Overdiagnosis in Some Cases?

Uncertainty is a major theme of The Politics of Autism.

At Autism, Stephen J. Blumberg and colleagues have an article titled "Diagnosis Lost: Differences Between Children Who Had and Who Currently Have an Autism Spectrum Disorder Diagnosis." The abstract:

Autism spectrum disorder diagnoses sometimes change due to misdiagnosis, maturation, or treatment. This study uses a probability-based national survey—the Survey of Pathways to Diagnosis and Services—to compare currently diagnosed (n = 1420) and previously diagnosed (n = 187) children aged 6–17 years based on retrospective parental reports of early concerns about their children’s development, responses to those concerns by doctors and other healthcare providers, the type of provider who made the first autism spectrum disorder diagnosis, and the autism spectrum disorder subtype diagnoses received (if any). Propensity score matching was used to control for differences between the groups on children’s current level of functioning and other current characteristics that may have been related to diagnosis loss. Approximately 13% of the children ever diagnosed with autism spectrum disorder were estimated to have lost the diagnosis, and parents of 74% of them believed it was changed due to new information. Previously diagnosed children were less likely to have parents with early concerns about verbal skills, nonverbal communication, learning, and unusual gestures or movements. They were also less likely to have been referred to and diagnosed by a specialist. Previously diagnosed children were less likely to have ever received a diagnosis of Asperger’s disorder or autistic disorder.

From the article:

Overdiagnosis may also occur if ASD diagnoses are substituted for other learning disabilities because of the availability of greater resources for treatment of ASD (Shattuck, 2006). Parents in this study confirmed that some children who reportedly did not have ASD—approximately one in four who lost their ASD diagnosis or 3% of all school-aged CSHCN ever diagnosed with ASD—were given the diagnosis due to their need for services.

Shattuck PT (2006) The contribution of diagnostic substitution to the growing administrative prevalence of autism in the US special education. Pediatrics 117(4): 1028–1037

 That observation is consistent with The Politics of Autism,  where I write:

Once parents get over their initial denial, they have strong reason to seek a diagnosis of ASD for troubled children, since it can trigger thousands of dollars in assistance. Allen Frances says: “Having the label can make the difference between being closely attended to in a class of four versus being lost in a class of 40. Kids who need special attention can often get it only if they are labeled autistic.” Roy Grinker quotes a leading scientist who also has a clinical practice: “I am incredibly disciplined in the diagnostic classifications in my research, but in my private practice, I'll call a kid a zebra if it will get him the educational services I think he needs.” 

The article has some sobering thoughts on recovery:

Few children ever diagnosed with ASD were said to have actually had and recovered from ASD. Only 21% of children with a lost diagnosis (or less than 3% of all schoolaged CSHCN ever diagnosed with ASD) were said to have lost it due to treatment or maturity. The true rate of recovery may be even smaller if some of these children were incorrectly thought by their parents or doctors to have lost their ASD because they now have well-developed coping skills and other strengths.

The authors appropriately note serious limitations to the study:

The impact of selection biases resulting from sampling, nonresponse, and lack of coverage of non-English speaking households and households without telephones is unknowable, but they are less likely to influence the magnitude of comparisons between matched pairs than absolute estimates of a population prevalence. Still, due to the low response rate, generalizing the results beyond the sample should be done with caution.

The absence of Spanish-only respondents is a serious limitation indeed, since underdiagnosis seems to be a problem among Hispanic families.

More broadly, such a study does not capture the flip side:  people who were incorrectly labeled as not having ASD.  As I write:

Bias can run in the other direction, too. If a school district does not really want to provide services, its psychologists might feel pressure to code behavior in a way that spells “normal.”

Losing the Diagnosis, but Not the Problems

A release from the American Academy of Pediatrics:

About one in 14 toddlers diagnosed with autism spectrum disorder (ASD) no longer met the diagnostic criteria in elementary school, but most continued to have emotional/behavior symptoms and required special education supports, according to a study to be presented Sunday, April 26 at the Pediatric Academic Societies (PAS) annual meeting in San Diego.

Previous studies have shown that ASD symptoms resolve in some children over time. It is not clear, however, if these children continue to have cognitive, behavioral or learning deficits.

Researchers, led by developmental pediatrician Lisa Shulman, MD, reviewed data on 38 children diagnosed with ASD in 2003-2013 whose symptoms had resolved when they were re-evaluated about four years later. The children were among 569 children living in the Bronx who had been diagnosed with ASD by a multidisciplinary team at a university-affiliated early intervention program.

The children came from racially, ethnically and socioeconomically diverse backgrounds, a population generally underrepresented in autism studies. Forty-four percent were Hispanic, 36 percent were Caucasian, 10 percent were African-American and 46 percent were on Medicaid.

Clinicians who made the original diagnosis also provided interventions and monitored response to treatment. Over time, they noted that ASD symptoms in some children resolved, but most continued to have other learning and emotional/behavioral symptoms needing attention.

"Autism generally has been considered a lifelong condition, but 7 percent of children in this study who received an early diagnosis experienced a resolution of autistic symptoms over time," said Dr. Shulman, director of Infant and Toddler Services and the Rehabilitation, Evaluation and Learning for Autistic Infants and Toddlers program at the Children's Evaluation and Rehabilitation Center/Rose F. Kennedy Center at Albert Einstein College of Medicine/Children's Hospital at Montefiore.

"The majority of the children at original diagnosis displayed intellectual disability but at the point of resolution of autistic symptomatology displayed normal cognition," Dr. Shulman added.

Although the social impairment of autism resolved and cognitive functioning (IQ) improved, researchers found that 92 percent of the children had residual learning and/or emotional/behavioral impairment. Only three of the 38 children had no diagnosis.
Language/learning disability was found in 68 percent, and nearly half had externalizing problems such as attention-deficit/hyperactivity disorder or disruptive behaviors. In addition, 24 percent had internalizing problems such anxiety, obsessive-compulsive disorder or selective mutism. Finally, nearly three-quarters of the children continued to require academic supports, such as a small class setting or resource room.

"When an early ASD diagnosis resolves, there are often other learning and emotional/behavioral diagnoses that remain," said Dr. Shulman, associate professor of clinical pediatrics at Albert Einstein College of Medicine and attending physician, Children's Hospital at Montefiore. "Understanding the full range of possible positive outcomes in this scenario is important information for parents, clinicians and the educational system."

Dr. Shulman will present "When an Early Diagnosis of Autism Spectrum Disorder Resolves, What Remains?" from 3:45-4 p.m. PT Sunday, April 26. To view the study abstract, go to http://www.abstracts2view.com/pas/view.php?nu=PAS15L1_2750.2

​This study was supported by a grant from the Children's Evaluation and Rehabilitation Center/ Rose F. Kennedy Intellectual and Developmental Disabilities Research Center.
- See more at: https://www.aap.org/en-us/about-the-aap/aap-press-room/pages/Some-children-lose-autism-diagnosis-but-still-struggle.aspx#sthash.bi4XtmYK.dpuf

Intervention and Recovery

At the Pittsburgh Post-Gazette, Mark Roth writes about early intervention, with an anecdote about 4-year-old Rocco Korch:

A 2008 survey of the scientific literature reported that between 3 percent and 25 percent of children diagnosed with autism eventually recover, and a study this year at the University of Connecticut rigorously tested 34 children who had been diagnosed with autism as toddlers and are now largely free of symptoms.

Rocco's story is typical in one other way, though. Most experts say that the earlier autistic children can begin therapy, the more progress they will make.

One of the best validated therapy techniques is the Early Start Denver Model, developed by Sally Rogers, now at the University of California, Davis, and Geraldine Dawson, who recently left her post as chief science officer at Autism Speaks and is now at Duke University.

In a 2009 study of their model, which features intensive one-on-one therapy in a naturalistic setting several hours a week, children in that group showed significantly greater gains in language and thinking ability over a two-year period than autistic children in a control group. In a follow-up study last year, researchers found the Early Start group also showed more brain activation when viewing faces -- a sign of responding to social cues -- while the control group showed more brain activity when viewing objects.

In an interview earlier this year, Ms. Rogers said the key to their approach is providing therapy in everyday settings the children would already be involved in.

In older behavioral therapy, she said, "if you wanted to teach the child the names of objects, you would figure out what that child is motivated for -- food, drinks -- and you'd put out a shoe and say 'Where's shoe?' and you'd give them a treat when they touched the shoe."

In the Early Start approach, "kids come in to therapy every day and when they take off their shoes, I would say, 'Where's that foot?' or 'Give me shoe?' " And at the end of the session I would say, 'Get Jason's sock; get Sally's shoe.'

A Story of Recovery

KTLA in Los Angeles reports on CARD and a story of autism recovery. (The station uses the less accurate term, "cure.")

Followup Story to "Optimal Outcomes" Study

The San Jose Mercury News writes of the "optimal outcomes" study that this blog has described:

There are no solid numbers showing what percentage of children diagnosed with autism eventually lose that diagnosis. But the idea that autistic children could recover began to gain traction in 1987 when UCLA psychology professor Ivar Lovaas said he saw a 47 percent recovery rate using intensive behavioral therapy. Many researchers, however, questioned whether some of the children in that and other studies truly had autism in the first place.

The new study, published in the Journal of Child Psychology and Psychiatry, has put these questions to rest, autism experts say.

For the study, a team of psychiatrists led by Deborah Fein of the University of Connecticut recruited 34 people who had been diagnosed before age 5 and had since lost their diagnosis according to the team's extensive interviews and behavioral observations. The team also solicited independent verification of the children's initial diagnoses.

Fein is quick to caution that the overwhelming majority of children with autism will not recover. "I've seen hundreds and hundreds of kids who got great therapy and excellent parenting," she said. "They all made progress, but very few of them reached that stage."

In general, she added, "it's very hard to predict who is going to respond rapidly to intervention."
Another unknown is how recovery comes about. Most families try several therapies, often several at once, making it difficult to tease out which are most important for producing optimal outcomes.

This report is more accurate than most.  But some psychologists would dispute the suggestion that the new study has ended all questions about Lovaas's findings:  some argue that the 47% figure was too high.

Misreporting the Study of Optimal Outcomes

A previous post described a study showing that some kids with ASD eventually lose the diagnosis.The study is available here.

The press has not always handled autism responsibly, and a number of news reports are inaccurately describing the study:

• BBC:  "While not conclusive, the study, in the Journal of Child Psychology and Psychiatry, suggests some children might possibly outgrow autism."
• Fox News:  "New research has found that some children diagnosed with autism actually ‘grow out’ of their symptoms – as well as their diagnosis – when they grow older, BBC News reported."
• US News and World Report: A new study published in the February issue of the Journal of Child Psychology hints that children diagnosed with autism could grow out of the disorder...The autistic children's original diagnoses were accurately reported, researchers found, leading them to believe that these children may have grown out of their disorders, or perhaps were compensating for their autism-related struggles, reports the BBC. 
• UPI headline: Study: "Some with autism 'outgrow it'"
• RTT News: "Some children may actually grow out of autism says a new study from researchers at the University of Connecticut."

The study does not use terms such as "outgrow" or "grow out of," which would suggest that the process is automatic, like losing baby teeth.  The New York Times quotes lead author Deborah Fein:

Dr. Fein emphasized the importance of behavioral therapy. “These people did not just grow out of their autism,” she said. “I have been treating children for 40 years and never seen improvements like this unless therapists and parents put in years of work.” [emphasis added]

Why point out misleading language in news reports?  Very few readers are going to read the NIMH release, much less  the original study.  They will see terms such as "outgrow" and may develop the false hope that the problems associated with autism will just go away by themselves.  An editorial in the same journal urges appropriate caution:

Opening the dialogue on optimal outcomes and using the word “recovery” as a possible outcome must be done responsibly. It cannot detract attention from those who do not fall in this group, those who make less progress than hoped or achieve much smaller gains. How are these children different? How do child characteristics interact with treatment characteristics to foster the best outcomes? Broderick (2009) cautions against a “binary conceptualization of hope” (p. 270) that represents outcomes other than optimal ones as tragic and hopeless. She suggests that other optimal outcomes include emergence from isolation into engagement with the world and full participation in an ordinary life, even while retaining significant symptoms. Hope for recovery as the only hope would be not only shortsighted, it would be unethical.

 

NIMH Study: Some Lose the Diagnosis

Is it possible for ASD people to "recover" or lose the diagnosis?  Some have questioned this idea, or at least suggested that recovery is very rare.  Research on outcomes is surprisingly scant, but a release from the National Institute of Mental Health discusses an important new study:

Some children who are accurately diagnosed in early childhood with autism lose the symptoms and the diagnosis as they grow older, a study supported by the National Institutes of Health has confirmed. The research team made the finding by carefully documenting a prior diagnosis of autism in a small group of school-age children and young adults with no current symptoms of the disorder.

The report is the first of a series that will probe more deeply into the nature of the change in these children’s status. Having been diagnosed at one time with an autism spectrum disorder (ASD), these young people now appear to be on par with typically developing peers. The study team is continuing to analyze data on changes in brain function in these children and whether they have subtle residual social deficits. The team is also reviewing records on the types of interventions the children received, and to what extent they may have played a role in the transition.

“Although the diagnosis of autism is not usually lost over time, the findings suggest that there is a very wide range of possible outcomes,” said NIMH Director Thomas R. Insel, M.D. “For an individual child, the outcome may be knowable only with time and after some years of intervention. Subsequent reports from this study should tell us more about the nature of autism and the role of therapy and other factors in the long term outcome for these children.”

The study, led by Deborah Fein, Ph.D., at the University of Connecticut, Storrs, recruited 34 optimal outcome children, who had received a diagnosis of autism in early life and were now reportedly functioning no differently than their mainstream peers. For comparison, the 34 children were matched by age, sex, and nonverbal IQ with 44 children with high-functioning autism, and 34 typically developing peers. Participants ranged in age from 8 to 21 years old.

Prior studies had examined the possibility of a loss of diagnosis, but questions remained regarding the accuracy of the initial diagnosis, and whether children who ultimately appeared similar to their mainstream peers initially had a relatively mild form of autism. In this study, early diagnostic reports by clinicians with expertise in autism diagnosis were reviewed by the investigators. As a second step to ensure accuracy, a diagnostic expert, without knowledge of the child’s current status, reviewed reports in which the earlier diagnosis had been deleted. The results suggested that children in the optimal outcome group had milder social deficits than the high functioning autism group in early childhood, but had other symptoms, related to communication and repetitive behavior, that were as severe as in the latter group.

The investigators evaluated the current status of the children using standard cognitive and observational tests and parent questionnaires. The optimal outcome children had to be in regular education classrooms with no special education services aimed at autism. They now showed no signs of problems with language, face recognition, communication, and social interaction.

This study cannot provide information on what percentage of children diagnosed with ASD might eventually lose the symptoms. Study investigators have collected a variety of information on the children, including structural and functional brain imaging data, psychiatric outcomes, and information on the therapies that the children received. Analysis of those data, which will be reported in subsequent papers, may shed light on questions such as whether the changes in diagnosis resulted from a normalizing of brain function, or if these children’s brains were able to compensate for autism-related difficulties. The verbal IQs of the optimal outcome children were slightly higher than those with high functioning autism. Additional study may reveal whether IQ may have been a factor in the transition they made.

“All children with ASD are capable of making progress with intensive therapy, but with our current state of knowledge most do not achieve the kind of optimal outcome that we are studying,” said Dr. Fein. “Our hope is that further research will help us better understand the mechanisms of change so that each child can have the best possible life.”

Reference

Fein D, Barton M, Eigsti IM, Kelley, E, Naigles L, Schultz RT, Stevens M, Helt M, Orinstein A, Rosenthal M, Troyb E, Tyson K. Optimal outcome in individuals with a history of autism. Journal of Child Psychology and Psychiatry DOI: 10.111/jcpp.12037. Project Number: R01 MH 076189