A remedy without rigour?
Here is a paper that is just about to be given to the British Dyslexia Association," said an email sent to the TES news desk. The message, from David Reynolds, Exeter university professor and author of the paper, continued: "It suggests that a cure for dyslexia may have been found, and I am absolutely certain of my data."
That paper and others by Professor Reynolds have helped to boost a multi-million pound expansion of a new treatment for dyslexia. It is an approach that was unheard of five years ago but is now sweeping across Britain, the United States and Australia. But it is also being pursued by academics who question what, if any, curative effect it has.
The expansion of dyslexia, dyspraxia attention deficit treatment (DDAT) dates from its appraisal by broadcaster Sir Trevor McDonald on prime-time ITV1 in 2002. Sir Trevor called it a "revolutionary breakthrough in the treatment of dyslexia".
But that view was not shared by the Independent Television Commission, which said the claims were "not sustainable", a judgment it repeated about a later item on DDAT on Channel 4's Richard and Judy.
Some 275,000 viewers asked for more details about DDAT after the ITV1 broadcast. DDAT websites claim it was a turning point in the treatment's reputation. They also emphasise and reproduce Professor Reynolds'
"independent studiesI (which) show remarkable progress in reading, comprehension, writing, social skills, self-esteem and co-ordination".
DDAT is based on the hypothesis that the cause of dyslexia is a fault in the area of the brain which controls balance and muscle movement: the cerebellum. It suggests that by stimulating the cerebellum with exercises, new neural connections will be formed and the brain will perform better in all activities, including reading and writing.
The theoretical jump from deficit to treatment is a controversial one and the only way to test it is on those with dyslexia. But it is not a new jump: other exercise-based programmes have claimed success for some years in the UK, the US and Sweden.
Professor Reynolds claims to have proven DDAT's success in a single experiment, described in four papers spread over two years.
The experiment took place at Balsall Common primary school in Solihull in 2001. The school's weakest readers - 35 pupils from Years 3, 4 and 5 - were given a well-known dyslexia "risk test" designed by Rod Nicolson, professor of psychology at Sheffield university.
Professor Nicolson, who for several years has been a paid consultant for DDAT's founder, the millionaire Wynford Dore, worked with Professor Reynolds on the analysis of the Balsall Common results. Professor Reynolds, formerly a paid director of another Dore company, told The TES that he now has "no financial connection of any kind with Wynford Dore".
Initially, half of the 35 Balsall Common pupils received the DDAT treatment, in which exercises are done at home in twice-daily, 10-minute sessions using equipment that includes a balance board and beanbags.
Six months later, the other half of the pupils did the same. The risk rating for each child was retested every six months over the following 18 months. Reading and annual national curriculum test results were also compared.
Professors Reynolds and Nicolson say the results were very encouraging. Not only did children's bead-threading and balance skills improve (not surprising, since they had practised those skills) but their literacy skills, which were not being directly exercised, also improved.
The papers said children's dyslexia risk fell substantially, particularly for those whose risk was higher originally. They also claimed that pupils'
reading skills improved in the dyslexia test, that their reading progress accelerated to near-normal and that their standardised and national curriculum tests in reading, writing and comprehension showed big improvements.
Professor Nicolson called the results "highly encouraging". They might, said Professor Reynolds, "bring us very close to a cure".
But critics challenge the experimental basis of DDAT and say it is flawed.
In an acrimonious exchange of papers, mostly published in the British Dyslexia Association's journal Dyslexia, most of Britain's foremost academics have argued that the Balsall Common results prove nothing at all.
They argue that the design and analysis of the experiment were so unscientific that the findings are unreliable.
Professor John Rack, head of research and assessment at the Dyslexia Institute, says: "I don't think they are seeking to disprove a hypothesis.
They seem to me to be just seeking further evidence to prove what they already believe."
DDAT's critics make the following points:
* The Balsall Common children were not dyslexic. Their scores on the Nicolson "risk test" ranged from "mild" and "moderate" to "severe". Some lagged only a few months behind their expected reading ages; some were even ahead of them. In this case, some academics argue that these pupils cannot reliably be used to test a treatment for dyslexia.
* The experiment was not a "fair test". No one checked whether all the children did the exercises in the same way. There was no control group being given an alternative or placebo treatment to make sure that improvements in the treated group were not simply due to maturation, getting extra help at school or being given special attention and encouragement.
* While 35 pupils began the experiment, as children left the school the number fell to 29. By the time of the final test results on which claims of success hang, the number of pupils had dropped to 22. Even 35 would not be a reliable number, says Carol Fitz-gibbon, emeritus professor at Durham university. She says the sample group was "disgracefully small".
* For such robust claims to be justified, children would need to be tested before and after the experiment - with the same test and by a tester unknown to them. Teachers are notorious for giving pupils the benefit of the doubt.
Most of the "proof" of pupils' improvement in the Reynolds papers is based on standardised reading-test scores marked and administered by Balsall Common teachers. The scores were taken from what the papers called Sats but which were in fact optional tests published by the Qualifications and Curriculum Authority and which inevitably varied from year to year.
But Professor Nicolson argues that these criticisms are unfair. He says:
"People have misinterpreted what was said. They were saying it should have been a clinical-type trial, whereas we were clear it wasn't a clinical-type trial. It was a value-added study."
Professor Reynolds believes the dyslexia establishment has ignored real results. "All treatments involve some kind of expectancy effect," he says.
"But if you look at the Balsall Common Sats results and the reading test, the treatment seems to offer more.
"Even in 20023, where no treatment is happening, these children still carry on improving more than their peers."
Other academics argue that strict laws control the testing of any drug claiming to be a cure before it is allowed on the market. That protocol includes replication: one experiment with one set of results is not enough.
Several bodies, including the Dyslexia Institute, have recommended that the exercises in DDAT be shared so that they can be tested objectively in a clinical-type trial with proper sample quality and control groups.
Professor Rack says: "What we need is a controlled study, then we could give some sensible advice to teachers and parents," He and others say that when they have asked for details of its exercises, DDAT has refused to release them on the grounds of "commercial confidentiality".
Meanwhile more and more parents are lobbying local education authorities to pay for DDAT treatment, says Professor Margaret Snowling of York university.
She believes that many parents are desperate and are being offered false hope.
"Dyslexia is hard to treat," she says. "It's a real struggle. I find it unbelievable that anyone thinks this is something which can be generalised to children who have very serious disorders."
'AMAZING RESULTS THAT WILL STAND UP TO ANY SCRUTINY'
Since the first DDAT clinic opened in Kenilworth, Warwickshire, in 2000, seven more have been launched in the UK - in Fulham, Cardiff, Southampton, Edinburgh, Sheffield, Manchester and Bedford.
There are five centres in the US and two in Australia, and more are expected to follow in Canada and New Zealand this year.
Some 17,000 adults and children have been treated so far, 90 per cent of them successfully, according to DDAT. Each pays fees of around pound;1,500.
DDAT's founder is Wynford Dore, a millionaire businessman who left school at 14 and whose dyslexic daughter attempted suicide. None of the companies - British, American or Australian - has made a profit, says Mr Dore. He adds: "I have poured millions into this. This is not a money-making business. I am desperate to stop another generation suffering as my daughter did."
Mr Dore refused to provide details of his firms' turnover to The TES. Proof of DDAT's efficacy comes not only from the Reynolds and Nicolson research, which he says is "totally independent", but also from "lots of studies".
He adds: "Many confidential, independent studies by local education authorities are going on. We have an army of wonderful medical researchers working on this."
He believes pique is behind the dyslexia establishment's criticism. "The fact that some professors don't like this research doesn't make it unscientific," he says. "I have pleaded with the Dyslexia Institute and the British Dyslexia Association to do research. I have never put restrictions on them. We are getting amazing results that will stand up to any scrutiny."