Following our posts of “Five Ways Not to Treat Dyslexia,” parts 1 and 2, we received a number of comments from individuals who disagree with my statements and advocate the use of the systems and treatments refuted in the articles. In “Five Ways” I discussed vision therapy (VT), Irlen lenses, ChromaGen lenses, low-plus reading glasses, and omega-3 fatty acids.
In this somewhat lengthy post, I would like to recap the comments, take a further look at the evidence for these therapies and offer some suggestions to help parents be more critical in their evaluations of dyslexia treatments.
Thank you to all of our readers and to those who wrote in reply to “Five Ways Not to Treat Dyslexia.”
Summarizing the comments
We heard from Leonard Press, O.D., a vision therapist who opines that the information I presented in “Five Ways” is “biased and distorted.” He says that my presentation “skews the issues to sell a phonics-based program.”
Dr. Press has elsewhere been an ardent critic of the 2009 American Academy of Pediatrics (AAP) Joint Policy Statement, Learning Disabilities, Dyslexia, and Vision, which I strongly endorse and to which I refer in “Five Ways.”
Dr. Press is a co-author of the American Optometric Association’s policy statement Vision, Learning, and Dyslexia. (Note: the optometrists label theirs a “joint” policy statement, which might cause it to be confused with the AAP Joint policy statement. However, whereas the AAP policy statement reflects an agreement between pediatricians, ophthalmologists, and pediatric ophthalmologists, the American Optometric Association/American Academy of Optometry “joint” statement reflects an agreement between the optometrists and…the optometrists.) As I noted in “Five Ways,” the optometric policy statement correctly says, “Vision therapy does not directly treat learning disabilities or dyslexia.” Dr. Press and his colleagues would have done well to stop there, but they go on to say that vision therapy improves “visual efficiency and visual processing, thereby allowing the person to be more responsive to educational instruction.” As I noted in “Five Ways,” those statements on their face sound reasonable if you buy the optometrists’ assumption that “visual efficiency” problems and “visual processing” problems play a major role in dyslexia. The problem is that there is no valid evidence that they do, and on the contrary, there is valid evidence that dyslexic readers have neither more nor fewer eye or vision disorders than nondyslexic readers.
Here are the facts. Dyslexia is a language processing problem, not a vision problem. Vision therapy does no direct harm to the child, but it does waste the parents’ money, which would be put to much better use by pursuing language therapy of the sort recommended by the International Dyslexia Association (IDA). And more importantly, by wasting valuable time, vision therapy does harm the child if, as is too often the case, the parents pursue vision therapy instead of the valid, language-processing therapy the child needs. Finally, regarding Dr. Press’s assertion that I “skew(ed) the issues to sell a phonics-based program,” please be aware that I have no financial interest in Lexercise nor in any other dyslexia diagnosis or remediation method, program, or company.
I appreciate Dr. Press’s interest in our blog, but I could not disagree with him more, and the evidence supports my position as stated in “Five Ways” and in the AAP Joint Policy Statement, not his. I stand by the facts as presented in “Five Ways.”
Among proponents of Irlen lenses, we heard from Helen Irlen, MA, who originated the still-unproven concept of “Scotopic Sensitivity Syndrome” (now known as “Irlen Syndrome”), for which her Irlen lenses are the supposed treatment. Ms. Irlen felt there were flaws in the very recent Pediatrics article I cited (which showed no benefit of Irlen lenses for reading disorders) and by way of rebuttal cited a flawed study that claims to show a benefit of her lenses.
(Note that Irlen International newsletters also credit Irlen lenses with helping people suffering from visual symptoms associated with head injuries, concussions, whiplash, perceptual problems, neurologic impairment, memory loss, language deficits, headaches, autoimmune disease, fibromyalgia, macular degeneration, cataracts, retinitis pigmentosa, complications from refractive surgery, depression, anxiety, schizophrenia, multiple sclerosis, and Asperger syndrome, among others.)
Ms. Irlen also cited SPECT scans as evidence of the effectiveness and validity of her syndrome and her lenses. I have searched without success for objective information regarding SPECT scans; I find them referred to only on sites of Irlen clinics and related facilities. I would be happy to review any objective information Ms. Irlen has regarding SPECT scans themselves, and any peer-reviewed evidence regarding their validity in diagnosing or treating dyslexia. I have been unable to find any. You can read Ms. Irlen’s comment in the section below the original post.
We heard from an occupational therapist who believes a) in vision therapy and b) that I am insufficiently educated because if I were sufficiently educated, I would believe in vision therapy, too.
We also heard from John Hayes, a representative (perhaps the inventor; I’m not sure) of See Right Dyslexia Glasses, who felt that not only was I wrong for saying that dyslexia is a language processing problem and not a vision problem, but that the vision therapy, Irlen and ChromaGen people have it wrong too, and that his glasses are the real treatment for dyslexia.
Reviewing the Evidence
Let’s look quickly at the evidence regarding each of the vision-based treatments.
Vision Therapy (VT)
There is no randomized controlled trial of vision therapy for dyslexia. However, in 2000 and again in 2008, the UK College of Optometrists reviewed the evidence regarding the effectiveness of VT for several conditions, including learning disabilities, and found insufficient evidence of effectiveness to recommend VT for learning disabilities. (See “A critical evaluation of the evidence supporting the practice of behavioural vision therapy.” Ophthalmic Physiol Opt. 2009 Jan;29(1):4-25.)
Please see the Pediatrics article referenced in the first half of “Five Ways”, which is the most recent of several well-designed studies that show the lack of legitimate effectiveness of these tinted lenses/filters for dyslexia. (You will note that under Responses to this article, Helen Irlen offered a rebuttal very similar to her reaction to my post, and the authors, Stuart J. Ritchie, et al, replied at length defending their research.)
The single best study of tinted lenses is “Do tinted lenses improve the reading performance of dyslexic children?” by Sheryl J. Menacker, Michael E. Breton, Mary L. Breton, Jerilynn Radcliffe, and Glen A. Gole in Archives of Ophthalmology, 1993, 111(2): 213-218. An abstract is available here.
A recent paper that is particularly strong in assessing the flaws in studies that claim to support the use of tinted lenses is “A review of three controversial educational practices: perceptual motor programs, sensory integration, and tinted lenses” by Keith J. Hyatt, Jennifer Stephenson and Mark Carter, in Education and Treatment of Children, May 1, 2009. The authors know the research methodology very well and can detect flaws in design and interpretation that most people (including me) might miss. Furthermore, they have no eye care connection, so they have no ax to grind on either side of the optometry/ophthalmology battles.
As cited in Part 2 of “Five Ways,” the only two studies claiming effectiveness are by the inventor, who obviously has a financial interest.
Low-plus reading glasses
Numerous non-optometric studies have shown no effect for these glasses, but perhaps the most convincing evidence is again from the UK College of Optometrists, who in their 2000 assessment of vision therapy techniques (cited above) reviewed critically the optometric literature regarding low-plus readers, evaluating the quality of the studies’ design to assess the strength of the studies’ conclusions. They found no convincing evidence that low-plus readers have any benefit.
Omega-3 fatty acids
Please see my comments in “Five Ways” part 2.
The Parents’ Dilemma
We know that parents are faced with hard decisions and persuasive voices as they choose among therapies. In fact, they may have to make a choice between views that are exact opposites of one another, and certainly, opposite views cannot both be right!
As a parent, how are you supposed to decide what’s right — and what’s right for your child? It’s almost overwhelmingly tempting to rely on anecdotes and testimonials from individuals promoting a method or from friends. But if you are able to assess the evidence for yourself, if you learn how to find the flaws in studies and claims, you won’t have to take anyone’s word — including mine — and you’re much more likely to make the right decision for your child.
Questioning the Evidence
There are levels of evidence in science and medicine ranging from randomized trials to anecdotes. The lower the level of evidence, the less reliable it is.
The lowest level of evidence is the anecdote or testimonial. That’s also the type of evidence we’re most familiar with. (Example: your best friend took a new cold remedy and her cold got better. She tells you it’s the greatest thing since sliced bread. It seems like a no-brainer that it would be worth trying the same remedy if/when you get a cold; after all, it “worked” for her.) It is almost impossible for any of us — myself included — not to be persuaded by anecdotal evidence: it’s right there in front of us, and we don’t have to do a bit of work to interpret it. She had a cold, she took the remedy and the cold got better. How much clearer can it get? (Note: just because something is anecdotal does not mean it’s false — it’s just the least likely to be true of all the forms of evidence.)
All of the study types listed on the graphic above are more likely to produce true results than anecdotal evidence; the higher you go on the list, the more reliable the results will be. However, unless you’re a researcher yourself, you almost certainly won’t be able to figure out whether the study you’re trying to read is a correlational study, comparison group, or a randomized controlled trial. But let me give you some general guidelines that you as a parent might find useful as you try to assess claims regarding therapies for struggling readers. Each of these is phrased as a question you might pose when you’re considering a claim.
Where’s the control group?
Consider a study in which 1000 people with colds were given a cold remedy and the cold got better in all 1000. The author claims a 100% success rate for the remedy, and we are tempted to believe that this remedy really works — after all, it has a 100% success rate! But wait a minute — was it the cold remedy that made them get better? No, they were going to get better anyway, because that’s what colds do! So what’s missing in this “study” is a control group: another group of about 1000 people (as similar to the first 1000 as possible) who had a cold and who did not get the cold remedy. Because colds get better, the study would now show 100% cure in both groups, which means that the remedy had no effect whatsoever on the cure rate.
So in any study that claims effectiveness for a therapy, always look for the control group. If you look only at the summary — “100% cure rate!” — you will be misled.
How was the condition defined, and who made the diagnosis?
In the cold remedy study, what if some of the people in the study had colds, but others had allergies and others had a sinus infection? That would really muddy the results. This is especially pertinent in studies of remedies for reading disorders. You need to look for how the research defined reading disorders (for example, did the subjects all have developmental dyslexia or did some have that and others have secondary dyslexia, due to inadequate IQ or inadequate instruction and did some have bad ADHD that made them unable to pay attention long enough to read, etc.), and who made that diagnosis. Only then can you know whether the remedy being reported is even treating the condition the authors say it’s treating.
How were success and failure defined in the study, and how was that determination made?
In the case of reading remedies, was success defined as an improvement in the rate of reading? Fluency? Comprehension? Something else? How much improvement was considered a success? 1%? 10%? 50%? Was the person making the assessment qualified to make the assessment? Did the subjects and/or the evaluators know which treatment they were getting? Subjects who know (or believe) they are receiving treatment will perform differently, and an evaluator’s assessment of the outcome will unavoidably be biased if he or she knows what treatment the subject received. Ideally, both the subject and the evaluator are “blind” to the treatment, meaning that neither knows whether the subject received the ‘real’ treatment.
Were the people in the study receiving any therapy other than the therapy that is being studied?
In a study of new cancer treatment — call it Treatment X — if the patients are getting Treatment Y at the same time they’re getting Treatment X, then how can you know whether the effect was due to Treatment X (as the authors of the study may claim) or Treatment Y? You can’t. If a group of children with dyslexia is given vision therapy and also given specific reading instruction, and after six months they read better, the authors may claim that the study shows that vision therapy helped their reading, when in fact it may have been the reading instruction itself that accounted for the improvement.
Does the author of the study have a financial interest in the outcome?
If the author has a financial interest in the outcome, it does not prove that the study is invalid — there could certainly be very valid studies, rigorously designed and conducted, that produce absolutely unbiased results. However, if the author has a financial interest, you have to interpret the claims of the study with even greater caution than usual, looking for evidence of bias, because now it matters to the author which way the study turns out.
To summarize (and thanks for your patience!):
- Don’t take someone else’s word that therapy works; learn to look critically at the evidence yourself.
- Resist the urge to rely on anecdotal evidence (testimonials); hold out for stronger evidence.
- Determine how the authors defined successful treatment, and how success or failure were established.
- In assessing studies regarding remedies for reading disorders, always ask: a) Where’s the control group? b) How did they define “reading disorder,” and who diagnosed it? c) Were the children in the study receiving any other treatment (such as reading instruction) besides the vision therapy, tinted lens, etc. and d) Did the author(s) have a financial interest in the outcome of the study?
Finally, keep this in mind: the burden of proof is not on those of us who would advise against controversial therapies such as those listed in “Five Ways.” Instead, the burden of proof is on those who would have parents pursue those therapies. As I have tried to demonstrate, the proof is not there.
I hope you find this helpful, and I hope this information prepares you to ask better questions as you’re considering your options for addressing your child’s struggle with reading, writing, and spelling.
Dr. Young has no financial interest in any dyslexia evaluation or remediation program, method, or company, including Lexercise.