Abstract
Return to driving is a valued activity among people who experience stroke. Health care providers, including occupational therapists, require evidence-based tools for driver screening post-stroke, validated for stroke with representation of diverse demographic subgroups. To identify tests supported in the literature predictive of fitness to drive after stroke and critically appraise the representativeness of extant research across demographic subgroups. A systematic literature review was conducted to address the objectives. Consistent with prior research, the Stroke Driver’s Screening Assessment and Trail Making Test-B were the most predictive of driver fitness. However, research has consistently underrepresented women, people younger than 55 years of age, and people from low-income countries. Further research is needed with (a) more detailed reporting of participant demographics and (b) increased representation of demographic subgroups within samples, to support culturally informed driver screening practices following stroke.
Plain Language Summary
Returning to driving is very important to people who have experienced a stroke. As a stroke can cause functional changes that make driving difficult, health care providers, including occupational therapists, must use tests to check fitness to drive. This review examined which tests are predictive of fitness to drive, for which subgroups of people who have had a stroke. Ideally, for a test to be appropriate to use with a patient, the test should be previously tested with study participants similar to the patient, for example, with similar age or education level. This review found that the Stroke Driver’s Screening Assessment and Trail Making Test-B are good predictors of fitness to drive after stroke. However, study samples in this research often did not include many women, people younger than 55 years of age, and people from low-income countries. More research is needed that includes people who are not represented in the studies to date.
Introduction
Return to driving is an important priority among people who have experienced a stroke (Griffen et al., 2009; Patomella et al., 2009; White et al., 2012). Within a Canadian context, best practice guidelines stipulate all persons who have experienced a stroke who wish to return to driving should be screened for functional impairments using valid and reliable assessments to inform recommendations (Mountain et al., 2020). In the United States, most states have similar requirements prior to resumption of driving following a stroke (Winstein et al., 2016). Australia and New Zealand also have robust best practice guidelines for return to driving after stroke (2025). Similarly, many countries around the globe are developing or calling for the development of best practices and more research regarding resumption of driving after stroke, including within India (Gupta et al., 2024), Saudi Arabia (Almosallam et al., 2022), Chile (Riesco et al., 2018), and South Africa (Bryer et al., 2010). Occupational therapists are ideally positioned to evaluate fitness to drive and inform recommendations for return to driving after a stroke (World Federation of Occupational Therapists, 2019). However, therapists need access to appropriate screening tools to inform recommendations regarding fitness to drive post-stroke (Cammarata et al., 2017).
Previous systematic reviews have identified screening tools most predictive of fitness to drive among stroke populations (Devos et al., 2011; Hanna et al., 2017; Hird et al., 2014; Murie-Fernandez et al., 2014). However, the most recent review was published 7 years ago and focused on screening for visual-perceptual impairments only (Hanna et al., 2017). Furthermore, existing systematic reviews examined screening tools validated for stroke as a diagnostic group and the extent to which historically underrepresented subgroups are included in the extant literature is unknown. As return to driving is of high importance to patients, it is a priority to investigate which demographic subgroups are represented in the literature to adequately provide equity, diversity, and inclusion informed care. As such, the need for an updated systematic review that critically appraises the representation of different demographic subgroups of individuals who have experienced a stroke within the existing literature is warranted.
In stroke research more broadly, the Heart & Stroke Association (2023) released a report highlighting system-level changes required to address the gaps in stroke research and health care for specific demographic subgroups, particularly for women. Moreover, although the overall incidence of stroke is decreasing in high-income countries, stroke incidence is increasing for particular subgroups including women, young people, and people from low-income countries (Ekker et al., 2019; Leppert et al., 2022; Scott et al., 2022), the same subgroups of people often neglected in existing stroke research (Heart & Stroke Association, 2023). As screening tools are ideally used with the specific subgroup population for which they have been validated, it is reasonable to question the validity of currently recommended tools for subgroups who may not have been included in existing validation studies (Fujii, 2018).
Within the field of neuropsychology, where many tests for driver screening were developed, there are calls for demographically appropriate tests, normative data, and testing procedures to enhance the validity of assessment (Franzen et al., 2021; Fujii, 2018). The ECLECTIC framework was developed to inform culturally appropriate assessment practices. This framework considers (E) education and literacy, (C) culture and acculturation, (L) language, (E) economics, (C) communication, (T) testing situation, comfort and motivation, (I) intelligence conceptualization, and (C) context of immigration, as cultural moderators of test performance and provides strategies to enhance validity of results, and cultural sensitivity (Fujii, 2018). The ECLECTIC framework recommends referencing normative data as specific to the subgroup membership of the patient as possible and using functional tests when available, as functional assessments are less likely to misclassify a person’s real-life abilities (Fujii, 2018). Within occupational therapy, this attention to culture in assessment is echoed in recent practice frameworks that emphasize cultural safety, including interpreting findings within cultural contexts (Restall et al., 2022). To provide equity, diversity, and inclusion informed care, occupational therapists need access to research which report and include diverse demographic subgroups (Suarez-Balcazar et al., 2023).
Understanding subgroup representation is necessary for therapists and drivers to work together to identify the most appropriate tools that can adequately identify underlying abilities. Study participants for fitness to drive research are often sampled from people who present to driving assessment centers for formal driving assessments, (e.g., Chua et al., 2012; Korner-Bitensky et al., 2000; Selander et al., 2010). However, this only represents the group of people identified for further assessment, and such decisions are informed by earlier driver screening (Korner-Bitensky et al., 2010; Sangrar et al., 2018). Furthermore, formal driving assessments are often a costly out-of-pocket expense (Petzold et al., 2010). Thus, recruiting from assessment centers excludes people who were not offered, or cannot afford formal driving assessments.
As such, the purpose of the current research was twofold: (a) to update and extend previous reviews which identified office-based screening tools for predicting fitness to drive following stroke and (b) to critically appraise the representation of different demographic subgroups in extant fitness to drive research following stroke. Specifically, we assessed the representation of subgroups in which stroke incidence is increasing including women, people aged 55 and younger, people residing outside of high-income countries, and members of racialized communities (Ekker et al., 2019; Leppert et al., 2022; Scott et al., 2022). Racialized communities refer to persons who belong to visible, non-White minority groups including South Asian, Chinese, Black, Filipino, Arab, Latin American, Southeast Asian, West Asian, Korean, and Japanese (Statistics Canada, 2021).
Method
Design
The methodology for this systematic literature review followed Cooper et al. (2009). A study protocol was prepared and registered on PROSPERO (ID: (ID: CRD42023469594)), and results were reported per the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRIMSA) guidelines (Page et al., 2021). The completed PRISMA checklist is presented in Supplemental Table A.
Locating and Selecting Studies
To meet the research objectives and after consultation with the Western University Research Librarian, the following six databases were searched: Scopus, Medline (OVID), CINAHL, Embase, Cochrane, and PsychINFO. Searches were completed in October 2023 and updated in January 2025 to include any new publications through to December 2024. The search strategy combined keyword as well as subject heading searches utilizing Boolean operators for the following concepts: (a) drive, driving, driver; (b) fitness to drive, driver fitness; and (c) stroke (including all medical terms for stroke [e.g., cerebral vascular accident] and including all stroke subtypes [e.g., ischemic, hemorrhagic]). The detailed search strategy is presented in Table 1. Inclusion criteria for the review were studies that: employed a pass-fail outcome as the dependent variable for driver evaluation (on-road or via driving simulator as the validation method); specifically assessed people with stroke (or if the study included other diagnoses, outcomes for stroke population were reported separately); and the study population included drivers >18 years who held valid driver’s licenses prior to their stroke. All studies that met the inclusion criteria were included regardless of year of publication.
Search Strategy.
Studies were not excluded based on publication language. Attempts were made to obtain full-text translations for any studies published in languages not spoken fluently within the research team. Authors were contacted directly to inquire about translated copies. Following recommendations from Walpole (2019), when translations were not available, Google Translate was utilized to obtain translations and facilitate data extraction.
Google Scholar was also searched to increase the likelihood of identifying studies published in languages other than English. The “cited by” feature in Google Scholar was used to identify papers citing a seminal systematic review and meta-analysis (Devos et al., 2011). Devos et al. (2011) is cited 191 times by entries in Google Scholar and utilized similar inclusion criteria to the present study. Furthermore, the reference list of every included article was hand searched to identify relevant articles missed in database searches and Google Scholar.
Screening of papers was achieved via a three-step process. First, the Covidence platform was used to remove duplicates from the search output. Next, the references were uploaded to ASReview, an open-source machine learning platform (van de Schoot et al., 2021). ASReview uses active learning technology that adapts to researcher input to sort references according to relevance. As articles are manually identified as eligible or ineligible by the researchers, the list is continually re-sorted to bring most relevant titles to the top. Title and abstract screening was completed using ASReview by two independent researchers (AVV and LJ) for eligibility, and studies were included by consensus. A recent simulation study indicates that 100% of relevant studies are identified in the first 30% of articles sorted by ASReview (van de Schoot et al., 2021). Screening concluded after each independent researcher screened the first 30% of the titles, and when several consecutive titles (a further 10% of all titles found; n = 72) were marked irrelevant, thus screening a total of 40% of the titles. Disagreement involved a third researcher (LA) to determine eligibility. Finally, after title and abstract screening was completed in ASReview, the titles were uploaded to Covidence once again for a second round of title and abstract screening (as Covidence enables more specific tracking of reasons for exclusion). Studies were selected for full-text review and data extraction by consensus, any discrepancies were resolved by the third researcher (LA).
Data Collection and Critical Appraisal
Data extraction was completed in Covidence by two independent researchers (AVV and LJ) and reviewed by the first author to consolidate the data extracted. Data extracted included: (a) screening tests utilized; (b) predictive validity statistics reported of the assessment/screening tool (including area under the curve, specificity, sensitivity, positive predictive value, negative predictive value, regressions and correlations, or between group differences between pass and fail on driving test) as reported (N.B. no calculations were completed for data extraction based on any raw data presented in the papers. For example, some studies included the 4x4 table, but only statistics explicitly reported by authors were extracted); (c) method for determining pass/fail (on-road or simulator); (d) recommended cut points for used screening tests; and (e) specific demographics of the study participants including: age, sex at birth, gender, geographical location of participants, racial/cultural identities and any other social demographics reported (i.e., income, education, employment status, where reported) and number of participants for each; and (f) geographic location of researchers (N.B. when demographic data does not clarify if reporting pertains to gender or sex at birth, it was presumed to be sex at birth as the driving literature tends to use both terms interchangeably). Quality appraisal followed a modified checklist of health care interventions originally developed by Downs and Black (1998). The modified checklist was developed by Duch et al. (2013) for studies that are not randomized trials. This checklist was further amended for the present study to exclude one item (regarding reporting precise probability values) which was not pertinent to all statistics reported in this review. The modified quality assessment checklist is presented in Supplemental Table 2. Quality appraisal was also completed by two independent researchers (AVV and LJ) and was reviewed by the first author to determine the final quality ratings.
Analysis
To synthesize results, data extracted from included studies was presented in an evidence table (Table 2). To address objective one, to identify tests most predictive of fitness to drive after stroke, any screening tests with findings from more than one study was presented in another table (Table 3). This table was created to optimize utility for clinicians to review screening tests commonly used in the literature, research that supports and does not support its use, the study design and sample size of relevant studies, and demographics of the study participants across studies were combined to show the subgroups represented in examination of the test. Table 3 also includes an assigned grade of practice recommendations (Burns et al., 2011) for each test which provides practice recommendations for clinicians. The grade of practice recommendations is assigned based on the (a) level of evidence (e.g., prospective versus retrospective designs) and (b) the consistency of the findings across multiple studies (Supplemental Table 3). Grade A (Strong Recommendation) is assigned when there is level I evidence, or consistent findings across multiple level II, III, and IV studies. Grade B (Recommendation) is assigned with generally consistent level II, III, or IV evidence. Grade C (Option) is assigned for inconsistent level II, III, or IV evidence. Grade D (Option) is assigned with only level V evidence or little to no empirical evidence. Further descriptions of the implications for grade of practice recommendations are included in Table 3.
Summary of Included Studies.
Note. M = mean; SD = standard deviation; PPV = positive predictive value; NPV = negative predictive value; rb = biserial correlation coefficient; r = Pearson correlation coefficient; MMSE = Mini-Mental State Examination; ROC = receiver operator characteristic; AUC = area under the curve; SE = standard error; OR = odds ratio; CI = confidence interval; EC = estimated coefficient.
Driver Screening Tests, Studies That Support and Do Not Support the Use of the Tests, Individual Study Details (Design and Sample Size), and Combined Participant Demographics.
Note. Grade of practice recommendation and descriptions are based on the American Society of Plastic Surgeons: Evidence-based clinical practice guidelines as cited in Burns et al. (2011). CBDI = Cognitive Behavioural Driver’s Inventory; DPAB = Dynavision Performance Assessment Battery; MVPT = motor-free visual perceptual test; SDSA = stroke driver’s screening assessment; VRST = visual recognition slide test.
For the second objective, to critically appraise the representativeness of demographic subgroups within stroke and fitness to drive research, each study in Table 3 included the subgroups used in the test investigation. In addition, we completed analyses of (a) the number and proportion of studies from countries across the globe, (b) the weighted mean age of participants, (c) the number and proportion of female and male participants, (d) demographic information reported across all included studies, and (e) weighted mean length of time since stroke.
Results
Following initial database searches and updated searches, 27 articles were included in the review. For complete details, including reasons for exclusion, see Figure 1 for the PRISMA flow diagram. The most common reasons for exclusion were (a) not including an on-road or simulated driving test with a pass/fail outcome (n = 14 papers excluded; examples include Heikkilä et al., 1999; McNamara et al., 2019), or (b) having dependent variables related to driving performance rather than a dichotomized pass/fail on a driving test (n = 10 papers excluded; examples include Mazer et al., 2003; Stapleton et al., 2012). Summaries of all included studies and quality appraisal ratings are presented in Table 2. The average of the quality appraisal ratings across included studies was 6/9 items, SD = 1.1, range 5 to 7. One study not published in English was included in the review (Korean language). We emailed the corresponding author to investigate if any translated copies exist; however, we did not receive a response.
Flow diagram of results of database searches, screening, full-text review, and study inclusion.
Suggested screening tests and assigned grade of practice recommendation along with descriptions of relevant studies for each study are presented in Table 3. The Stroke Driver’s Screening Assessment (SDSA) and Trail Making Test-B (TMT-B) received the highest grade of practice recommendation in the current review. Overall, studies examining the SDSA (all versions) have tested 410* participants (66 female [16%]) and the weighted mean age of participants was 61.52 years ± 5.9. (*N.B. the total number of participants in studies examining all versions of the SDSA is actually n = 520, however, Björkdahl [2015; which included n = 110] did not report sex of participants and thus was not included in this calculation). Similarly, studies that found the TMT-B to be predictive of pass/fail driving outcomes included a total of 549 participants (104 female [19%]) and the weighted mean age was 58.2 ± 2.74. Taken together, in the studies examining both measures, men comprised >80% of the sample participants consisting primarily of individuals over the age of 58 years. Moreover, all studies were completed in high-income countries in except for one study conducted in Malaysia (Munin et al., 2023).
A test commonly used in clinical practice which received a lower practice grade recommendation is the Motor-Free Visual Perceptual Test (MVPT). Although the MVPT was previously recommended for driver screening in best practice guidelines (Heart & Stroke Foundation of Canada, 2019), this test was assigned Grade C in the present study. The studies examining the MVPT had inconsistent, low-level evidence for its use, with more studies (3 out of 5) indicating it is not predictive of fitness to drive. Early studies of MVPT pointed to predictive validity, but recent studies have not replicated those findings.
To critically assess the representation of demographic subgroups within the entire body of literature, the demographics of participants in all included studies were analyzed. The results revealed only 21% (n = 479/2,270) of the study participants were female, the weighted average age of participants was 61.6 ± 5 years, and 99% (n = 2,226/2,246) of the participants live in high-income countries. The mean time since stroke was reported for a very small portion of participants (5%; n = 109/2,270), and the weighted mean for time since stroke was 12.3 months (SD = 11.6) among reporting studies. For other demographic information, only six (22%; 6/27) included studies reported the education level of participants, two (7%; 2/27) reported language, two (7%; 2/27) reported racial identity and only one (4%; 1/27) reported employment status. Among the six studies that reported specific participant education levels, most participants had post-secondary education (64.5%; n = 142), then secondary education (34.5%; n = 77) and the fewest with primary education (1%; n = 2). However, education level was only described for 10% (n = 220/2,270) of participants in included studies. The only participant racial identities reported included Southeast Asian (0.4%; n = 9/2,270), South Asian (0.09%; n = 2/2,270), Asian (0.4%; n = 9/2,270), Caucasian (2.2%; n = 50/2,270) and African American (1%; n = 22/2270), which indicates the racial identities of 97% (n = 2,198/2,270) of participants are not reported. Many studies reported “gender,” but there was no information about gender beyond a male and female binary and thus it is possible that only sex at birth was collected. No studies reported on other variables such as income, or type of employment of participants.
Discussion
The objectives of this systematic review were to (a) identify screening tests that are predictive of fitness to drive following stroke, and (b) critically appraise the representation of different demographic subgroups in extant fitness to drive research following stroke. The screening tests identified in the present review as most predictive of pass/fail for stroke populations on a driving test are consistent with the findings of previous systematic reviews. Stroke Driver’s Screening Assessment (SDSA; Nouri & Lincoln, 1992) which has American (Akinwuntan et al., 2013), Nordic (Lundberg et al., 2003) and Malaysian versions (Munin et al., 2023) was rated Grade A or Strong Recommendation. The Trail-Making-Test B (TMT-B; Partington, & Leiter, 1949) was rated Grade B, or Recommendation. A previous systematic review completed by Marshall et al. (2007) recommended Trail Making Test-B, along with Rey Osterreith Complex Figure (Rey & Osterrieth, 1941) and Useful Field of View (UFoV; Edwards et al., 2006). The systematic review by Hird et al. (2014) also recommended SDSA, UFoV, and Rey Osterreith Complex Figure. Finally, the most extensive review and meta-analysis by Devos et al. (2011) also recommends use of TMT-B, as well as Road Sign recognition test and the Compass Test, which are components of the SDSA.
Although findings regarding SDSA and Trails-B are consistent, other tests such as the Rey Osterreith Complex Figure, Motor-Free Visual Perceptual Test (MVPT; Colarusso & Hammill, 1972), and UFoV were graded as Level C or Option in this study for their predictive validity of pass/fail driving outcomes. It is likely that these tests were rated at only Level C in the present review due to studies excluded based on the dichotomous pass/fail inclusion criteria. Nonetheless, it appears that tests previously identified as predictive of fitness to drive following stroke have remained stable from the earliest review (Marshall et al., 2007) to present. Importantly, there is a significant caveat to these findings- the tests continue to be investigated using same limited subgroups of stroke populations, namely, the vast majority are older men from high-income countries.
The lack of representation of women, people younger than 55 years of age and people from lower income countries is concerning, considering these are the precise groups in which stroke incidence is on the rise (Ekker et al., 2019; Leppert et al., 2022; Scott et al., 2022). This review provides further evidence of the gender gap in stroke research (Heart & Stroke Association, 2023) and raises important considerations about how evidence supporting these tests is interpreted and communicated. Furthermore, many studies did not report other relevant demographic information of participants including gender, racial identities, level of education, employment status, and income. As these demographics are not reported in the literature, the extent to which many subgroups of people who experience stroke are represented in the literature is unknown. Not only must research include a diversity of participants to accurately reflect the population of people who experience stroke, it is also important to explore if demographics affect driver screening test results to optimize validity, reliability and cultural relevance of tests for specific subgroups (Fujii, 2018).
Stroke best practice guidelines underscore the necessity to use valid and reliable screening tools (Mountain et al., 2020; National Stroke Foundation, 2010; Winstein et al., 2016). Many tests have been validated for stroke as a diagnostic group, however much of the existing literature does not report specific participant demographic information. Consequently, diverse subgroup representation of people who experience stroke is not evident in the fitness to drive literature which poses limitations for applying research for diverse patient identities. As with all practice areas, the application of evidence-based practice occurs within the practical confines of existing research, and occupational therapists have expressed concerns about the limitations and applicability of research to clinical practice in general (Upton et al., 2014). Clinical reflection is a useful tool to situate assessment results- when selecting a driver screening tool, therapists are encouraged to reflect on the unique characteristics of the patient and the shortcomings of the literature when interpreting assessment findings (Restall et al., 2022).
The ECLECTIC framework (Fujii, 2018), which emphasizes how characteristics such as income, education, and culture impact neuropsychological testing results, may prove a useful framework within occupational therapy to guide selection of screening tools and interpretation of results. As decisions regarding fitness to drive are highly important to patients and a critical source of contention, optimizing the validity and reliability of assessment strategies is crucial (Betz et al., 2016; Korner-Bitensky et al., 2010; Sangrar et al., 2018). Indeed, a meta-synthesis of patient preferences surrounding driving conversations with health care providers found that patients want providers to include and discuss objective evidence and testing results to substantiate their recommendations (Betz et al., 2016). Thus, adopting frameworks such as ECLECTIC may help to enhance the validity and ultimately acceptance and credibility of health care provider driving recommendations among patients.
Given the limitations in demographic reporting, the current research does not provide sufficient data to wholly apply the ECLECTIC framework for occupational therapy driver screening practices. However, ECLECTIC can be applied to development of future primary research to work toward more detailed demographic reporting. Specifically, to implement the ECLECTIC framework in practice (Fujii, 2018), future primary driving research should prioritize (a) detailed reporting of participant demographic information (including income, education, language fluency, racial, and cultural identities) and (b) purposeful recruitment for subgroups of stroke populations not currently represented in research, particularly women, people younger than 55 years of age, people from racialized communities, and people from low-income countries. We also advocate for collecting and reporting of both sex and gender, as it is unclear how each impact driving related behaviors and performance on clinical tests, which warrants further inquiry (Keay et al., 2018; Özkan & Lajunen, 2006). In addition, we recommend reporting participant employment status, type of work, and working hours (e.g., shiftwork), as work-related factors also affect driving-related outcomes (Knott et al., 2020; Lyu et al., 2018). Moreover, both gender and employment factors provide further socio-cultural description of participants to inform decisions about the clinical applicability of research findings (Fujii, 2018; Restall et al., 2022).
Funders and policymakers are encouraged to incentivize such research priorities to address gaps in predicting fitness to drive after stroke. More clarity surrounding demographic details about participants will also increase transparency and awareness of groups included (and excluded) in research studies to inform questions of applicability of results for specific patient subpopulations. Most of the research was conducted with patients later in their stroke recovery trajectories (12.3 months post-stroke; SD = 11.6) and as such, more research is also needed with patients <1 year after stroke, when fitness to drive determinations are often made (when leaving acute care or rehabilitation settings; Mountain et al., 2020).
Limitations
The present review sought to include studies that evaluate fitness to drive using a dichotomized pass/fail criterion on a driving test. Although pass/fail driver tests are the clinical gold standard in driving assessment practice (Bédard & Dickerson, 2014), some research was not included that may still contribute to the evidence predicting driving ability or performance following stroke. Similarly, research with diagnostically heterogeneous samples (or where results for stroke were not reported separately) were excluded from the review (e.g., Krishnasamy & Unsworth, 2011; Unsworth et al., 2012). Stroke best practice guidelines indicate screening tests should be validated for stroke populations (Mountain et al., 2020). As such, other relevant studies/screening tools for driver fitness with other diagnostic groups (such as dementia) were excluded from this review (e.g., Eramudugolla et al., 2022 [evaluated the Mini-Mental State Examination]; Kwok et al., 2015 [evaluated the Montreal Cognitive Assessment]).
Another related limitation of the current review (and systematic reviews in general) is that the application of very specific inclusion criteria can perpetuate the same studies and findings over time (Greenhalgh et al., 2018). Indeed, the current review revealed similar findings to previous reviews and included similar studies. The research question for the current study is specific and thus is coherent with the strict inclusion criteria of a SLR methodology (Greenhalgh et al., 2018); however, a narrative review may be a useful future direction to explicate the findings of this work, particularly surrounding subgroup representation.
Finally, artificial intelligence generated translation was utilized to complete some data extraction. The gold standard for using translated works is to procure costly professional translations. However, using creative strategies to include literature may represent a cost-effective way forward to address the English language privilege in systematic reviews (Walpole, 2019).
A limitation in the body of evidence is that most studies did not provide cut points for fit and unfit to drive. Evidence for fit/not fit thresholds are most useful for clinical practice, but unfortunately, this review was not able to provide updated recommendations. The best evidence for cut point data remains within the systematic review by Devos et al. (2011) which also completed meta-analysis to calculate the cut points. The present study did not plan a meta-analysis and did not request access to raw data from included studies to complete further analysis, and thus data was extracted from papers as reported. Further research is warranted to examine cut offs for driver fitness. As clinicians are often tasked to make dichotomous decisions, such as if a report is required to be sent to the local driver’s licensing authority or not (Canadian Council of Motor Transport Administrators, 2020), research that mirrors the same dichotomous dependent variable is needed.
Conclusion
This systematic review found that although many of the recommended screening tests for driver fitness after stroke have remained stable overtime, the subgroups reported to be represented in the literature are consistently older men from high-income countries. Furthermore, research studies rarely provide detailed demographic data to characterize the subgroup representation within study samples, which further obscures the representativeness of the body of literature. The present review highlighted screening tests recommended in the literature along with the specific subgroups the findings were based on, where reported. More primary research is warranted that (a) reports detailed demographic information about participants and (b) includes participants belonging to demographic subgroups underrepresented in current research.
Supplemental Material
sj-docx-1-otj-10.1177_15394492251344518 – Supplemental material for Screening Fitness to Drive After Stroke Across Demographic Subgroups: A Systematic Review
Supplemental material, sj-docx-1-otj-10.1177_15394492251344518 for Screening Fitness to Drive After Stroke Across Demographic Subgroups: A Systematic Review by April Vander Veen, Leaha Johnston, Jeffrey Holmes, Patricia Tucker and Liliana Alvarez in OTJR: Occupational Therapy Journal of Research
Supplemental Material
sj-docx-2-otj-10.1177_15394492251344518 – Supplemental material for Screening Fitness to Drive After Stroke Across Demographic Subgroups: A Systematic Review
Supplemental material, sj-docx-2-otj-10.1177_15394492251344518 for Screening Fitness to Drive After Stroke Across Demographic Subgroups: A Systematic Review by April Vander Veen, Leaha Johnston, Jeffrey Holmes, Patricia Tucker and Liliana Alvarez in OTJR: Occupational Therapy Journal of Research
Supplemental Material
sj-docx-3-otj-10.1177_15394492251344518 – Supplemental material for Screening Fitness to Drive After Stroke Across Demographic Subgroups: A Systematic Review
Supplemental material, sj-docx-3-otj-10.1177_15394492251344518 for Screening Fitness to Drive After Stroke Across Demographic Subgroups: A Systematic Review by April Vander Veen, Leaha Johnston, Jeffrey Holmes, Patricia Tucker and Liliana Alvarez in OTJR: Occupational Therapy Journal of Research
Supplemental Material
sj-docx-4-otj-10.1177_15394492251344518 – Supplemental material for Screening Fitness to Drive After Stroke Across Demographic Subgroups: A Systematic Review
Supplemental material, sj-docx-4-otj-10.1177_15394492251344518 for Screening Fitness to Drive After Stroke Across Demographic Subgroups: A Systematic Review by April Vander Veen, Leaha Johnston, Jeffrey Holmes, Patricia Tucker and Liliana Alvarez in OTJR: Occupational Therapy Journal of Research
Footnotes
Data Availability Statement
The authors confirm that the data supporting the findings are within the article, supplementary material, and on the Covidence platform.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
