The value of objective research.

Medical research helps ensure that we provide the best possible care for our patients. When we try to interpret data without an objective comparison group, our senses often lead us astray. Some of the examples of misconceptions about children with autism illustrate this very well.

Disclaimer: The examples discussed below illustrate why careful research is needed. The emphasis is on statistical design issues rather than autism itself. As additional research is done, the medical community's opinion about some of the theories discussed below may change.

What is autism?

Autism is a general term used to define a wide range of disorders (including core autism, atypical autism, Asperger's Syndrome, Childhood Disintegrative Syndrome, and Rett's syndrome). These disorders typically have some of the following components:

• severe delays in language development
• problems with social interactions
• abnormal responses to sensations
• marked restriction of activities and interests

Patients with autism vary widely in the severity of these disorders and the degree to which they are affected. Some are totally uncommunicative, others have very limited communication ability, and others have close to normal abilities. Some patients require constant supervision, and others can hold down jobs and live on their own.

Most cases of autism begin in early infancy, although the condition is most frequently diagnosed around one or two years of age, when the deficits in language and social interaction become more obvious. In some cases, however, infants appear to develop normally, then suddenly lose many of their communication and social interaction skills.

A diagnosis of autism can have devastating psychological consequences for a parent. It is hard for any parent to be objective about the condition that affects their children, but it must be especially difficult for the parent of an autistic child.

There have been many misconceptions about autism. The misconceptions developed through poorly designed research that lacked a fair comparison group. Good research with objective comparisons has driven out these misconceptions.

Misconceptions about bad parenting

A Freudian view of the cause of autism was popular in the 50s and 60s. Bruno Bettelheim claimed in two books, The Empty Fortress and Infantile Autism, that parents who are cold, distant, and unaffectionate--caused their children to develop autism.

[T]he precipitating factor in infantile autism is the parent's wish that his child should not exist. --Bruno Bettelheim.

These researchers often placed most of the blame on the mothers and characterized them as "refrigerator mothers." The development of this theory was based on weak research with no good comparison group.

• Parents were evaluated after the fact. The limited communication and social interaction skills of the autistic child may have caused the parents to appear cold and unaffectionate, rather than the coldness and lack of affection causing the autism.
• Researchers were aware of the diagnosis of autism and might tend to over-interpret minor
• No comparison groups were used: either parents of normal children or other siblings of parents of autistic children.

The belief that bad parenting was the cause of autism led to tragic consequences. Parents had to deal with a double whammy, the diagnosis of autism in their children, and the accusation that they were the cause of this condition. In many cases, parents lost custody of their children.

Careful research has demonstrated that bad parenting could not be the true cause of autism. There were several comparisons that led to this finding.

First, the siblings of autistic children are almost always unaffected. While it is possible for a parent to be cold and unaffectionate with only one child, you would expect it to be far more common that a parent would be cold and unaffectionate towards each child.

Second, the rate of autism is higher among boys than girls. There is no evidence that parents tend to be less affectionate with boys rather than girls.

Third, there are subtle, though often unnoticed, manifestations of autism from the time of birth, before parenting could have had any impact.

Thanks to objective research, we no longer believe that "refrigerator mothers" cause autism.

Misconceptions about facilitated communication

Facilitated communication is the use of an assistant to help a non-verbal individual to type messages on a keyboard. Rosemary Crossley founded the DEAL Communication Centre in Melbourne, Australia to train autistic children and others with severe communication disorders to use facilitated communication and other alternative means of communications.

I'd only seen one child before 1986 who'd been diagnosed as autistic, a breathtakingly cute hyperactive seven-year-old with appalling behavior. His mother was a friend, and she'd asked me if I could help him to communicate (he was totally mute). I'd worked with Jonathon for several years, and by the time DEAL opened he was able to type short sentences (although his willingness to do so was very variable indeed). Initially I'd held his hand while he typed (partly to stop him running away) but he had gradually moved beyond that, although he still generally needed a hand on his shoulder to maintain focus. Having only seen one of him, however, I didn't know whether his problems were typical of children diagnosed as autistic. The word got around, though, and after DEAL opened we were approached by a number of parents asking if we would see their autistic children and try and find some way for them to communicate. -- Rosemary Crossley, quoted at http://home.vicnet.net.au/~dealccinc/Spaut1.htm.

What appears to be communication, however, is just the autistic child picking up on subtle and possibly subconscious cues from the facilitator. The belief that facilitated communication represented actual communication with the autistic child was based on weak research with no good comparison group.

• the individual facilitators would often be the ones who would evaluate the success of their facilitation
• there was often no formal assessment of language skills prior to the initiation of facilitated communication

The belief in facilitated communication has led to tragic consequences. Parents and guardians of autistic children have been accused of child abuse through the use of facilitated communication.

Careful research has demonstrated that the messages in facilitated communication originate from the facilitator and not from the autistic child. This research had several comparison groups.

First, sessions where the facilitator knew the information being asked of the patient were compared to sessions where the facilitator did not know, and to sessions where the facilitator was misled about the information. In this research, the success of the technique rose and fell in sync with the amount of information given to the facilitator.

Second, communications skills after facilitator training were compared to communication skills before facilitator training. Some of the successful cases of facilitated communication had actually represented patients who had the ability to communicate non-verbally prior to the intervention. When communication skills were measured before therapy started, the success of facilitated communication was no better than the baseline skill level.

Finally, communication was compared with different facilitators. There was little agreement among the responses provided by different facilitators.

Thanks to objective research, the use of facilitators to communicate with autistic children has been discredited.

Misconceptions about vaccines

In a 1998 Lancet journal article, Andrew Wakefield et al noted an association between autism and the MMR (Measles, Mumps, Rubella) vaccine.

BACKGROUND: We investigated a consecutive series of children with chronic enterocolitis and regressive developmental disorder. METHODS: 12 children (mean age 6 years [range 3-10], 11 boys) were referred to a paediatric gastroenterology unit with a history of normal development followed by loss of acquired skills, including language, together with diarrhoea and abdominal pain. ... FINDINGS: Onset of behavioural symptoms was associated, by the parents, with measles, mumps, and rubella vaccination in eight of the 12 children, with measles infection in one child, and otitis media in another. All 12 children had intestinal abnormalities, ranging from lymphoid nodular hyperplasia to aphthoid ulceration. --Wakefiled et al. Lancet 1998 Feb 28;351(9103):637-41

Although the authors do emphasize the preliminary nature of their findings, they discuss the possible link between MMR vaccination and autism at length and conclude in their final paragraph that

We have identified a chronic enterocolitis in children that may be related to neuropsychiatric dysfunction. In most cases, onset of symptoms was after measles, mumps, and rubella immunisation. Further investigations are needed to examine this syndrome and its possible relation to this vaccine.

The link between the MMR vaccine was based on weak research without a good comparison group. Problems with this research include:

• Patients were referred to the clinic and are not typical of any population of autistic children.
• There was no comparison group, either of unvaccinated children, or of children without autism.
• Onset of symptoms was assessed subjectively by either a parent or the child's physician.

This belief has led to tragic consequences. Public acceptance of the MMR has declined leading. Lower vaccine coverage has led to an increased possibility of an outbreak of these diseases.

Careful research has shown that there is no link between the MMR vaccine and autism. This research used large carefully defined groups of autistic children and included objective comparisons.

The first comparison involved autistic children diagnosed before and after 1987, the year that the vaccine was widely distributed. Although the cases of autism did rise over time, there was no sudden surge of cases in 1987.

The second comparison looked at autistic children who receive the MMR vaccine early in life are compared with autistic children who receive the MMR vaccine later in life. For these two groups, there was no difference in the onset of symptoms.

Thanks to objective research, there is substantial assurance that there is no link between the MMR vaccine and autism.

Further reading

1. A History of Facilitated Communication: Science, Pseudoscience, and Antiscience: Science Working Group on Facilitated Communication. Jacobson JW, Mulick JA, Schwartz AA. American Psychologist 1995, Vol. 50, No. 9, pp. 750-765. http://www.apa.org/journals/jacobson.html
2. Facilitated Communication and Autism. Separating Fact from Fiction. Smith MD, Belcher RG. http://www.csaac.org/pub-fac.htm
3. Autism and measles, mumps, and rubella vaccine: no epidemiological evidence for a causal association. Taylor B, Miller E, Farrington CP, Petropoulos M, Favot-Mayaud I, Li J. Lancet 1999; 353: 2026-2029.

Category: Statistical evidence