Autism and Insurance

3 Mar

ResearchBlogging.orgOne of my favorite autism researchers is a guy named David Mandell. The reason is simple: he just asks good questions.

For example, he studied Vocational Rehabilitation and compared the results for autistics and non-autistic adults. There’s a guy looking at issues that will matter to me all too soon, and already matter to a lot of people already (people all too often forgotten even within greater the autism community: adults).

He looked at adult populations in psychiatric hospitals and found that many adults diagnosed with schizophrenia may be autistic.

Now he has addressed a big question: how much would adding insurance mandates for autism increase the premiums?

This question comes up a lot. For example, in California one of the big questions has been how can Kaiser get away without providing insurance coverage, even though California has an autism insurance mandate (AB88).

The YouTube video is about the first person to win coverage for therapies like Speech, Occupational and ABA from Kaiser:

Kaiser said this will “Significantly increase the cost” of insurance, with “their actuaries” estimating would be $5 to $7 per member per month.

As an aside, the Kaiser person in this meeting was dodging the question, and it is annoying that the interviewer let her do it. The question isn’t about how much it will cost, but the fact that California already mandates that the insurers cover the therapies and that kaiser is avoiding it’s legal responsibility.

As another aside–this is what insurance is for. We pay a little bit to share the risk. If it costs us all a little bit so that some small group gets help when they need it. We don’t question it when we are talking about, say, therapies for a stroke victim or someone in a bad accident. Why do we question it when it comes down to children with developmental delays?

Pennsylvania recently passed a mandate requiring autism insurance coverage with a $36,000 cap per year. Dr. Mandell works in Pennsylvania and used his state as an example in his paper, Quantifying the Impact of Autism Coverage
on Private Insurance Premiums.

That’s a RBQ (really big question) that comes up a lot when people are working on getting autism insurance mandates in their state: what will the cost be? Insurers, as one could imagine, claim the costs will be big. (again, avoiding the question of whether it is the right thing to do).

I won’t go into the details of the model Dr. Mandell’s team used. It was actually pretty straightforward, just as you would probably expect. Instead, let’s take a look at results. Figure 1 shows the increase in insurance premiums if autism therapies are covered.

Figure 1 from paper

Figure 1 from paper

The y-axis is the percent increase in insurance premiums, and the x-axis is the average expenditure per child with autism (in $1000’s). The model gives estimates for average expenditures from $10,000 to $36,000 (the cap in the Pennsylvania insurance mandate). Estimates are given for 3 different autism prevalences: 2.0 per 1000, 4.0 per 1000 and 6.7 per 1000. Note that 6.7 per 1000 is the same as 1 in 150.

Take a look at the highest estimate: $36,000 per child, 1 in 150 prevalence. Increase in insurance premiums? 2%. Yep, 2%.

Or, in the words of the study authors:

Even in the unlikely event that treated prevalence were to rise to the accepted community prevalence of 1 in 150 children, and per capita expenditures rose to $36,000 per year, the increase in the family contribution would reach $6.53 a month, or $78.31 per year

This may be the same amount as the Kaiser actuaries claimed. They claimed about $5-7 per “member”. Usually “member” means the primary insured (I.e. the parent whose job gives the insurance coverage). If, instead, they mean $5-7 per every insured family member, then Kaiser’s estimate is likely about 4x higher than prof. Mandell’s team’s calculation.

Prof. Mandell’s team recognizes that this $78.31 per year is likely an overestimate.

To that extent, the estimates presented here may overestimate actual increases to premiums, given that some healthcare expenditures would remain the same but now would be associated with an ASD diagnosis.

One reason they give is that many children with autism may already be receiving insurance paid medical treatment, but under a different diagnosis. I.e. doctors may be avoiding the autism label in order to get reimbursed. So, for some kids and some therapies, an “autism” mandate may just shift the costs already being paid by the insurance companies from some other diagnosis to autism.

Another reason why this estimate may be low comes to my mind, even though it isn’t discussed in the paper. I know this is anecdotal, but medical expenses are likely highest for younger kids. That’s when the OT, ST, and ABA type therapies will be most common. It doesn’t make sense that the average expenses would be the same for older kids as for younger kids. So, when they calculate based on $36,000 per child per year, they are likely overestimating the expenditures for the older kids (say ages 10-20).

Another possible cause of over-estimating the expenditures. When they estimate for the full CDC prevalence (1 in 150, or 6.7 per 1000), they are including all ASD’s. People with, say, Aspergers or PDD-NOS have significant challenges, no doubt. However, my guess is that they do not require the same level of medical expenses as someone with autistic disorder. So, by assuming the prevalence for all ASD’s, Dr. Mandell’s group may have overestimated the expenses.

Now, it is worth noting that when one includes ABA in the picture, $36,000 per year may not cover everything. A 20 hour week program could cost more than $36,000.

But, leave all of that out for now. Take the high end estimate and ask, will a 2% increase in premiums to make such a big difference? I know there is a risk of this discussion getting sidetracked into an ABA discussion. But, consider adding 2 sessions a week of speech therapy and/or 1 session a week of OT to a young autistic kid’s life. Think of the difference that could make. These are life-long benefits. Isn’t that worth something comparable to, say, the amount of insurance premiums we pay to cover heart surgeries or other very expensive medical interventions for adults?

Or to put it another way: if my insurance company said, “Sure, we can give your autistic kid these therapies now. But, if you ever have a heart attack, you are going to die on the table because we won’t pay for heart surgery.” I’d sign on the bottom line. No, I am not asking everyone to make that choice, or to sacrifice their own benefits for other people’s kids. But, isn’t improving the life of a young child worth at least as much as what we spend to improve the life of older adults?

James N. Bouder, Stuart Spielman, David S. Mandell (2009). Brief Report: Quantifying the Impact of Autism Coverage on Private Insurance Premiums Journal of Autism and Developmental Disorders DOI: 10.1007/s10803-009-0701-z

This paper has also been discussed in the Translating Autism blog, which is where I found out about it.

11 Responses to “Autism and Insurance”

  1. David Mandell March 4, 2009 at 02:40 #

    Thanks so much for the positive press. I just want to point out that this study and the formula we used is really the brain child of Jim Bouder, the first author on this paper and a great autism advocate in Pennsylvania. I just helped him format it for publication.

  2. Sullivan March 4, 2009 at 06:26 #

    Dr. Mandell,

    thank you for taking the time to respond. Just in case, I’d like to stress that when I state that the model is straightforward, I didn’t imply that it is simplistic. Just that it makes sense and that I didn’t want to discuss the details here as I wanted to concentrate on the conclusions.

  3. Do'C March 4, 2009 at 06:39 #

    Dr. Mandell (or Bouder), I’d be curious on your perspective of medical versus educational intervention. I’m all for autistic children receiving the most appropriate services. Consider speech therapy. Is it a “medical” treatment, or an educational endeavor? I don’t expect that this would provide for black and white answers, but I would be curious about your thoughts on the subject – and the implications on public educational funding.

  4. Bunny March 4, 2009 at 14:21 #

    How would insurance companies differentiate between “approved” treatments and “alternative” treatments? Will they be pressured to pay for chelation, methyl B12 shots and hyperbaric chambers? Is this dilemma partly to blame for the fact that insurance companies are balking at covering autism?

  5. Jim Bouder March 4, 2009 at 15:31 #

    I also thank you for the positive press, and David for his kind words. My only criticism of David’s words is that they are overly humble.

    Regarding the “medical vs. educational” debate, while I think there might be some gray area here and there, I think the answer is more straightforward than you might expect. First, the goals of special education and the goals of health care are not one and the same. The goal of special education is to enable a child with a disability to access the regular education curriculum to the maximum extent appropriate. The goals of health care are to relieve pain, cure disease, and improve functioning. There is no Federal mandate (or state mandate I’m aware of) that requires schools to treat the symptoms of a child’s disability – this remains the domain of health care. In the case of children with autism, educational benefit can be realized as a result of treating and ameliorating the varied symptoms of autism. In my experience, clinicians are generally better trained to address the clinical needs of children with autism than special educators. This is not to say that quality special education programs are not important or beneficial – I am merely suggesting that clinical services delivered under a medical model of care are necessary for many of our kids and, especially when delivered in cooperation with special education, will improve the child’s overall outcome.

    Second, I think that the argument that ABA services are educational in nature confuses the application of the behavioral analytic discipline with the nature of the discipline. If you watch “Criminal Minds” (a crime drama about the FBI’s Behavior Analysis Unit), you will witness the application of behavioral principals toward the end of solving crimes, but I doubt anyone would argue that ABA is a law enforcement discipline. Similarly, Organizational Behavior Management principals are widely used by management and human resources professionals to improve employee performance, but this application of behavioral principals does not render it a human resources discipline. Applied Behavior Analysis is a specialized psychological discipline that can be used to enhance many other disciplines concerned with human behavior. Regarding ABA and autism (and other developmental disabilities), if you review the literature, you will see that the vast majority of ABA research is conducted by persons with backgrounds in behavioral psychology.

    To sum up, clinical applications of ABA are aimed at improving the functioning of people with autism, whereas special education aims at enabling the child to make meaningful educational progress. I believe these are distinct goals, but the first certainly effects the second. Also, ABA is a psychological discipline with a broad array of applications, including (not surprisingly) clinical applications.

    I hope this helps clarify the issue for you. Thanks again for your kinds words regarding the paper.


  6. isles March 4, 2009 at 17:14 #

    Bunny, that’s a really good question, I think, and if it is the case that insurers are holding back from covering autism therapies generally because it would mean they’d have to pay for quackery, that’s just another example of how the acts of the mercury militia hurt all autistics.

  7. Another Voice March 5, 2009 at 17:07 #

    Probably the first reason that insurance companies don’t want to pay for autism treatment is because they have not charged any premiums to cover those costs. Once the premiums have been increased to cover the projected outlay arguments against payment should subside.

    I see the problem of defining what services will be covered to be more lingering and difficult. Many people do not want chelation to be covered, but here we have a medical procedure that will not be covered by medical insurance. State regulators do not take a stand on doctors performing the procedure. But would we expect health insurance companies to take a stand and say this is not recommended medical practice; therefore we will not pay?

  8. alyric March 5, 2009 at 17:26 #

    @ Jim

    “Regarding ABA and autism (and other developmental disabilities), if you review the literature, you will see that the vast majority of ABA research is conducted by persons with backgrounds in behavioral psychology.”

    Since the basic principles are poorly known and not really accepted outside ABA land, that’s hardly surprising is it? It’s long been known that behaviour analysis was quite confined to ABA style research methods into developmental disabilities – to the great detriment of any understanding of autism or autistic people I might add.

    Be interested to know what ‘clinical’ applications ABA has? For me, clinical is ‘medical’ so what are they?

  9. Jim Bouder March 5, 2009 at 17:54 #

    In a nutshell, I’d say (1) the clinical applications include addressing the varied symptoms of developmental disabilities, mental health, and traumatic brain injury, (2) the educational applications include addressing learning disabilities, and designing and evaluating instructional technology, and (3) OBM focuses on business and industry issues.

    As a dad, my experience with ABA is much different. I found no better way to learn about my son (who has autism) than to equip him with the tools to tell me what he’s thinking.

    Bringing this back around to insurance, I think the larger issue is whether the research satisfies the evidence-based coverage criteria used by insurers to approve other treatments. In the case of ABA, I believe it does (as does the Independent Panel assembled to review Pennsylvania’s legislation). I believe the paper demonstrates that cost is not an overwhelming issue – this opens the field of debate to whether insurers are consistently applying their standards for approving new treatment technologies and, if not, whether it is appropriate for legislatures or regulators to intervene.


  10. alyric March 6, 2009 at 00:28 #

    “In a nutshell, I’d say (1) the clinical applications include addressing the varied symptoms of developmental disabilities, mental health, and traumatic brain injury,”

    Could you be a little more specific? What symptoms and how does a psychologist of minimal to zero biological background address such?

  11. Jim Bouder March 6, 2009 at 04:52 #


    Out of respect to Sullivan, I’d prefer not to get into an ABA debate in this thread. You may feel free to email me at if you’d like to continue this discussion. If I can’t answer your questions, I’m sure I can direct you to someone who can.

    I’d be happy to answer any questions you have regarding the paper.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: