Irlen syndrome: expensive lenses for this ill defined syndrome exploit patients

Philip Griffiths is right to point out that interpretation of the results of our double-masked study is compromised by the high rate of attrition. I made mention of this in my first response and alluded to such criticism in my second. It should be borne in mind that this study was conducted 20 years ago before the CONSORT guidelines were first published.

It is perhaps unwise to conclude that “the study can only be taken as evidence that precision tinted lenses do not improve real world reading in subjects with reading difficulties”. First, in this study at least, the population was not selected on the basis of reading difficulty; second, the lenses optimally tinted and those suboptimally tinted, used a control, both improved reading rate compared with baseline; third, if the study is indeed compromised by attrition, so are all the inferences that can be made from it, both positive and negative.

It is important to remember that a scientific study, no matter how good, cannot demonstrate the absence of an effect; it can, however, fail to demonstrate the presence of an effect, which is not the same thing. In this instance, our failure to show an effect of tints on reading speed was most probably due to the variability in measurement of reading rate introduced by comprehension of the text. Subsequent work over the ensuing years has developed alternative measures of reading rate that avoid this source of variability and are clearly sensitive to visual factors, including font design, prismatic correction, and, yes, coloured filters.

Although the precision tints may help with the poorly defined syndrome identified by Meares and Irlen, there is little point in comparing a rose tint with Irlen filters, as suggested by Sam Lewis, partly because the Irlen lenses have not been described in a scientific publication and the rationale behind their design is unclear, and partly because there is evidence that a one-size-fits-all rose colour is not generally beneficial.

In my view, the burden of evidence of benefit from precision tints is sufficient to justify a properly designed fully funded clinical trial, but if public money is to be spent on such a trial it might be best to begin with an investigation of photophobia in migraine, a relatively well-defined condition which is common, disabling and has a considerable economic burden. The need for a trial of precision tints in migraine prophylaxis is indicated by (1) anecdotal evidence of patients whose migraine headaches have been reduced by precision tints; (2) two studies that show, in the visual cortex of patients with migraine, an abnormal haemodynamic response to visual stimuli, a normalisation of this response when precision tints are worn, and no such normalisation with control tints; (3) a small-scale randomised control trial of precision tints in migraine with weak but positive outcome; (4) reports from optometrists that their patients have reported they can prevent or abort their migraine aura by wearing precision tints.

Over five of the last 20 years I have tried repeatedly to get such a study under way, but my applications for funding have not met with success.

In his “Author’s Response” of 30th July 2014 modified by The BMJ on 4 August 2014, Mr Gwyn S Williams says:
“This was highlighted by two randomised placebo controlled trials of coloured lenses which demonstrated improvement in all groups but with the greatest improvement, interestingly, in the control/sham groups6,7.”

As one of the authors of the randomised controlled trial numbered 7 in the above quotation, I should point out that the results were not as Mr Williams quotes them, but rather the reverse. He should look again at Figure 2. The symptoms were greatest in the control, not the active group. The study may be criticised on other grounds, but surely not by quoting the reverse of the results that we in fact obtained. In any event, the results of this trial relate to tints prescribed from the Intuitive Colorimeter system, and not those provided by Irlen.

As an "Irlen's sufferer" I find myself quite torn with this article, as I can understand the concern being addressed in regards to some people being misdiagnosed.

I was diagnosed when I was 8 years old, I recall it very well. I was given no indication why I was seeing this consultant, I was also given no prior knowledge of any symptoms. My parents were very skeptical at first, but were deeply worried of my performance in school so it was a last measure. They were very away that my performance didn't make sense since I was described by teachers are being very articulate for my age group, I have old school reports of teachers claiming my verbal spelling ability was well above my age, yet I could barely read a sentence under 10 minutes with writing being out of the picture completely, which obviously makes little sense. This led some teachers to believe I was "Lazy" for which I was actually bullied by said teacher.

When it come to being diagnosed I hit the nail on the head with every single symptom, such as "If I look away and look back fast enough, I can catch the letters", "The heads move about" so there is little doubt there is something very wrong with certain people such as myself. My parents were instantly convinced and like many I got my glasses, I can't say with a straight face they completely worked, but without a doubt they helped. I like to say I had a window of opportunity without glasses of about 15 minutes until its effectiveness wore off and about 30 with them on.

In the same breath I can't say I'm exactly like other Irlen's sufferers since I can read a computer screen with little issue (No movement, Shaking, Merging of text etc etc) Whereas other sufferers seem to have it bad on both mediums, although I will clarify that although I can read text in books now albeit very slowly, it doesn't look remotely the same as it does on either. So basically I had to learn to read the same text twice because of the differences, and my writing capability although very slow and tedious to do, I manage to do "almost" entirely by muscle memory. Sadly this leaves my writing very untidy which gets progressively worse the longer I write.

So even though I do find your article somewhat offensive, I can understand your concern in regards to over diagnosing people. I don't actually bother with the lenses myself since I feel its the equivalent to duct taping a leaky pipe. Its temporary and as suggested in the article ITS VERY EXPENSIVE!, which for me personally is little improvement for the price of them. Then again I was last tested at aged 8, I may have picked them based on preference of colour rather than effectiveness? Who knows, but it would be nice for these very real issues to be looking into by trained health professionals and hopefully help find a more effective method of treatment.

Research using the Intuitive Colorimeter has been cited as supporting the Irlen filters both in the recent correspondence and on the Irlen website. There are fundamental differences between the methods used by the Irlen centres and those of the Intuitive Colorimeter system, and in consequence of these differences, research using the Intuitive Colorimeter should not be taken as supporting the Irlen system. The research does show that coloured filters can have dramatic beneficial effects for some individuals, however.

The Intuitive Colorimeter is an apparatus that illuminates text with coloured light and permits the separate and precise control of the hue, saturation and brightness of the light. While working at the Medical Research Council Applied Psychology Unit, I designed the original version of the instrument to investigate the claims made by Irlen and the reports of benefit from patients using her lenses, which are proprietary. The colorimeter allows the effects of a colour to be studied while the eyes are adapted to that colour and the absence of surface colours avoids contamination of the judgment by colour constancy. It is simple to assess the consistency with which individuals choose a colour and to do so without their awareness. This is important because individuals who are inconsistent in their choice are not good candidates for tints.

Using this instrument I discovered that some individuals with reading difficulty reported distortions of text that abated when the text was illuminated by light of a particular colour. The colour was remarkably specific and different for each individual. Eventually it became apparent that coloured lenses were a satisfactory alternative but only if they provided an exact match to the colour of light chosen in the colorimeter. With Tim Noakes I then evolved a system for tinting lenses the precise shade. We used seven stable rapidly absorbed primary dyes with colours evenly spaced around a hue circle. The system can obtain a very close approximation to any colour in a large gamut. The dyes were chosen to have the maximum transmission possible. They vary smoothly across the visible spectrum, so as to minimise the changes in colour (metamerism) that can occur when the lenses were worn under lighting with a peaked spectral power distribution. The colorimeter was redesigned so that its coloured light had a spectral power distribution identical to that experienced when the lenses were worn under conventional white fluorescent lighting (CIE F3). This meant that anomalous colour vision could be accommodated. White fluorescent light is stable over time and has a colour midway between the yellowy white of incandescent light and the bluey white of daylight. The lenses were therefore appropriate for classroom and office use and the colour the patient experienced differed minimally under other lighting.

The Colorimeter system has been used in two double-masked trials. The Colorimeter permits a double-masked protocol because the patient adapts to the coloured light and is unaware of its precise shade and strength. Two types of lenses were provided. One matched the chosen colour and the other differed by 6 just-noticeable differences, a colour difference just sufficient to negate any benefit. The lenses were provided, each for at least a month in random order, and patients were later unable to say which was which. (Memory for colours is poor unless the shade can be easily named). Both studies showed a reduction in symptoms on days when the correct lenses were worn relative to days when the control lenses were worn. The first study has been criticised for an unacceptably high attrition, and the second for its small sample size, and further large-scale randomised controlled trials will be necessary.

There is little to suggest that the benefits of coloured lenses are due to the optics of the eye, in particular accommodation. The lag in accommodation has been shown to be greater in individuals with visual stress, but colour does not appear to influence the lag by virtue of chromatic aberration. Instead the pathophysiological mechanisms appear to be cortical. There is convergent evidence that in patients with migraine the cortex is hyperexcitable. There is an abnormally large haemodynamic response in the visual cortex when patients with migraine observe patterns of stripes, and this is reduced to normal levels when the stripes are observed through lenses tinted using the Colorimeter system. Control lenses (grey and coloured) are without effect. This suggests that precision tints may be a useful treatment for some patients with migraine, those with photophobia whose attacks are visually triggered.

Whatever tinting system is used, the assessment is subjective and dependent on the clinical skills and experience of the examiner. At present the only objective physiological correlate of likely benefit is the reduction in abnormal brain oxygenation as revealed by the haemodynamic response referred to above. Sometimes the benefit extends to an improvement in reading speed, however, and this can be assessed using the rate at which a paragraph of randomly ordered common words is read aloud (Wilkins Rate of Reading Test). This rate has been used to investigate the precision in colour necessary for optimal effect, and by implication, the number of trial lenses necessary in any tinting system. Both in children and adults an improvement in reading rate is commonly observed with coloured filters, even by investigators who are unsympathetic to their use.

Unfortunately, there is sometimes a tendency for examiners to over-prescribe, perhaps in the desire to help. Although Richie et al failed to find an improvement in reading rate in clients diagnosed with Irlen syndrome by an Irlen diagnostician, this cannot be taken as evidence that coloured filters, appropriately selected and prescribed, are unhelpful for the patients that need them.

For further details, please see the following reviews: Wilkins, A.J. (2012). Origins of visual stress. In J. Stein and Z. Kapoula (ed.) Visual aspects of dyslexia. Oxford University Press: Oxford. pp 63-78 and other articles, particularly those entitled Vision and Reading Difficulties 1-5, downloadable from www.essex.ac.uk/psychology/overlays/publications2.htm .

Arnold Wilkins, University of Essex