Impact of vision disorders and vision impairment on motor vehicle crash risk and on-road driving performance: A systematic review

Author: Wood JM, Black AA, Dingle K, Rutter C, DiStefano M, Koppel S, Charlton JL, Bentley SA.

Geographical coverage: USA, UK, Canada, Nigeria and South Africa.

Sector: Burden of disease

Sub-sector: Epidemiology

Equity focus: None

Study population: Car drivers with visual impairments or visual disorders.

Review type: Other review

Quantitative synthesis method: Narrative review

Qualitative synthesis method: Not appropriate

Background: Vision is important for safe driving and, due to an ageing population, an increasing proportion of older drivers are likely to have visual difficulties. However, there is limited understanding regarding the impact of vision disorders on driving ability and safety – which is needed to determine, for example, what aspects of vision to include in driving tests.

Objectives: This review aimed to evaluate and summarise evidence on the impact of vision disorders and impairment on motor vehicle crash (MVC) risk and on-road driving performance across seven databases.

Main findings:

Eight studies met the inclusion criteria for MVC risk (n = 36), on-road performance (n = 9), and both MVC risk and on-road performance (n = 3). Of these studies, less than half were rated as ‘good’ quality. Due to the small number of studies and often conflicting findings, authors did not draw firm conclusions for most vision disorders. However, evidence from several ‘good’ and ‘fair’ quality studies suggested increased MVC risk with binocular visual field impairment. There was mixed evidence regarding the impact of cataract, glaucoma, age-related macular degeneration and homonymous field loss on MVC risk and no evidence of increased MVC risk with mild VA impairment.

The review suggests that different vision disorders impact on driving ability and MVC risk in different ways and it is important that clinicians are aware of this issue when advising their patients with visual impairment. People are often unaware that their vision and driving performance is changing. Regular eye examinations serve not only to detect eye diseases and vision impairment early, but also provide the opportunity for clinicians to inform patients of relevant changes in their vision and how this may affect their driving. This is important to foster insight, encourage self-regulation, compliance with vision health management protocols to reduce disease progression where possible and to support transitioning to alternative mobility options when driving independently is no longer possible.

This review highlights the need for well-designed future studies to further explore the impact of vision disorders and impairment on driving outcomes to inform evidence-based policy and fitness to drive guidelines.

Methodology:

Articles were selected for review if they were original, peer-reviewed research in English, had full-text available, included drivers with vision disorders or impairments, and control drivers without such conditions. They needed to assess crash risk or on-road driving test outcomes, and use quantitative methods for data collection and analysis. Exclusions were made for studies using only driving simulators, non-drivers, qualitative methods, or were commentary manuscripts, reviews, case studies, conference abstracts, proceedings, or dissertations.

A search was performed in seven electronic databases including Cochrane Library, Ovid Medline, PsycInfo via EbscoHost, EMBASE, CINAHL PLUS to locate relevant studies up to 2 April 2020.

Three of the authors worked independently and in duplicate to screen all of the remaining titles and abstracts for inclusion. Any queries raised were resolved by discussion with the team. For the full-text screening stage, three authors worked independently and in duplicate to include/exclude studies using the study criteria and coded the reasons for exclusion. Disagreements were resolved by consensus. The data extracted by one of the reviewers was independently verified by another reviewer to identify and rectify any errors.

Risk of bias was assessed independently by two authors using The National Heart, Lung and Blood Institute Quality Assessment Tools (NHLBI 2014). Two reviewers assessed each study independently and in duplicate and did not assess the quality of their own studies.

A narrative review rather than a meta-analysis was undertaken due to the vastly different outcome measures, sample characteristics, and study designs included in the evidence synthesis. Where a study did not report any statistical findings relevant to the research question of this systematic review, the authors calculated a crude OR and 95% CI using online statistical tools.

Applicability/external validity: The weaknesses in study design and analysis for the included studies may have limited the extent to which the reported findings are more broadly applicable.

Geographic focus: Studies were derived from a range of countries, primarily with a high income. There was no consideration of how the reported findings (and their implications) might vary in different national or economic contexts.

Summary of quality assessment:

The methods used to identify, include, and critically appraise studies had several limitations. The search was confined to published, English-language studies, and there was no indication that the reference lists of the articles considered were reviewed. The review lacked a discussion on the appropriate time frame. While the data analysis was generally robust and focused on study quality, the choice of a narrative approach, although suitable in this context, limited the extent of the analyses. Due to these limitations in the study identification, inclusion, and appraisal methods, we have low confidence in the review’s findings.