Testing acuity of vision in general practice: reaching recommended standard

BMJ 1994; 309 doi: (Published 26 November 1994) Cite this as: BMJ 1994;309:1408
  1. J C Pandit, registrar in ophthalmologya
  1. a Ophthalmology Department, Torbay Hospital, Torquay, Devon TQ 2 7AA
  1. Correspondence to: Nuffield Laboratory of Ophthalmology, Oxford OX2 6AW.
  • Accepted 21 July 1994

Measurements of acuity of vision are highly reliable with a reliability in excess of 98% according to one important study1—because a standard protocol exists. The British Standards Institution specifies the minimum illumination for externally illuminated charts and the distance that patients should be from the chart.2 Compliance with this standard helps to ensure comparable results in the measurement of acuity of vision. Patients' acuity of vision measured in the eye department at this hospital, however, sometimes differs from the measurement given in their referral letter. I examined whether general practice surgeries complied with standards on illumination and distances.

Methods and results

I randomly selected 24 of the 48 main general practices that fell within the boundaries of the former Torbay District Health Authority. Branch surgeries were excluded from the survey. In September 1993 I visited 22 of the 24 surgeries and measured the distance between the patient and the test chart with a 7 m metal tape measure to the nearest centimetre and the illumination falling on each chart with a luxmeter to the nearest 10 lx.

The 22 surgeries had 67 rooms with a fixed Snellen chart (the most commonly used chart for testing acuity of vision). All the surgeries had at least one chart, either a chart designed to be read at 3 m (54 rooms) or one to be read at 6 m (13). Two of the 6 m charts were of the reversed type—that is, the chart is behind the patient, who reads it in a mirror directly opposite.

The distance the patient had to stand from the chart was shown in only one of the examination rooms (which had a 6 m chart). None of the general practitioners or practice nurses was aware of the minimum lighting requirement of 480 lx.

The distance that the patient had to stand from a chart ranged from 206 cm to 480 cm (mean 305) for the rooms with 3 m charts and from 310 cm to 600 cm (512) for those with 6 m charts. External illumination ranged from 60 lx to 1600 lx (447) for the rooms with 3 m charts and from 100 lx to 725 lx (325) for those with the 6 m charts.


Few tests can quantify loss of any sensory function as easily as the test of acuity of vision. Loss of acuity may be the only indicator of the onset of serious eye disease. Testing acuity of vision is particularly important, for example, during the screening of diabetic patients for retinopathy including macular oedema. Even a one line change in acuity may be of importance and should arouse suspicion.

Driving is a key activity for people in isolated rural areas, both for employment and for daily living. The standards of vision for driving are strict; if testing of vision is inadequate public safety may be at risk or people may lose their livelihood unnecessarily.3



Snellen charts for testing acuity of vision should be the recommended distance from the patient for the size of chart used—that is, 300 cm for 3 m charts and 600 cm for 6 m charts. The distance should be marked so that it cannot be erased, and the patient should stand behind this mark.


The minimum illumination for externally illuminated charts is 480 lx. This can easily be achieved by directing a spotlight at the chart.

This study shows that most general practitioners estimate rather than measure the distance between their patients and the vision test chart, and none of the doctors was aware of the minimum standards on illumination (see box for recommendations). Although I did not study the effect of the variation in distances and illumination on patients' recorded acuity of vision, it would be fair to assume that the reliability and reproducibility of measurements of acuity of vision made in general practice may be in doubt.

I thank Miss Jayne Walker for typing the manuscript and Dr Peter Featherstone for advice in preparing it for publication.