Autism: Taming a nightmare

In many ways, autism is a nightmare condition for health plans. Its prevalence appears to be skyrocketing, it typically requires long-term treatment, diagnoses are being made at younger and younger ages, there is a plethora of therapies (some evidence-based, some entirely fanciful), and some therapies can be very expensive.

Oh, and there’s no cure for the condition, nor even a firm evidence-based etiology for it.

Fortunately, a few things do mitigate what might otherwise seem like an overwhelming situation.

How many cases?

Most prominent is the fact that, despite some scare-mongering, autism has not actually surged out of control.

Talk of an “autism epidemic” is misleading, says Allison Singer, MBA, president of the Autism Science Foundation. “We know there’s an increase in measured prevalence, but we don’t know if there’s an increase in incidence.”

When Singer’s older brother was diagnosed with autism in 1968, she explains, autism’s prevalence was estimated at 1 in 10,000. It was a time when diagnostic criteria for autism were much more restrictive than those currently used.

Autism in the 1960s was still often diagnosed as “childhood schizophrenia.” It wasn’t until 1980, and the publication of the Diagnostic and Statistical Manual (DSM)-III that autism was distinguished as a separate condition. In 1994, the DSM-IV added Asperger syndrome, and in 2013, the DSM-V rolled all subcategories into the autism spectrum disorder (ASD) diagnosis.

So it should have been no surprise when the Center for Disease Control and Prevention’s (CDC) Morbidity and Mortality Weekly Report for March 28, 2014, estimated that (per 2010 data) one in 68 U.S. 8-year-olds has an ASD diagnosis (not classic, severe autism per se.) Even though the CDC noted that the 1-in-68 figure was not necessarily reflective of the U.S. population as a whole, the number was large enough--and enough of an increase over the 2012 estimate of one 8-year-old in 88--to fuel renewed media coverage of an “autism epidemic.”

Next: ASD cases might actually be on the rise

Singer says the term ‘autism’ is clinically meaningless because the diagnostic criteria are so much broader now. “It doesn’t describe a distinctive cluster of symptoms any more,” she notes. Its quasi-replacement, ASD,  is also fairly meaningless, she contends, because it provides no guidance for therapy. (Does the patient exhibit aggression? Self-injury? Does the patient have speech difficulties? A low IQ? A high one?)

The evolving definitions of autism in successive DSMs have wrought havoc on diagnosis and on research, Singer says.

“The definition [of autism] has changed drastically over time,” says  Susan Levy, MD, MPH, of the Children’s Hospital of Philadelphia and member of the autism subcommittee of the American Academy of Pediatrics’ Council on Children with Disabilities. She agrees with Singer that there is no autism epidemic, that is, no sudden increase in cases traceable to a risk factor, though she’s more amenable to the changed diagnostic criteria in DSM-V, saying, “It’s better science.”

Another factor in the general increase in ASD prevalence, Levy suggests, is that the average age at diagnosis was age 5 or 6 years about 10 years ago, whereas now it’s starting to dip below age 4 and has the potential to drop closer to age 2.

Finally, ASD cases might actually be on the rise. Dr. Levy believes that environmental exposures to chemicals are a likely culprit.“There has been a major shift in how we define, recognize and even conceptualize autism,” starting in the 1970s and throughout the 1990s, says Jon Baio, Ed.S. an epidemiologist with CDC’s developmental disabilities branch.

A major concern, he says, is “whether the rapid and recent increase in autism prevalence estimates is attributable to changes in etiologic risk factors or simply to our shifting classification and perception of autism, our increased focus on autism in science and the media, and our expanded funding and legislation for autism services.

“It seems unlikely that the steady rate of increase in reported ASD prevalence estimates is still closely tied to the introduction of DSM-IV 20 years ago. There are clearly other factors at play,” Baio says.

Regarding the CDC, he says, “the biggest question is not about ‘measured’ vs. ‘actual’ prevalence, it is why such wide variation exists among different U.S. communities, racial/ethnic groups, and socioeconomic classes in how individuals are getting evaluated, diagnosed, and served for their autism, and what can be done to reduce these disparities.”

Next: Costs and reforms

Costs and reforms

Behind all the questions about autism’s prevalence, of course, looms the issue of cost.

In 2005, after Judith Ursitti’s son was diagnosed with autism at age 2 as a result of routine developmental screening, Ursitti was referred to a pediatrician who prescribed several therapies including speech therapy, occupational therapy, physical therapy (for a delay in walking) and applied behavior analysis.

Ursitti’s health insurer denied coverage for all these therapies, because autism was specifically excluded under her policy. The interventions cost the family $3,000 to $5,000 per month at that time, says Ursitti, who is now director of state government affairs for Autism Speaks, an advocacy group.

As a result of that experience Ursitti, then living in Texas, became an advocate for a law in that state requiring that evidence-based ASD therapies be covered by state-regulated insurers. The law was enacted, initially covering children ages 2 through 5, though the upper age was later raised to 10, and now there is no age limit, says Ursitti.

Meanwhile, the family had moved to Massachusetts, where Ursitti again got involved in an effort to pass state-level autism insurance reform. The law was enacted in 2010, so Ursitti’s family finally has insurance coverage for autism interventions.

Ursitti says 37 states now have similar laws. And as a gauge of costs, the average premium impact on state-employee health plans has been 31 cents per month per member, she says.

Insurers’ initial fears that premiums would surge, Ursitti adds, were based on “a misunderstanding of the autism spectrum.”

Meanwhile, Medicaid has been adding its voice. On July 7, the Centers for Medicare & Medicaid Services (CMS) and Children’s Health Insurance Program Services issued an informational bulletin clarifying Medicaid coverage of services to children with autism. Though the bulletin does not create new coverage requirements, it might cause some state Medicaid programs to provide “additional or expanded ASD treatment services,” with an emphasis on the early and periodic screening, diagnosis and treatment requirement, according to the American Academy of Pediatrics (AAP). California, according to the AAP, has begun to cover applied behavior analysis services as a result of the CMS bulletin.

Next: Not insurers alone

Not insurers alone

Another mitigating factor on the ASD front is that mainstream advocacy groups aren’t pushing for reimbursement for anything beyond evidence-based therapies. “We’re not asking for dolphin therapy,” says Ursitti, nor chelation therapy, nor experimental treatments, nor supplements, nor hyperbaric oxygen, all of which are unproven.Levy adds that some non-evidence-based ASD interventions can be dangerous, including stem cell transplants, exclusion diets and chelation.

A June article in JAMA Pediatrics reported findings by a team of researchers from the United States and United Kingdom, who had estimated the lifetime direct and indirect economic costs for one individual with ASD. They found that for individuals with autism and intellectual disability, the average lifetime cost was $2.4 million in the United States and a surprisingly similar--given how different the two countries’ health systems are--$2.2 million in the United Kingdom.

For individuals without intellectual disability, the average lifetime cost was $1.4 million in either country. For children with ASDs, the largest costs were for special education and parents’ lost wages, and for adults with ASDs, they were residential care and lost wages.

Medical services costs comprised 30.9% of the average total annual costs for an adult with ASD with intellectual disability and 27% of such costs for an adult with ASD without intellectual disability. The corresponding portions for children were 16% and 13.4%, respectively.

One lesson that can be drawn from this study is that autism has costs and ramifications far beyond the healthcare realm.

For example, Ursitti notes that the federal Individuals with Disabilities Education Act (IDEA) mandates certain kinds of assistance--not healthcare, but instead aiming at educational goals--for children with disabilities, including ASD.

“In theory, that’s a great idea,” she says. “In practice, there’s no money.”  Still, Ursitti adds, “We’re not asking insurers to handle it all.” That is, part of the ongoing ASD cost burden will be borne by the educational system, local social services and potentially Medicaid, she says, adding, “it’s not appropriate to expect health insurers to cover everything.”  

Scott Baltic is a freelance writer based in Chicago.