Protans and driving safety

EDITOR: An unfortunate comparison between drunk drivers and protans was made in a Letter to the Editor in Clinical and Experimental Optometry,January 2002.‘ It has been cited in at least two articles in yourjournal since it Ifeel compelled to write as the comparison is invalid, dangerous if taken seriously and offensive to those with a red disability. One needs to be cautious and sensitive when commenting on any disabled group, even if their disability seems trivial. Dr Vingrys started with the well-known relationship between blood alcohol concentration (BAC) and the factor by which the risk of causing a road crash is increased with increasing BAC. For a driver with a level of 0.05 per cent BAC, the risk of accident is twice that of a driver with zero per cent BAC. He has used the graph found on the NSW Motor Accident Authority Website http://www.maa.nsw .gov.au/road-safety/drin k/. On the same axes, he has plotted values that he says are based on data in a paper by Verriest and colleague^.^ (Figure 1). He says that the values he has plotted represent the risk of accident for ‘severe’, ‘protan’ and ‘protanope’. In Verriest’s study 2,058 traced ‘accidentdrivers’ responsible for accidents around Dresden between November 1976 and May 1978 were tested for colour perception and divided into groups depending on the type of accident they caused. The result was that all colour perception groups were represented among the group of accidentdrivers in almost exactly the same proportion as their representation in the community. End of story? No. The five short paragraphs in this paper by Verriest and colleagues4 have been re-analysed to come up with the conclusion opposite to that of the author. The errors Dr Vingrys makes include but are not limited to the following: 1. He has quoted from Verriest and colleague~,~ accident risk in accidents per 100,000 kilometres but there is no mention of this unit in the paper, nor can such figures be derived from any part of it. 2. Dr Vingrys’s quotation of Verriest and colleagues4 is wrong; The red defectives were not over-represented in accidents at intersections controlled with traffic lights or among all accidentdrivers in the study. 3. He has calculated a figure that he says represents the increased risk of accident for protans. His figure really is the factor increase in the chance that a protan has over a colour normal that the class of accident caused is a rearend accident. What he does not say, and what is really much more interesting, is that the risk that a protan is involved in all other kinds of accident is much less than for a colour normal. Other classes of accident, such as head-on smashes, running off the road or hitting obstacles, are of far more import than rearend smashes when one considers the much higher morbidity and mortality as~ociated.~ 4. The group Vingrys calls ‘dichromats’ or ‘severe’ on his graph are given an ‘accident rate’ of ‘1.7 to 1.9 times greater’ than those he calls ‘normal observers’. It is uncertain how this was derived or what he means but as this group was over-represented in accidents at traffic lights, as it is the only other group for