International Standards:

Vision Requirements for Driving Safety

Section 4 – Which Visual Functions Might Be Tested?

Some of the current visual requirements for obtaining or keeping a driving license are listed in the Appendices.

Appendix 1 lists the responses of members of the ICO to a questionnaire sent in 2002.

Appendix 2 surveys the situation in the USA.

These tables are, of course, incomplete and variable.  However there are some general tendencies. The most common requirement is a visual acuity of 20/40 (0.5, 6/12).

There appears to be a widespread consensus that this provides an adequate safety margin between visual acuity measured in the office and driving safety as practiced on the road.

Requirements for visual field are more variable. Requirements for other visual functions are rare.

We will discuss various visual functions that might be tested.

Visual acuity

Visual acuity is the visual parameter that is most easily and therefore most widely measured.  It is the reciprocal of the magnification needed to bring a high contrast detail to the threshold of visibility.

It is often mistaken for a general measure of vision, although it only tests the small foveal area onto which the test letter is projected. For optical problems (defocus, opacities) this is adequate, since foveal blur predicts perifoveal blur.

For retinal problems (also prevalent in an older population) visual acuity is only a partial measure, since foveal function does not predict perifoveal function.

The 20/40 (0.5, 6/12) standard is the criterion most widely used.  We believe this to be reasonable, not because one becomes an unsafe driver at 20/50 (0.4, 6/15) but because it includes a safety margin for adverse conditions.

This criterion implies that subjects who can read 20/40 on a well-lighted, stationary chart, are generally assumed to still be safe drivers in a moving environment and under adverse conditions, such as after dark, in rain or in fog.

Many consider 20/40 as "half of normal".  This is not true; normal visual acuity is 20/16 (1.25, 6/5) or 20/12.5 (1.6, 6/4), so a person with 20/40 needs 2.5 x or 3 times more magnification than normal. Drivers with 20/40 visual acuity must come 3 times closer to a road sign to read it and this reduces the time available to react to one third of normal.

Visual acuity measurement should of course be standardized. We make a plea for the use of charts with a logarithmic progression, such as in the EDTRS standard, as was recommended in the 2002 ICO/IFOS resolution ^[1].

Normal vision is binocular vision, as recently stressed by a WHO Consultation on Characterization of Vision Loss ^[20]. The criteria should thus be based on binocular vision (both eyes open). In the rare cases where binocular vision is worse (e.g. diplopia), good monocular acuity should not have precedence.

Contrast sensitivity

Contrast sensitivity may be reduced due to optical factors, as in cataract patients.  Contrast problems may also result from retinal problems (AMD, glaucoma, etc.) that are also common among the elderly.

If contrast sensitivity loss is caused by optical problems (defocus, scatter), both visual acuity and contrast sensitivity will be affected.

When contrast sensitivity loss is caused by retinal problems, visual acuity loss and contrast sensitivity loss are not necessarily correlated. It is possible to have poor visual acuity with good contrast vision; it is also possible to have good visual acuity with poor contrast vision.

Brabyn et al. (2001) ^[21] showed that some people in an elderly population may have 20/20 (1.0, 6/6) acuity on a high contrast chart in good illumination, but may easily drop to 20/200 (0.1, 6/60) or below with low light, low contrast and glare.

Mäntyjärvi & Tuppurainen (1999) ^[22] strongly suggest to include simple tests for contrast sensitivity and glare sensitivity in the requirements for a driving license in older drivers.

Various tests are available. The Pelli-Robson contrast sensitivity test chart is standardized, but not often included in clinical exams. A recent variant is the Mars chart for hand-held use (Arditi, 2004) ^[23].

For any test the recommended safety margin for safe driving needs to be defined. Appendix 3 lists a suggestion for a cheap and simple screening test.

Glare sensitivity

Glare sensitivity may similarly result from optical problems, such as cataract, or from retinal problems. In the first case straylight and disability glare are important; in the latter case, glare recovery time is also important. A recent European study ^[24] validated the use of a new straylight meter in an international population study.

Visual field

A major peripheral impairment of the visual field is a reason for denying a driving license in most states in the USA and this condition is also included in the Guidelines of the European Commission ^[25].

Central scotomata reduce visual acuity and are thus caught under the visual acuity requirement. No good data exist on the effect of mid-peripheral scotomata.

Yet, some states do not list field requirements at all, and when requirements are listed the way the field should be measured is often not indicated. Financial considerations may play a role. Reliable testing is expensive and the yield in accidents prevented is limited.

Use of standardized nomenclature is important. We need to specify what we mean. To most ophthalmologists a "30° field" means a 30° radius, to non-ophthalmologists it may mean a 30° diameter. Does a 120° field require 60° to the left and 60° to the right, or does 30° left and 90° right suffice? Some US rules specify "70° in either eye".  They probably intended 70° to either side of fixation, especially if it is followed by "140° binocularly".

Clinical visual field testing is aimed at the diagnosis of underlying disorders.  Sophisticated field testing equipment and algorithms have been developed for this purpose; each eye is tested separately and eye movements invalidate the test. Yet, functional vision is binocular vision and functional use of the visual field is impossible without an effective scanning strategy.

As stated before, functional criteria should be based on binocular vision, i.e. vision with both eyes open.  This is especially important for visual fields, where good areas in one eye may compensate for scotomata in the other eye.

Since there is no equipment that allows binocular testing with binocular monitoring, monocular testing with later superimposition of the monocular fields can be used as recommended in the Vision-99 Guide ^[26] and in the current AMA Guides to the Evaluation of Permanent Impairment (5th edition, 2001) ^[27]. However, this does not yet address the importance of adequate scanning strategies ^{[28, 29]}.

Appendix 3 contains a suggestion for a type of field test aimed at detecting functional consequences, rather than at diagnosing underlying disorders.

Useful Field of View (UFOV)

This test evaluates to what extent an object in the periphery of the visual field is not only seen but also perceived. As such, it evaluates a combination of visual and non-visual factors (attention).

According to a study by Owsley (1994) ^[30] the UFOV test had better sensitivity and specificity than visual sensory or mental status tests in identifying older drivers at risk for accidents.

Perceived driving disability (PDD) assessed by a questionnaire, was strongly correlated with visual acuity, contrast sensitivity and UFOV (van Rijn et al., 2002) ^[31]. These studies, however, need to be confirmed. The UFOV test has, as far as we know, not yet been accepted for general use and is not incorporated in the European guidelines.

The UFOV test points to the importance of factors, such as attention, that are not strictly visual.

Diplopia

A few jurisdictions mention diplopia. A recent Canadian proposal recommends the absence of diplopia within the central 40° (ie: 20° left, right, above and below fixation).

Individuals vary greatly in the degree to which they are bothered by diplopia. Some are able to suppress the unattended image when looking through a monocular telescope or other monocular device, others close one eye.

Traditional tests do not address this issue. We recommend that this issue be left to the gray zone of individual consideration.

Color vision

The Guidelines of the European Commission ^[25] have dropped color vision requirements.  They are still in use in some states in the USA, in Bulgaria, Columbia and provinces in Canada.

Studies by Verriest et al (1980) ^[32] have shown that abnormal color vision is not incompatible with safe driving. The problem of recognizing traffic lights is overcome by the standardized position of the different lights, appropriately chosen colors and in some countries by the differences in their sizes. 

Night vision

Problems with night vision are not limited to hemeralopia. Patients with hemeralopia are aware of their problems and will usually avoid driving at night. 

Patients with IOL’s or who underwent refractive surgery may experience glare problems and contrast loss when a wide pupil exposes the edges of the IOL or of the ablation zone. Such problems will never be detected in daylight testing. Also, patients with cataract will complain of glare and be extremely handicapped by the lights of cars driving towards them.

Especially older drivers will be blinded by oncoming cars because of increased intra-ocular light scattering and glare sensitivity, but also because of prolonged photostress recovery time (Mainster & Timberlake, 2003) ^[33].

There are a number of tests that evaluate problems with night vision, including the Mesotest (Oculus) and the Nyktotest (Rodenstock) (van Rijn et al, 2002) ^[31]. None of them are included in guidelines, although some countries include a crude estimation of night vision in their criteria.

Next: Vision Requirements for Driving Safety: Section 5 – Suggested Criteria and Rules

Also see: Table of Contents

Download the report on Vision Requirements for Driving Safety as a PDF file (369 KB)