This blog post has been a while coming, because I wasn’t too sure I wanted to write it. I’m not particularly interested in critiquing case studies, since all they really show is promising leads for future systematic research, without interesting methodological issues for me to get my teeth into. I particularly don’t like critiquing case studies of named non-anonymous individuals, especially children, and especially where there’s a ‘happy ending’ and maybe it would be more comfortable for everyone concerned to just leave things be – but this case has been publicised heavily by the people involved & is the basis of one of the most popular introductory books on homeopathy. My Mum took me to a homeopath for a year or so, & whilst her prescriptions didn’t seem to do anything, she was friendly & would have made a good counsellor, & the treatment wasn’t unpleasant. There isn’t medication for most Specific Learning Difficulties (SpLDs) or Autistic Spectrum Disorders (ASDs) & I’ve not heard of homeopaths suggesting that people should stop using behavioural support, social skills training, and other psychological interventions, & homeopathy doesn’t have to take a long time or be distressing for the child, so it’s not doing much harm.
This post isn’t about homeopathy, it’s about the limitations of single case-studies, & like much of the blog about how to read & understand the evidence. As with all case-studies all I can really do is offer alternative explanations for what I know of the case – which doesn’t prove anything any more than the original explanations did. But for such a heavily promoted case history, people should have access to relevant information to make up their own minds.
With that out of the way, here’s the case I’m looking at: A claim by John Melnychuk (http://www.paloaltohomeopathy.com/homeopathy.html) to have ‘cured autism’. In particular Amy Lansky’s son in her book ‘The Impossible Cure’ (http://www.impossiblecure.com/). [I’m not too comfortable with using a child’s full name on the Internet, even though it’s published in the book and a lot of personal information is on their website, so I’ll stick to calling him ‘M’]. Unfortunately I’ve not been able to get hold of the book in the UK & can’t afford a copy from the US – if anyone would like to loan me one for a month or so I’d be most interested, & if you have read the book & can provide more information & clarification that would be most welcome. Meantime the fullest account I’ve been able to find is here: http://www.renresearch.com/ and here: http://www.renresearch.com/autism.html .
In outline the story runs something like this:
- 1991 M born
- ‘by age two, he was still not speaking. He knew the entire alphabet, could count to twenty, and could stack blocks like nobody’s business, but he did not know more than about 10 words’ ‘When he was two, we enrolled him in a 2-day/week 2-hour preschool program. At about this time, our nanny suddenly left without notice … Our new nanny was kind but, in retrospect, detached emotionally and probably did not engage with M very much.’ ‘he could not sit at circle-time unless held in someone’s lap; he would tend to wander off to other places in the classroom or to other classrooms; lowered eye contact; self-stimulation activities such as spinning; and unresponsiveness to questions. At the same time, he clearly showed signs of great intelligence. If he was engaged in something that interested him, he could sit for very long stretches with intense concentration. He could build complex structures with amazing skills of symmetry and balance; he also showed sophisticated strategic and dexterity skills at computer games. At age 2.5, when our usual nanny was gone for a month and a far more engaging replacement nanny stepped in, M finally began to talk more. It was single words, but it was a start.’
- ‘As M approached his third birthday … we opted against having M tested at a highly-medicalized child development center attached to the university. Instead, we took him to see a respected speech and language therapist’ ‘Donna did not label M as autistic or as anything at all’ ‘we instituted several other changes’ ‘At age three, he finally built two-to-three word sentences. He was clearly more “present” than before’
- 1994 ‘developmental/behavioral/language problems’, ‘mild autism, probably as a result of vaccine damage’, ‘M was making slow progress’ with a variety of interventions including speech & language therapy’ ‘1994, M continued with his speech and language therapy and, after testing, qualified for special education benefits’ ‘The teachers felt he was bright and merely a quiet child’ ‘M continued to make slow and steady progress in his speech therapy’ ‘echolalia’ ‘He could watch things on TV and perfectly imitate nuances in various characters’ behavior and mannerisms’ ‘M also showed an uncanny musical, dancing, and acting ability’
- Jan 1995 ‘He began [homeopathic] treatment at age 3.5’ ‘During the interview, he got very upset because he didn’t successfully write “Mom” on a piece of paper the way he wanted to’ ‘for about 1.5 years’, ‘Two years later, you would never suspect he had been autistic’
- ‘By the time he was five … For example, his language production continued to be awkward at times. In times of stress (e.g. if he was sick), he would still retreat into himself and utilize echolalia as a speech strategy. He was also still a bit hyperactive at times, not being able to sit in his seat through an entire meal.
- Even now, at age six, all of this is still true of M, though to a lesser degree.’ – ‘1996 … he still has ways to climb to full recovery’ ‘M showed some decline in speech’ – ‘1997 … I am noticing a subtle decline in speech’
No doubt I’ll be accused of taking these out of context, so I really would suggest that you read the full story as linked to – I’ve cut & pasted the bits that jump out given my interest in developmental psychology.
First, a look at diagnosis. I don’t know much about how ASD are diagnosed in the USA, but in the UK then health visitors (~ community nurses who care for children & families in the gap between midwife & school nurse) will administer the Checklist for Autism in Toddlers, at 18 months (details here: http://www.autism.org.uk/nas/jsp/polopoly.jsp?d=1419&a=2226). This is not a diagnostic test, it is a screening tool to ‘flag up’ children ‘at risk’ for further evaluation, and as such will tend to be overinclusive – it is better for a child to be wrongly spotted as ‘possibly’ having an ASD & be sent for testing than for a diagnosis to be missed. It is difficult to make a definite diagnosis in young children, as ASD are diagnosed on the basis of social impairment, and typically developing young children may not be able to reliably demonstrate social skills, especially when faced with a stranger doing tests. The earliest sign of an ASD is often difficulty with ‘joint attention’ – child will not draw an adult’s attention to interesting things. Full diagnosis of ASD are made on the ‘Triad of impairments’ in social interaction, communication, & imagination. The DISCO & ADI assessments, structured clinical interviews which last for hours & involve both the person being diagnosed & a caregiver, which *can* be used for formal diagnosis, tend to be used only in children older than 3 (DISCO isn’t validated for under-3s, ADI has more questions after ages 3+, 4+ 5+). Although there is *fascinating* work going on into screening & diagnosis as early as the first year, using tests based on things like attention shifting & sensory processing, as yet these aren’t specific or researched sufficiently to be used in clinical practice. There’s more on types of screening test here: http://www.autism.org.uk/nas/jsp/polopoly.jsp?d=1419&a=3280 .
Patterns of ASD diagnoses are an interesting thing. With the idea that ASD are a spectrum has come the idea of the ‘broader phenotype’ – basically people who share some characteristics of ASD, but not enough so (or function too well for) a diagnosis. There’s some interesting ideas about emphasising/systemising (decent newspaper article here: http://www.guardian.co.uk/print/0,3858,4649492-111414,00 , academic intro here: http://www.autismresearchcentre.com/research/project.asp?id=6 ) the idea being that people with ASD will be on the far end on ‘systemising’, but that everyone will be somewhere on the line of being more interested in people vs being more interested in things (this is a huge oversimplification, I’d really suggest looking at the Autism Research Centre which will point you to useful papers & explanations: http://www.autismresearchcentre.com/research/percog.asp). [Incidentally, I was experimented on for some of the early research into emphasising / systemising whilst at 6th form – and yes, Prof. Baron-Cohen really is ‘Ali G’s’ cousin, but he is the stereotype of a quiet Cambridge academic, & gets upset if you mention it].
There’s also the idea of ‘assortative mating’ – with increasing social & geographical mobility & greater mate choice, high-systemisers are more likely to end up having children together, and those children are more likely to show high-systemising traits associated with ASD. This is supported by findings such as more parents & grandparents of children with ASD being engineers, and ‘clumps’ of ASD diagnoses around places like Cambridge & ‘Silicon Valley’, where many people work in technical occupations where being high-systemising is an advantage (this article might be interesting, though it gets a few bits wrong: http://www.wired.com/wired/archive/9.12/aspergers_pr.html , if you have ATHENS access try: http://adc.bmj.com/cgi/content/full/91/1/2 ). Both Amy Lansky and her husband have PhDs in CompSci & technical careers, their other son is studying computing at Stanford, and according to information on the family’s home page M enjoys computer animation & attends a school where ‘Education in mathematics and computer science is particularly strong’. Being a high-systemiser can have many advantages, and does tend to come with ‘broad phenotype’ ASD traits – Amy Lansky mentions things like the ability to concentrate for long periods and excellent visual search pattern-matching skills. These things would suggest a genetic rather than vaccine-related cause for M’s difficulties, and also suggests that M is flourishing in an environment which values systemising skills – no bad thing, but also an environment likely to bring out the skills and minimise the weaknesses of someone with a broad ASD phenotype.
The ‘broad phenotype’ of ASD refers to the likelihood of relatives of people with ASD to have sub-clinical ASD traits – not quite far enough along the spectrum to be diagnosed or to cause problems, but fairly subtle differences which might be picked up with psychological tests designed to measure traits associated with ASD such as field-independence (the ability to pick a small part out of a complex pattern) or rapid attention shift, or maybe somebody who has subtle difficulties with interpersonal functioning or ASD-like patterns of developmental delay, strengths & weaknesses, but by definition are able to function well enough to avoid a full diagnosis. People with a ‘broad phenotype ASD’ aren’t always going to be significantly socially impaired, particularly if they are intellectually able to compensate for social difficulties, or if they have a supportive environment which values their strengths (I’d probably qualify for ‘broader phenotype’ myself with several diagnoses of ASD in the family, & recieved educational & behavioural support when younger for it). Given a supportive family, lots of attention, speech & language therapy, and a social & educational environment building on what he’s good at, there’s no reason why M couldn’t have overcome early difficulties to reach the same level as his peers. I’m delighted he seems to be doing so well, but homeopathy isn’t necessary to explain this.
M apparently started homeopathic treatment age 3 1/2, without any mention of a formal ASD diagnosis by any professional at that time (is this in the book?). I would be wondering if M scored above the cut-off on a screening test, which would indicate a need for further investigation but might suggest to worried parents that he may have ASD, but wasn’t actually fully diagnosed. Before & during treatment M was receiving speech therapy, extra parental input, spent time at a Montessori school (where the structured environment may be easier for children with ASD), and apart from speech seemed to be developing at or ahead of schedule in many areas (eg writing, knowing the alphabet). The only thing that really stands out from the accounts is possible language delay, remedied by lots of formal & informal behavioural support. So just how language-delayed was M anyway? As a 2-year-old, he knew ‘about 10 words’, as well as the alphabet & counting to 20 – which I reckon is maybe 50 ‘words’ in all. 50 words is around what you might expect for a typically developing child of maybe 18-24 months. Between 2-3 years, you’d typically get a ‘language explosion’ where children suddenly acquire new vocabulary daily, with sentence length increasing from around 1 to 3 words over the course of the year. M ‘began to talk more … single words’ at about 2 1/2 years, with a ‘more engaging’ nanny. ”At age three, he finally built two-to-three word sentences’. This was when M was taken to a speech therapist. At 3 1/2, M could write. At four years old, ‘The teachers felt he was bright and merely a quiet child’. [Age 5 1/2] ”you would never suspect he had been autistic’. There is some suggestion of subtle ongoing speech problems, but not clearly described & sub-clinical. The Lanskys are clearly an academically able family. I would wonder if perhaps M’s language development was never so far off the normal developmental trajectory, but perhaps slower than other family members, giving them cause for concern. For more on speech milestones see: http://www.childdevelopmentinfo.com/development/language_development.shtml, http://www.nidcd.nih.gov/health/voice/speechandlanguage.asp#mychild
Diagnosis by Internet is a really stupid idea, even if I was qualified to do so, which I’m not. However, I haven’t been able to find anything from Amy Lansky’s accounts on the Internet which suggests that M was ever that far from a typical development curve. There’s no accounts of professionals other than Melnychuk who saw M having diagnosed him with an ASD at the time, and teachers who have knowledge of typical child development thought M was ‘a quiet child’. I’m not, repeat not, accusing Amy Lansky of lying. ASD are complicated & and suspecting your child might have one must be scary. Being told by Melnychuk, who according to his own website has no qualifications in conventional medicine, psychology, or fields other than homeopathy, that one’s child is ‘autistic’ must be distressing, particularly given that other case studies of autism provided by Melnychuk do not suggest that he recommends behavioural therapies to help manage symptoms of ASD, or supportive counselling for distressed & suicidal parents. Giving M homeopathy was followed by developmental progress (as you’d expect unless homeopathy was harmful – typical children show developmental progress, often stepwise), and giving M homeopathy when he was having difficulty was followed by progress (again, it is to be expected that children will progress slowly at times & faster at other times, and doing anything not actively harmful during times of slower progress will be followed by faster progress).
The problem with case studies is that they can’t usually prove or disprove much. So, how about research going beyond case studies? The only hit I can find on Medline for the use of autism in homeopathy is for ‘homeopathic secretin’ in the British Homeopathic Journal, which didn’t seem too relevant & found a non-significant worsening of symptoms during treatment. (Robinson, T.W. 2001. Homeopathic Secretin in autism: a clinical pilot study. Br Homeopath J. 90(2):86-91).
A Cochrane metanalysis of four studies of homeopathy for ADHD found ‘The forms of homeopathy evaluated to date do not suggest significant treatment effects for the global symptoms, core symptoms of inattention, hyperactivity or impulsivity, or related outcomes such as anxiety in Attention Deficit/Hyperactivity Disorder.’ (Coulter MK, Dean ME. Homeopathy for attention deficit/hyperactivity disorder or hyperkinetic disorder. Cochrane Database of Systematic Reviews 2007, Issue 4. Art. No.: CD005648. DOI: 10.1002/14651858.CD005648.pub2) – seems to be free to the public at http://mrw.interscience.wiley.com/cochrane/clsysrev/articles/CD005648/frame.html , so please read the evidence for yourself & make your own mind up.
I guess I’m going to get a lot more stick from this post than most things I write. Again, this isn’t about homeopathy, it’s about how to read & interpret a case study, so people can evaluate the evidence for themselves. If anyone can lend me a copy of ‘Impossible Cure’, or knows anything relevant to the case that I’ve not picked up online, I’d be most interested – thanks.
This case study is a pleasure to read in many ways. M is obviously a happy, healthy young boy, who has been given every opportunity to thrive & seen through any difficulties by parents who care about him very much. It’s a success story in terms of appropriate early intervention & a supportive environment in enabling children with specific learning needs to not just cope but find their ‘niche’ and thrive, not just in spite but even because of their differences. There should be more stories like this. It should be just as ‘OK’ to come under the ‘broader phenotype’ of autism as to be very chatty or artistic or athletic – it’s equally within the realms of normal human variation and should be equally encouraged & accomodated. But this case does not show that homeopathy made M ‘normal’, and it’s not fair to people who struggle with more severe ASD symptoms to pretend that there is an easy way to ‘fix’ them, even if that was a good idea.