regarding a lack of awareness of hemianopia. The authors of this review feel this has the potential to be an important factor for the adaptation process, as a lack of awareness could affect a persons' ability to adapt and compensate. In a prospective study of patients with homonymous visual field defects (Celesia, Brigell, & Vaphiades, 1997), 62% of patients were found to have hemianopic anosognosia, defined as the unawareness of visual loss in the homonymous hemifield. Celesia et al. (1997) suggest that this anosognosia is most often related to a failure of discovery of the deficit, occasionally due to severe visual hemineglect, a generalized cognitive impairment or a combination of these factors. A further study of anosognosia for visual field defects reports a lower incidence of 19% of patients failing to recognize their defect (Baier et al., 2015). A multicentre cohort study by Rowe and the VIS group (Rowe et al., 2013) supported this finding and reported 16% of their 479 patients with a visual field loss as not complaining of visual field loss specifically. In this cohort of patients, 10.6% of those with visual field loss were reported as not complaining of any visual symptoms of any type.

| Presence of inattention
Although no studies were identified in highlighting inattention as a factor influencing the adaptation process, one paper by Cassidy et al. (1999) reports on the reduced prognosis for patients presenting with inattention in combination with hemianopia. They report on the presence of visual field defects being associated with a more severe form of visuospatial neglect in the first week after stroke, than those without visual field loss. This fact has potential to influence the adaptation process, particularly in the early poststroke stages.

| Effect of age
Two studies provide observations around the effect of age on compensation strategies. Older age is generally considered to have an adverse effect on functional outcome following acquired brain injury; therefore, older age is considered likely to be a factor that has potential to influence the adaptation process to poststroke visual field loss. Schuett and Zihl (2013) report findings from their study to determine the effect of age on reading and visual exploration impairments, following compensatory oculomotor treatment. They report that older patients achieve the same post treatment improvements in reading and visual exploration as younger patients, concluding that age does not appear to be a critical factor for functional outcome when focusing on compensatory treatments of visual field defects. These findings suggest that older age is not necessarily associated with a reduced level of adaptation. However, a study by Tant, Cornelissen, Kooijman, and Brouwer (2002) compared scanning performance for healthy subjects on two different occasions, comparing subjects' own normal performance with their own performance when a hemianopia was simulated. They observed age-related processes in compensating for the simulated hemianopia. During eye movements recordings, they report a reduced level of compensation to visually elicited disabilities, in the older age ranges. Tant et al. (2002) suggest that disabilities in scanning performance are more pronounced in an older age group, suggesting a possible reason for this as differences in important factors for the compensation process (such as perceptual and intellectual abilities) which tend to decrease with age. There are limitations of this study in that the hemianopic visual field defects assessed were simulated and not true defects caused by brain injury, but the authors of this review feel it warrants a mention as having potential significance for adaptation. Tant et al.
compared scanning performance in the simulated hemianopia individuals, the same individuals without the simulated hemianopia, and real hemianopia patients. They reported clear parallels between simulated and real hemianopia, suggesting that hemianopic scanning behavior is elicited by the visual field defect and not by the additional brain impairment. The relationship between age and adaptation requires future exploration if all aspects of the adaptation process are to be considered.

| Environment
One study (Loverro & Reding, 1988) detailed the effect of environment, more specifically bed orientation on the outcome for hemianopic patients. Loverro and Reding (1988) found no relationship between bed positioning and rehabilitation outcome in patients with poststroke homonymous hemianopia or visual inattention. In this study, patients with hemianopia or inattention were randomized to have their impaired or unimpaired side directed toward the side of stimulating environment. This article was considered as low quality during the quality assessment process (48%) and the topic of bed positioning and environment is an area that lacks evidence and should be the focus of future research. The authors of this review consider that environment and side of stimulation have the potential to be an important factor in the adaptation to poststroke hemianopia.

| Interventions that may influence adaptation
Included in this section were thirty studies (1,477 participants-with 1,411 of these having poststroke visual field loss). This number includes one Cochrane review relating to interventions available for visual field loss following a stroke (Pollock et al., 2011). In view of the rigorous methods employed for Cochrane reviews, the findings have been summarized in this review, followed by a narrative overview of additional articles not included in the Cochrane review. Table 3

| Quality assessment
The quality of evidence was assessed for each of the 30 articles included in this section (Supporting Information Tables S4-S7). In summary, two articles scored 100% for quality of evidence presentation in the opinion of the reviewers (Gall & Sabel, 2012;Ong et al., 2012). Twenty five articles scored between 75% and 99% and therefore deemed as good quality evidence. Three scored between 50% and 74% on the relevant quality checklists.
Interventions for visual field defects are proposed to work in multiple ways, as detailed by a Cochrane review of such interventions (Pollock et al., 2011 , 1977, 1979). The Cochrane authors concluded there is some limited evidence to support the use of compensatory scanning therapy to improve scanning and reading outcomes in this patient group.
At the time of review publication, there was not sufficient evidence to support the impact of this compensatory scanning therapy on functional activities undertaken by the stroke survivor. In addition, there was insufficient evidence to reach conclusions regarding the benefits of visual restitution training (VRT) or prisms for this cohort of patients. Over 3 weeks Multimodel audiovisual exploration training appears to be more effective than exploration training alone.

| Compensatory treatment
Kerkhoff et al. Marshall et al. Note. Articles taken from Cochrane review (n = 9) included for information only and are not included in this adaptation review.

TA B L E 3 (Continued)
with well-designed randomized control trials, to explore the effects on visual field loss, in both the areas of recovery and adaptation.
The majority of studies relating to compensatory treatments are concerned with the improvement of eye movements and scanning into the affected field (Bergsma et al., 2011;Roth et al., 2009;Aimola et al., 2014;Hazelton, Pollock, Walsh, & Brady, 2015

| Substitutive treatment
Substitution interventions involve adaptation to visual field loss using optic devices, mechanical aids, or modifications to the immediate environment. Studies included in this review concerning substitutive treatments describe the use of prisms for hemianopia (Bowers, Keeney, & Peli, 2014;Giorgi, Woods, & Peli, 2009

| Restitution treatment
Restitutive interventions include those where there is direct training or repetitive stimulation of the impaired visual field (Pollock et al., 2011). Visual restoration therapy (VRT) is one form of restitution treatment that is the most commonly reported in the literature. The aim of VRT is the improvement of visual field loss by stimulating the border along the area of visual field loss; along the boundary between remaining, normal visual field and damaged, impaired visual field. Pollock et al. (2011) conclude that there is insufficient evidence to draw conclusions about the effectiveness of VRT as compared to placebo, control, or no treatment when focused on visual field outcomes. This is further supported by Roth et al. (2009) andReinhard et al. (2005) who examined whether VRT has the potential to change absolute hemianopic field defects, reporting none of their seventeen patients to have an explicit change in defect after training. The latter study was not included in the review as its focus was on recovery of visual field following VRT and not adaptation. In trials where eye movement recording was not undertaken, improvement in visual field due to eye movements cannot be excluded (Reinhard et al., 2005;Schmielau & Wong, 2007). However, studies where eye movements were measured did confirm visual field recovery, arguing against the hypothesis that compensatory eye movements alone can explain vision recovery (Gall et al., 2016;Kasten, Bunzenthal, & Sabel, 2006).
There is significant variation in the treatment dose, duration, and field outcome for these studies.
Although the aim of VRT is restitution and not adaptation specifically, the practices of VRT are reported to affect quality of life measures (Gall & Sabel, 2012), hence, having potential to influence the adaptation process. The effect of VRT on the absolute visual field defect is outside the aims and objectives of this review.

| CON CLUS ION
There is substantial evidence that patients can be supported to compensate and adapt to visual field loss following stroke using a range of strategies and methods. However, this systematic review highlights the fact that many unanswered questions remain: what does adaptation to visual field loss mean to the patient, carer, and clinician? How can adaptation be measured over time? Why do some people adapt more effectively and at a quicker rate than others, despite seem- It is important to note that some studies in this review observed a mixed caseload and therefore did not focus on a specific stroke survivor population. However, in the authors' opinion, the cause of visual field defect is not likely to be a crucial factor for the adaptation process, but indeed a range of other factors will show a greater influence. Future research should consider the factors that could be important for the adaptation process, seeking views of stroke survivors themselves and their families/carers to identify aspects they feel are important for their own adaptation journey, as well as clinicians responsible for the rehabilitation of this population group.
As clinicians working with this group of patients, we are expected to make a clinical judgment on whether a person has adapted to their loss of peripheral vision. This is particularly true for a situation where a person wants to consider a return to driving with a hemianopia under the exceptional cases rule for visual field loss. One of the Driving and Vehicle Licensing Agency (DVLA) (DVLA, 2018) requirements for consideration for the exceptional cases ruling to return to driving despite having a significant visual field loss is "clinical confirmation of full functional adaptation" to the visual field loss. There is currently no guidance on what this actually means or how clinicians can test for this, creating inconsistent approaches for patients and inconsistent care and decision making regarding referral of patients for driving assessment. This is an area that must be addressed in the interest of equality for those with visual impairment.
It is vital that the factors important for adaptation be identified to allow clinicians to recognize which people are likely to have difficulty adapting and target interventions specifically within these areas, as well as to develop methods for assessing adaptation and monitoring change over time.

ACK N OWLED G M ENTS
The authors would like to thank the VSURP (Vision and Stroke User Reference Panel) group of stroke survivors who were instrumental in the planning and development of this review and subsequent research. They would also like to acknowledge the help and support of colleagues within their research team Dr Lauren Hepworth and Ms Kerry Hanna for support throughout the review process.

CO N FLI C T O F I NTE R E S T
The authors report no conflicts in relation to this review. Claire Howard and Fiona Rowe are funded by National Institute for Health Research Fellowships to carry out research, which includes this systematic review. The views expressed are those of the authors and not necessarily those of the NHS, the NIHR, or the Department of Health.