Applied Behavioral Analysis Treatment for Autism Spectrum Disorder - The Fight for Coverage under ERISA Plans

Bloomberg Law®, an integrated legal research and business intelligence solution, combines trusted news and analysis with cutting-edge technology to provide legal professionals tools to be...

Tracey S. Bellapianta | Bloomberg Law As a result of the large number of Autism Spectrum Disorder (ASD) diagnoses, and no doubt the related cost, there have been a multitude of claims to health care plans governed by the Employee Retirement Income Security Act (ERISA)1 for Applied Behavioral Analysis (ABA) treatment. Over the years, the trend has been for insurance companies to deny ABA treatment on grounds that it is experimental or investigative and thus excluded from the plans. As a result, plaintiffs have filed class actions against the insurance companies arguing that the treatment is not experimental or investigative. This article will explore the claims for class certification brought by three similar classes of plan participants seeking such benefits, and the various district courts' treatment of the same. In Churchill v. Cigna Corp.2 and Potter v. Blue Cross Blue Shield of Michigan,3 the District Courts of Pennsylvania and Michigan, respectively, granted class certification. In Graddy v. Blue Cross Blue Shield of Tennessee, the Eastern District of Tennessee declined to do so.4 In each case, plaintiffs asserted claims for benefits due under their respective plans pursuant to ERISA § 502(a)(1)(B).5 These decisions will be discussed in detail.

Autism Spectrum Disorder and ABA Treatment

Autism is a neurobiological disorder that affects a child's development by severely limiting his or her ability to interact with others.6 Federal regulations define autism as a "developmental disability significantly affecting verbal and nonverbal communication and social interaction, generally evident before age three, that adversely affects a child's educational performance."7 Autism is part of a larger class of Autism Spectrum Disorder (ASD) which refers to a continuum of related cognitive and neurobehavioral disorders characterized by severe and pervasive impairment in several areas of development: reciprocal social interaction skills; communication skills; or, the presence of stereotyped behavior, interests and activities.8 Symptoms of ASD include lack of interest in establishing relationships, obliviousness to others or their needs, lag in development of spoken language or language comprehension, and stereotyped body movements like clapping, finger flicking, rocking or swaying, or walking on tiptoes.9 As ASD has no cure, many treatments focus on addressing the development impairments caused by the disorder.10 Applied Behavioral Analysis (ABA) is used in many cases to treat ASD. ABA is described as a "systemized process of collecting data on a child's behaviors and using a variety of behavioral conditioning techniques to teach and reinforce desired behaviors while extinguishing harmful or undesired behaviors[.]"11 The treatment focuses on the use of rewards or reinforcement to encourage desired behaviors and the elimination or reduction of unwanted behaviors by removing their positive consequences. As new skills are acquired, they are generalized into other settings with the intent that the child learns to employ that skill in new situation without being prompted.12 Studies have suggested that ABA therapy be initiated at an early age, for a minimum of 20 to 40 hours a week, and continue for two to four years.13 According to some studies, ABA therapy performed in a clinical setting can cost as high as $50,000 per year.14 The Centers for Disease Control and Prevention (CDC) estimates that between about 1 in 80 and 1 in 240, with an average of 1 in 110, children in the United States have an ASD.15

Proposed Class Definitions

The class definitions proposed by the class plaintiffs were comprised of two distinct components -- those participants who were enrolled in plans administered by the defendant insurance company who made claims for ABA in connection with ASD, and were denied on grounds that such treatment was deemed to be investigative or experimental (Claims Made Class), and those participants enrolled in plans administer by the defendant insurance company who did not make claims for ABA in connection with ASD in light of defendants' policy that such treatment was deemed to be investigative or experimental (No Claims Made Class). The plaintiffs in both Churchill and Potter proposed both the Claims Made Class and the No Claims Made Class, whereas the plaintiffs in Graddy only included the Claims Made Class.16

Class Claims

A party seeking to certify a class action must first meet the four requirements of Fed. R. Civ. P. 23(a): (1) the class must be so numerous that joinder of all members is impracticable; (2) there must be questions of law or fact common to the class; (3) the claims or defenses of the representative party must be typical of those of the class; and (4) the representative party must fairly and adequately protect the interests of the class. These requirements are generally referred to as numerosity, commonality, typicality, and adequacy. A party who satisfies the Rule 23(a) prerequisites must then meet the requirements of one of the subsections of Rule 23(b). In these three cases, certification was sought under: (1) Rule 23(b)(2), which allows certification when the opposing party “has acted or refused to act on grounds that apply generally to the class,” and/or (2) Rule 23(b)(3), which, allows certification when the court finds that the questions of law or fact common to the members of the class predominate over any questions affecting only individual members, and that a class action is superior to other available methods for the fair and efficient adjudication of the controversy.17

Claims Made Class


In Potter, the Eastern District of Michigan held that the numerosity requirement was satisfied, noting that identifying the exact number of members was not required, so long as “general knowledge and common sense indicate[d] that the class [was] large.”18 The court looked to a Michigan Public Schools Autism Prevalence Report which stated that nearly 14,000 children in Michigan (1 in 150) were diagnosed with ASD in 2008 and then assessed defendant Blue Cross Blue Shield’s (BCBS) market presence: BCBS is the largest non-profit health insurer in Michigan, and covers nine times as many Michigan residents as its next largest competitor, and more than 60% of Michigan’s commercially insured population. More than ninety patients have been treated with ABA therapy through Beaumont’s CARE program alone in the last six years. Assuming sixty percent are insured by BCBS, the number from the CARE program alone exceeds fifty. And many, if not most, of these patients are undoubtedly insured through an employer-sponsored health plan covered by ERISA. The class is not limited to those that received treatment from the CARE program, but includes those that received treatment at the Children’s Hospital of Michigan Autism Center, the Judson Center, and the University of Michigan Autism & Communication Disorders Center, among other programs nationwide.19 The court thus rejected BCBS’s contention that the class was too small to warrant class treatment. In Churchill, the Eastern District of Pennsylvania overlooked the fact that plaintiff did not provide specific information about the size of the class and similalry assessed defendant Cigna’s customer base and the incidence of ASD in the general population to conclude that the numerosity requirement was met: Cigna is the fourth largest health insurer in the United States, with more than 11 million participants, and the Center for Disease Control and Prevention (CDC) has estimated that 1 in 110 children in the United States have autism. . . . Assuming the incidence of autism in Cigna Plan participants is similar to the incidence of autism among in the general population and assuming each Cigna member’s health care plan is covered by ERISA, the proposed class could consist of as many as 100,000 individuals. If even a small fraction of these Cigna members filed claims for ABA, Subclass A would be far too large and geographically diverse to render joiner practicable.20 In Graddy, the Eastern District of Tennessee did not consider the numerosity requirement, as it ultimately held that plaintiffs could not meet the commonality or typicality requirements.

Commonality and Typicality

The district courts reached different conclusions when determining whether the classes met the commonality and typicality requirements: while the Potter and Churchill courts determined that commonality and typicality was present, the Graddy court held otherwise. — Commonality In Potter, the court held that the claims of the proposed class depended on a common, central issue – the determination that there was “no reasonable basis for disagreement that ABA treatment is mainstream medical treatment and not experimental.”21 The court held that this, along with the contention that BCBS uniformly failed to provide claimants with evidence and an explanation for its determination that ABA treatment was experimental, would form the entire basis of the lawsuit and generate answers to resolve the class members’ claims. The court found that the only dissimilarity among the class members was the amount in benefits to which each may be entitled, but that this fact lacked the potential to “impede the generation of common answers” that would resolve the litigation.22 Thus, the Potter court found that commonality was satisfied. In Churchill, the court similarly held that the central question to consider was “whether Cigna’s denial of medical coverage for ABA as treatment for ASD on the basis that such treatment is investigative or experimental was proper, and the answer to this question will resolve each class member’s individual claim.”23 The court noted that, while class members might: (1) be entitled to different benefits, (2) be participants in different Cigna-administered plans, or (3) have children who would benefit from ABA in varying degrees, these “differences [were] not significant enough in this litigation to defeat commonality.”24 Thus, since the entire class was allegedly harmed by Cigna’s uniform policy, and because resolution of the class members’ claim hung on whether such a policy was proper, commonality existed.25 In contrast, the Graddy court held that plaintiffs' motion to certify the class should be denied because the common issue in all the cases – “whether ABA treatment is experimental per se” – did not predominate over the other factual and legal issues that likely would arise in the cases of other plaintiffs.26 The court held that, “Other questions, including whether ABA treatment was experimental as to a specific plaintiff's condition and whether that plaintiff would, in fact, have benefitted from ABA treatment are not common and would require a highly individualized determination because individuals suffering from ASD and autism ‘may exhibit the characteristic traits of autism and ASD in any combination, and in different degrees of severity.’”27 — Typicality In Graddy, the court held that plaintiffs could not meet the typicality requirement set forth in Rule 23(a) because, even though all of the plaintiffs may have been subjected to some or all of defendant's alleged wrongful practices, this fact did not eliminate the need for an individualized assessment as to the ultimate propriety of the benefits decisions affecting each and every class member.28 The court continued, noting that in light of the required individualized assessment of each class member’s ERISA claim, it did not follow that plaintiff's success in establishing defendants liability would also apply to the class.29 In contrast, the Potter court held that plaintiff’s claims were typical of those of the class because, if plaintiff was able to prove that he was improperly denied benefits on the ground that ABA treatment was experimental, he would be proving that the claims of the class members, who were denied benefits on that same ground, were been improper as well. 30 In so ruling, the court again rejected defendants’ argument that plaintiff’s claims were not typical because he failed to exhaust his administrative remedies by not filing a second-level review. The Potter court addressed the Graddy court's determination that commonality and typicality were not met, noting that “[t]o the extent that Graddy is indistinguishable from this case, the Court respectfully disagrees with its analysis.”31 It then went on to reiterate its finding that individualized questions did not exist because BCBS had made an across-the-board determination that the ABA treatment was experimental, and therefore not a covered benefit. The Potter court went on to explain that the insurance company was not making its benefits determination after considering the individualized medical needs of the members, but based solely on its policy determination that ABA was experimental in all of its applications. If BCBS’s sole basis for denying a claim for ABA treatment is that it is, as a matter of company policy, experimental, and BCBS makes the decisions without consideration of each claimant’s individual medical needs, and the Court concludes that this ground of decision has no basis in fact, then the claimants will necessarily be entitled to benefits. The class consists only of those persons whose claims were denied because of BCBS’s determination that ABA was experimental. Whatever the nature of the individualized determinations the district court in Graddy believed destroyed the typicality and commonality requirements, such determinations need not be made here.32 In Churchill, the court also held that the claims met the typicality requirement because the same conduct – Cigna’s routine denial of reimbursement for ABA – affected all class members, including Churchill. However, since Churchill was not a member of the plan at the time of the lawsuit, the court held that he was not a typical representative of those potential class members who were current participants in a Cigna-administered plan.33 The court noted that the distinction was important since ASD is an ongoing condition, and that current plan participants have an interest in seeking prospective injunctive relief, while Churchill would not have this interest and thus would not benefit from having Cigna change its policy going forward. Since his interest would only be for economic damages in the past, the court narrowed the proposed class to include only former Cigna members, whom Churchill was qualified to represent.34

Adequate Representative

In Potter, the court rejected defendants' argument that Potter would be an inadequate representative due to his failure to exhaust his administrative remedies because he did not seek a second-level review.35 Again, the court held that the exhaustion issue did not pose a bar to judicial review of Potter's claims. The Churchill court reiterated that Churchill would be an inadequate representative for current Cigna customers since he would not be incentivized to request injunctive relief from which he could not benefit.36 Accordingly, the court found Churchill to be an adequate representative of the modified class, which only included former Cigna plan members.

Rule 23(b)(2)

After finding that plaintiffs could not be certified under Rule 23(a), the Graddy court further held that plaintiff could not maintain a class action under Rule 23(b)(2) because the proposed class did not meet the homogeneity requirement inherent in such a class. The court held that “the varied behavioral disorders exhibited by patients with ASD, and the question of whether such behavior disorders may not be treated by ABA, fail to meet the requisite homogeneity requirement needed to protect the interests of unnamed class members because such an assessment would require an individualized review of the insured’s claim. “37 In Churchill, the court also held that certification under Rule 23(b)(2) was inappropriate, based on Churchill's status as a non-plan participant. The court reasoned that, since the only relief available under that section was primarily injunctive or declaratory, this relief would be inappropriate for Churchill (he would only be interested in monetary relief) and therefore certification under this section was not granted.38 In Potter, the court held that certification under Rule 23(b)(2) was appropriate because BCBS’s actions affected the class as a whole. Thus, injunctive relief in the form of requiring BCBS to pay benefits, or to provide evidence as an explanation for why BCBS considers ABA treatment to be experimental, would be appropriate if entitlement to the relief was established.39

Rule 23(b)(3)

In Potter, the court held that certification under Rule 23(b)(3) was appropriate because the common issues of whether ABA was experimental and whether BCBS provided the class members with the opportunity for full and fair review predominated over any individualized issues. The court was further influenced by these additional factors: (1) BCBS acted in the same manner with respect to each member regardless of the individualized situations; (2) the class was sufficiently limited to include only those whose claims were denied on the grounds that the treatment was experimental, and not for any other independent or adequate reasons; and (3) to the extent that there were differences in the amount of benefits class members were entitled, those differences did not establish that individual questions predominate over common questions.40 In Churchill, the court rejected Cigna’s argument that individual issues would predominate over common class issues since each proposed class member’s claim depended on the terms of his or her plan, the type of ASD for which an ABA treatment claim was being made, and the nature of the ABA treatment which was received. Distinctions in class members’ Cigna plan terms do not preclude predominance. While each member’s plan may have different exclusions, it appears the “investigative or experimental” exclusion was the basis for Cigna’s denial of ABA coverage. The Cigna Medical Coverage Policy excludes coverage for ABA on this basis, and Cigna has not provided evidence to show other ERISA-covered plans altered this policy to either allow coverage for ABA or to deny coverage on a different basis. Similarly, while there may be differences in the types of ASD that affected each class member’s child and the types of ABA treatment that child received, Cigna made a class-wide determination that ABA was experimental in all cases. The propriety of this determination-specifically, whether it violates ERISA-can easily be litigated in a single forum.41 Thus the Churchill court also held that class certification under FRCP 23(b)(3) was appropriate. 42

No Claims Made Class

In Potter, the court certified the subclass of beneficiaries who did not make a claim for ABA therapy for ASD in light of defendant's policy that such treatment was deemed to be investigative or experimental.43 The Potter court held that the “theoretical difficulty” of determining whether a person was a member of that particular subclass was overcome by the presumption that any insured who failed to submit a claim did so based on the insurance carrier’s policy to deny ABA coverage, because “[t]he Court can fathom no other reason" to explain the failure to submit such a claim.44 The Churchill court did not agree with the reasoning in Potter and declined to certify the proposed class of individuals who failed to submit a request for benefits. The Churchill court held that the proposed class of individuals did not exhaust his or her administrative remedies, and also failed to prove that such action would be futile, noting that the Third Circuit has denied use of the futility exception when an ERISA plaintiff did not request the contested benefit, even when the plan has a blanket policy of denying all such requests.45 The court noted that “[c]ourts are reluctant to excuse exhaustion when a plan member fails to make a claim because the futility exception requires an individualized analysis of each non-claimant’s behavior before filing a lawsuit.”46 Thus, the court concluded that, in the class context, it would not be able to find clear evidence of futility since a class action, by its very nature, precludes consideration of individual circumstances. Speaking to the Potter decision, the Churchill court noted that the Eastern District of Michigan's reasoning was unpersuasive, citing the “multitude of reasons” why a beneficiary might fail to file a claim, including a beneficiary’s belief that a different provision of the policy might render reimbursement unavailable, such as the education exclusion.47 The Churchill court noted that, in this circumstance, Cigna’s policy of deeming ABA “investigative and experimental” would therefore not be the actual cause of harm to the claimant.


Of the three class actions brought by ERISA plan participants for ABA in connections with ASD, two district courts granted certification, and one denied class certification on grounds that a review of the claims would require too individualized of a review and thus could not meet the commonality or typicality requirements of Rule 23(a). Of the two cases where class certification was granted, only one of the decisions allowed claims to go forward for plan participants who had not previously brought claims for benefits under their benefits plans. With such different results from these three district court opinions, it will be interesting to see how other courts handle similar cases, which, given the prevalence of ASD, likely will be filed. DisclaimerThis document and any discussions set forth herein are for informational purposes only, and should not be construed as legal advice, which has to be addressed to particular facts and circumstances involved in any given situation. Review or use of the document and any discussions does not create an attorney-client relationship with the author or publisher. To the extent that this document may contain suggested provisions, they will require modification to suit a particular transaction, jurisdiction or situation. Please consult with an attorney with the appropriate level of experience if you have any questions. Any tax information contained in the document or discussions is not intended to be used, and cannot be used, for purposes of avoiding penalties imposed under the United States Internal Revenue Code. Any opinions expressed are those of the author. The Bureau of National Affairs, Inc. and its affiliated entities do not take responsibility for the content in this document or discussions and do not make any representation or warranty as to their completeness or accuracy.©2014 The Bureau of National Affairs, Inc. All rights reserved. Bloomberg Law Reports ® is a registered trademark and service mark of The Bureau of National Affairs, Inc.

Request Bloomberg Law®