McHenry v. PACIFICSOURCE HEALTH PLANS

679 F. Supp. 2d 1226, 48 Employee Benefits Cas. (BNA) 2512, 2010 U.S. Dist. LEXIS 321, 2010 WL 56106

• Court: District Court, D. Oregon
• Decided: January 5, 2010
• Docket: CV-08-562-ST
• Status: Published

Opinion

OPINION AND ORDER

STEWART, United States Magistrate Judge:

INTRODUCTION

Plaintiff, Lisa A. McHenry (“McHenry”), is a participant in the Metro Area Collection Service, Inc. Group Health/Dental Plan, which is insured by defendant, PacificSource Health Plans (“Pacific-Source”). McHenry’s minor son, J.M., suffers from autism and receives Applied Behavioral Analysis (“ABA”) therapy. This therapy has been effective in treating J.M.’s autism but at a substantial cost. PacificSource is the claims administrator and has denied coverage for J.M.’s ABA therapy. McHenry brings this action under the Employee Retirement Income Security Act of 1974 (“ERISA”), 29 USC §§ 1001-1461, to compel coverage.

On May 5, 2009, 643 F.Supp.2d 1236, this court ruled that because the Plan did not unambiguously grant PacificSource the power to determine eligibility, interpret Plan language, or making binding benefits determinations, the de novo standard of review applies to PacificSource’s denial of benefits (docket # 27).

The parties have filed cross motions for summary judgment (dockets # 41 & # 47). All parties have consented to allow a Magistrate Judge to enter final orders and judgment in this case in accordance with FRCP 73 and 28 USC § 636(c). For the reasons set fourth below, McHenry’s motion is denied and defendants’ motion is granted.

UNDISPUTED FACTS

J.M. was diagnosed with autism in May 2006, at the age of one year and nine months. On or about November 20, 2006, J.M.’s pediatrician, Rupa K. Shah, M.D., submitted to PacificSource a request for coverage for ABA therapy. J.M. began receiving ABA therapy from Emily Hoyt, a Board Certified Behavior Analyst (“BCBA”), in January 2007. Hoyt submitted invoices to PacificSource for payment of services provided to J.M. from January through April 2007. SR 16-18. 1

In June 2007, PacificSource denied payment of these billings, explaining that the “[pjrovider is not eligible on this plan.” SR 16. Later that same month, McHenry submitted to PacificSource an Initial Grievance of the denial. SR 20. In her grievance she inquired “what would make a therapist eligible to provide [ABA therapy] on our plan[?]” and whether Pacific-Source “offerfed] a plan that inelude[d] ABA therapy?” Id. She requested that her claim receive a medical, not administrative, review.

PacificSource submitted McHenry’s grievance to its Medical Grievance Review Committee (“Grievance Committee”). SR 50-53. On August 2, 2007, the Grievance Committee notified McHenry that it had upheld PacificSource’s denial of her claim on three bases: (1) the Plan “specifically exclude[d] coverage for experimental or investigational procedures, services and treatments;” (2) “the plan exclude[d] academic or social skills training;” and (3) BCBAs, “while professionally educated, are not medically trained clinicians and are not eligible providers for PacificSource.” SR 54. It then explained:

This determination is based on the above exclusions and a lack of sufficient evidence-based peer-reviewed literature and other supporting data to establish this as a standard of care of coverage. The committee determined that Applied Behavior Analysis meets the plan definition of an experimental or investigational procedure.

Id.

McHenry appealed this decision on August 6, 2007. SR 70-71. She disagreed with the conclusion that ABA therapy was experimental or investigational in nature and cited to an article listing the many medical professionals, medical organizations, and government agencies that had accepted it as a scientifically based treatment for children with autism. SR 70, 72-77 (Erick V. Larsson, Ph.D., Intensive Early Intervention using Behavior Therapy is No Longer Experimental, available at http://rsaffran.tripod.com/ieibt.html) (last accessed Jan. 5, 2010). 2

PacificSource submitted her appeal to its Policy and Procedures Review Committee (“Policy Committee”). SR 93. By letter dated August 28, 2007, the Policy Committee informed McHenry that it had upheld the denial, explaining that “[a]fter reviewing all of the available information in this case, the committee concluded that the services provided by ABA therapy are educationally based social/interactive skill training services” which were “specifically exclude[d]” by the Plan. Id. If McHenry believed any covered services were being provided “in adjunct to ABA therapy,” she would need to submit those services for a payment decision, but to be covered, “eligible services would need to be provided by an eligible medical or mental health provider....” Id.

On September 24, 2007, McHenry submitted her written appeal of the Policy Committee’s decision, disputing the conclusion that ABA therapy was primarily educational or social skills training. SR 108. She noted that while some of the results of the therapy included improvement in educational and social skills, “ABA therapy programs include speech and several hundreds of other therapeutic goals that are essential activities of everyday life.” Id (emphasis in original). She compared the focus and improvement of everyday activities provided by ABA therapy to that provided by therapy for an orthopedic disability. Id. Additionally, she submitted letters in support of her claim from Dr. Shah and from Karen Grant, Psy.D., a psychologist with the Oregon Health Sciences University, Child Development and Rehabilitation Center Autism Clinic. SR 109-12.

PacificSource acknowledged McHenry’s appeal by letter October 1, 2007, and informed her that the next and final level of PacificSource’s internal review process was a hearing before the Membership Rights Panel (“MRP”). SR 196. McHenry appeared before the MRP on November 7, 2007. SR 219, 350. She presented testimony and documents which she believed refuted each of the three bases that had been cited for denying her claim at the three previous levels of review. SR 224-347.

On November 21, 2007, PacificSource notified McHenry of the MRP’s conclusion that ABA therapy was “behavioral-educational social skill training” specifically excluded by the Plan. SR 351. It also informed her that she could request an independent external review. Id.

McHenry requested that review, and PacificSouree randomly selected Independent Medical Expert Consulting Services, Inc. (“IMEDICS”) to conduct it. SR 368. On December 12, 2007, IMEDECS notified McHenry that because her dispute did not involve an adverse determination based on medical necessity, experimental or investigational treatment, or continuity of care, Oregon external review law did not apply, and it would conduct no review. SR 381.

Having exhausted her remedies with PacificSource, McHenry filed this lawsuit on May 5, 2009.

STANDARDS

The parties have filed motions for summary judgment pursuant to FRCP 56. However, it is clear from the parties’ briefing that they desire the court to issue final judgment based upon the stipulated record and the additional evidence submitted with their supporting memoranda. In an ERISA case, under the de novo standard of review, “[t]he court simply proceeds to evaluate whether the plan administrator correctly or incorrectly denied benefits.” Abatie v. Alta Health & Life Ins. Co., 458 F.3d 955, 963 (9th Cir.2006) (en banc.) In conducting this review the court “can evaluate the persuasiveness of conflicting testimony and decide which is more likely true.” Kearney v. Standard Ins. Co., 175 F.3d 1084, 1095 (9th Cir.1999) (en banc), cert. denied, 528 U.S. 964, 120 S.Ct. 398, 145 L.Ed.2d 310 (1999). Moreover, given the nature of the issues in this case, to rule in favor of either party, this court must make factual findings by weighing the evidence in the record. Accordingly, FRCP 56, with its “genuine issue of material fact” standard, is inappropriate. See id. Instead, the proper procedural mechanism is a motion for judgment on the record pursuant to FRCP 52. See Thompson v. Ins. and Benefits Trust, 670 F.Supp.2d 1052, 1054-56 (E.D.Cal.2009); Rodgers v. Metro. Life Ins. Co., 655 F.Supp.2d 1081, 1085-86 (N.D.Cal.2009). The court construes the parties’ motions as being brought pursuant to FRCP 52 and will decide this matter based upon the evidence contained in the stipulated record and such other evidence it finds is clearly “necessary to conduct an adequate de novo review.” Mongeluzo v. Baxter Travenol Long Term Disability Benefit Plan, 46 F.3d 938, 944 (9th Cir.1995) (citation omitted).

DISCUSSION

I. Background

A. Nature of Autism

Autism is a neurobiologieal disorder that affects a child’s development by severely limiting his or her ability to interact with others. See SR 267-68 (Dep’t of Defense, Report and Plan on Services to Military Dependent Children with Autism 5 (July 2007) (“DOD Report”).) 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.” 34 CFR § 300.8(c)(1)®.

Autism is part of the larger class of Pervasive Developmental Disorders (“PDD”) or Autistic Spectrum Disorders (“ASD”), synonymous terms which refer 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.” Diagnostic and Statistical Manual of Mental Disorders 69 (4th ed. text revision 2000) (“DSM-IV-TR”); SR 931 (Pauline A. Filipek, et al., Intervention for Autistic Spectrum Disor ders, 3 NeuroRX 207, 207 (April 2006)). These conditions are present from birth or early in development and are typically diagnosed in early childhood. The cause of autism is unknown and may have “multiple etiologies that are currently grouped together under this diagnostic umbrella because of the similar core behavioral symptomatology.” SR 931-32 (Filipek, supra, at 207-08).

As its physiological etiology is unknown, autism is diagnosed by the behavioral symptoms it causes. Specifically, diagnostic criteria for autism require the presence of six symptoms from three categories of behavior: impaired reciprocal social interaction, impaired communication, and restricted, repetitive, or stereotyped behaviors. DSM-IV-TR at 75. Examples of these symptoms can include a lack 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. Id. at 70-71.

B. ABA Therapy

Autism has no known cure. Because its etiology is not fully understood, it is not surprising to find no etiology-based treatment methods. See SR 892 (Kostas Francis, Autism Interventions: A Critical Update, 4ÍI Developmental Med. & Child Neurology 493 (2005)). Thus, many treatments focus primarily on addressing the developmental impairments caused by the disorder. See SR 903 (Patricia Howlin, The Effectiveness of Interventions for Children with Autism, J. Neural Transmission, Supplement 69, at 101 (2005)). ABA therapy is one such treatment.

“ABA describes a systematized 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.... Practically speaking, it is the application of behavioral principles to shape behaviors and teach new skills in an individual.” SR 270 (DOD Report at 8). ABA is not unique to autism; its methods are derived from Skinnerian behavioral psychology and have been applied to community development, social work, nursing, industry, education, and medicine. See SR 1322 (Karola Dillenburger, Parent Education and Home-Based Behavior Analytic Intervention: An Examination of Parents’ Perceptions of Outcome, 29 J. Intellectual & Developmental Disability (2004)). It was first studied and applied as a potential treatment methodology for autistic children by O. Ivar Lovaas at UCLA. See SR 991 (O. Ivar Lovaas, Behavioral Treatment and Normal Educational and Intellectual Functioning in Young Autistic Children, 55 J. Consulting and Clinical Psychology 3 (1987)).

ABA employs “operant conditioning” and “discrete trial training” among other behavioral psychology techniques to teach basic life skills one small step at a time. Throughout the treatment, “the focus is on the use of rewards or reenforcement to encourage desired behaviours and the elimination or reduction of unwanted behaviours by removing their positive consequences by means of ‘time out,’ ‘extinction,’ or punishment.” SR 894 (Francis, supra, at 495). As new skills are acquired, they are “generalized” into other settings with the intent that the child learns to employ that skill in a new situation and without the encouragements or “prompts” initially relied upon. Following these methods over a period of several years, Lovaas’s study found that it was possible for some autistic children to acquire the skills needed to enter into and successfully complete first grade in an “ordinary” classroom unassisted. Over 40% of the participants in his experimental group were reportedly indistinguishable from non-autistic children.

Although Lovaas’s methods and results are not without their critics, multiple studies over the past two decades have confirmed his findings that ABA is generally beneficial to children diagnosed with PDDs. See, e.g., SR 979 (Glen O. Sallows & Tamlynn D. Graupner, Intensive Behavioral Treatment for Children with Autism: Four-Year Outcome and Predictors, 110 Am. J. on Mental Retardation 417 (2005)); SR 1209 (Howard Cohen, et al., Early Intensive Behavioral Treatment: Replication of the UCLA Model In A Community Setting, 27 J. Developmental and Behavioral Pediatrics 145 (2006)); SR 1335 (Tristram Smith, et al., Intensive Behavioral Treatment for Preschoolers with Severe Mental Retardation and Pervasive Developmental Disorder, 102 Am. J. on Mental Retardation 238 (1997)); SR 1471 (Bob Remington, et al, Early Intensive Behavioral Intervention: Outcome for Children With Autism and Their Parents After Two Years, 112 Am. J. on Mental Retardation 418 (2007)). Since Lovaas’s study, ABA has expanded and grown as research has continued to test its efficacy in different populations and in clinical or non-clinical settings and practitioners have attempted to standardize best practices. See SR 1228-30 (Robert Horner, et al., Problem Behavior Interventions for Young Children with Autism: A Research Synthesis, 32 J. Autism & Developmental Disorders 423, 424-26 (2002)).

While the degree of ABA’s efficacy is the subject of current research and debate, “[djecades worth of scientific research provide clear and convincing support” for its use as an “effective intervention.” SR 926 (William J. Barbaresi, et al., Autism: A Review of the State of the Science for Pediatric Primary Health Care Clinicians, 160 Archives of Pediatrics & Adolescent Medicine 1167, 1171 (AMA 2006)). These studies indicate that ABA should be initiated at an early age, for a minimum of 20 to 40 hours a week, and for two to four years. Id.; SR 996 (Lovass, supra); SR 1210 (Cohen, supra); SR 1252 (Svein Eikeseth, et al., Outcome for Children With Autism Who Began Intensive Behavioral Treatment Between Ages 4 and 7, 31 Behavior Modification 264 (2007))

ABA therapy is costly and demands a substantial investment of a family’s time and money. Family involvement is a critical component, and it is common for parents to be trained in its methods to continue its application at home. See SR 1252 (Eikeseth, supra.) The financial cost of ABA therapy services in a clinical setting can easily reach as high as $50,000 per year. SR 979 (Sallows, supra, at 418).

A defining feature of ABA intervention is treatment directed by a professional with advanced formal training in behavioral analysis. Oregon has no certification procedure for these professionals. Shaw Deck, ¶ 2 & Ex. A. The nationally accredited certification agency, the Behavior Analyst Certification Board (“BACB”), provides a standardized certification as a BCBA. See SR 1061-64 (BACB, Standards for Board Certified Behavior Analyst® (BCBA®), available at http://www.bacb. conVbecom_frame.html) (last accessed Jan. 5, 2010). A BACB certification as a BCBA requires, at a minimum, a masters degree and several hundred hours of graduate level instruction or mentored or supervised experience with another BCBA. Additionally, multiple universities throughout the United States provide advanced degree programs in ABA therapy which involve a combination of course work and practical experience.

C. ABA Therapy Provided to J.M.

J.M. began receiving ABA therapy from Hoyt in January 2007. Hoyt received her Masters Degree in Behavior Disorders/ABA from Columbia University in New York. SR 187. She is a certified BCBA and has worked with autistic children since 1998. Id.; SR 1188.

Hoyt provides ABA therapy through Building Bridges, a clinic in southeast Portland. SR 187. Its services include “comprehensive home programs for young children on the autism spectrum.” SR 188. Each child is given an individual assessment and a plan specifically tailored to his or her needs. The ABA therapy is targeted at the child’s communication, cognitive skills, academics, social skills, lay skills, and self-help and fine motor skills. Treatment is provided in home through two-hour, one-on-one sessions with a therapist, and multiple sessions a day are recommended. Parents are trained in the techniques used by the therapist in order to apply the elements of the treatment to daily interactions with their child. Id.

According to McHenry, J.M. has benefited greatly from the ABA therapy provided by Hoyt. SR 1159.

D. Coverage for Autism Under the Plan

At the time McHenry began seeking services for J.M., the Plan dated November 1, 2006 (“2006 Plan”), specifically excluded benefits for PDDs. SR 1732. It did, however, provide benefits for services related to conditions which may be symptoms of autism, such as speech, physical, and occupational therapy. SR 1717-18. PacificSource paid benefits for treatment J.M. received along these lines in early 2007, prior to the Plan’s annual renewal date of November 1, 2007. SR 14-15. PacificSource never cited the pervasive developmental disorder exclusion in its denials of reimbursement for J.M.’s ABA therapy.

The status of this exclusion was brought into question by legislation effective shortly after J.M.’s diagnosis. In August 2005, the State of Oregon enacted the Mental Health Parity Act (“Parity Act”), which went into effect on January 1, 2007. See Or. Laws 2005, c. 705, § 1, codified at ORS 743.556 (renumbered ORS 743A.168). 3 The Parity Act mandated that “[a] group health insurance policy providing coverage for hospital or medical expenses” must “provide coverage for expenses arising from treatment for ... mental or nervous conditions at the same level as, and subject to limitations no more restrictive than, those imposed on coverage or reimbursement of expenses arising from treatment for other medical conditions.” Id. This language required PacificSource to abandon its prior exclusion for PDDs in the 2006 Plan.

After its passage, PacificSource announced on its website that:

[bjeginning January 1, 2007, Pacific-Source will be managing mental health and chemical dependency treatments consistent with the implementation of Oregon’s new parity rules. We will apply utilization criteria and benefits for both mental health and chemical dependency in a manner similar to those applied to other medical benefits and treatment reviews.

SR 1745.

Pacific Source also provided a table of covered and non-covered diagnoses under the Parity Act and listed autism (299.0) as a covered mental health diagnosis. SR 1746. Accordingly, PacificSource provided coverage in its 2007 Plan effective Novem ber 1, 2007, for “medically necessary services for the treatment of mental and nervous conditions” including autism. SR 1747, 1778. As amended, the 2007 Plan offered coverage for autism in compliance with the Parity Act. However, it retained several exclusions at issue here.

II. Preliminary Issues

As a threshold issue, PacificSource asserts that McHenry is not entitled to reimbursement for the ABA therapy provided to her son before November 1, 2007. McHenry admits that the 2007 Plan did not take effect until November 1, 2007, but argues that PacificSource was obligated to provide coverage when the Parity Act became law on January 1, 2007. ' Moreover, PacificSource expressed its intent to amend its policy language by stating on its website that it would be “managing mental health ... treatments” in compliance with the Parity Act “beginning January 1, 2007.” PacificSource’s actions throughout 2007 repeatedly affirmed that intent. First, PacificSource explicitly relied on the 2007 Plan language in denying McHenry’s claim at each level of review, though it was not technically operative until the time of her second appeal. SR 54-58, 93-97, 351. Second, throughout the course of processing McHenry’s claim, PacificSource employees routinely referenced the 2007 Plan, both internally (SR 89-92, 103-05, 199, 348-49, 390-91) and in communications with the Oregon Insurance Division regarding her claim (SR 94-97, 124, 394-95). Finally, McHenry argues that because PacificSource never relied on the 2006 Plan’s exclusion for autism as a basis for denying her claims throughout her administrative appeals process, it is barred from doing so now.

ERISA requires an employee benefits plan to set forth the specific reasons for an adverse benefits determination at the time of its decision. 29 USC § 1133; 29 CFR § 2560.503—1(g); see Booton v. Lockheed Med. Benefit Plan, 110 F.3d 1461, 1463 (9th Cir.1997) (“If benefits are denied in whole or in part, the reason for the denial must be stated in reasonably clear language, with specific reference to the plan provisions that form the basis for the denial[.]”). In view of this requirement, a plan administrator is not permitted to assert rationales during litigation that it “adduces only after the suit has commenced.” Jebian v. Hewlett-Packard Co. Employee Benefits Organization Income Protection Plan, 349 F.3d 1098, 1104 (9th Cir.2003), cert denied, 545 U.S. 1139, 125 S.Ct. 2956, 162 L.Ed.2d 887 (2005); see also McCoy v. Fed. Ins. Co., 7 F.Supp.2d 1134, 1145 (E.D.Wash.1998) (defendant waived its right to raise an argument on de novo review where it had the opportunity to raise it during the ERISA review process but did not do so, and plaintiff did not acquiesce in defendant’s raising of the issue)

PacificSource never cited the lack of autism coverage under the 2006 Plan as a reason for denying McHenry’s claim during its administrative review. In fact, at every step of the review, it acted as if the 2006 Plan provided coverage and even cited language to McHenry from the 2007 Plan as the basis for its denial. This court declines to now entertain PacificSource’s belated argument that autism was not a covered diagnosis prior to November 1, 2007.

III. Analysis

To be entitled to reimbursement for J.M.’s’treatment, the parties agree that ABA therapy must be medically necessary, a covered benefit under the Plan, and provided by an eligible provider. McHenry has the burden to prove that ABA therapy is a covered benefit under the Plan, and PacificSource has the burden to prove that it falls within an exclusion. See Mario v. P & C Food Mkts., Inc., 313 F.3d 758, 765 (2nd Cir.2002).

A. Medically Necessary

J.M.’s pediatrician, Dr. Shah, has thrice written to PacificSouree indicating that ABA treatment was medically necessary to treat J.M.’s autism. SR 1189-92. Pacific-Source has not challenged J.M.’s diagnosis or Dr. Shah’s opinion that ABA is a medically necessary treatment. Therefore, she satisfies that requirement for coverage.

B. Covered Beneñt

Even if ABA therapy is medically necessary, PacificSouree argues that it is not a covered benefit because it falls under the Plan’s exclusions either for: (1) experimental or investigational procedures; (2) educational services; or (3) academic and social skills training.

1. Experimental or Investigational Procedures

The Plan excludes services for “[experimental or investigational procedures,” defined, in part, as:

Services, supplies, protocols, procedures, devices, chemotherapy, drugs or medicines or the use thereof that are, in PacificSource’s judgment, experimental or investigational for the diagnosis and treatment of the patient. For purposes of this exclusion, experimental or investigational services and supplies include, but are not limited to, services, supplies, procedures ... or the use thereof which at the time they are rendered and for the purpose and in the manner they are being used: ...

75 Are not of generally accepted medical practice in the state of Oregon or as determined by PacificSouree in consultation with medical advisors, medical associations, and/or technology resources; [or]

75 Are not approved for reimbursement by the Centers for Medicare and Medicaid Services[.]

SR 1789.

a. Standard of Review

Despite this court’s earlier ruling on the standard of review, PacificSouree argues that its decision with respect to this exclusion is still entitled to deference because the 2007 Plan commits the determination of which “services” are experimental and investigational to “PacificSource’s judgment.” Id. It points out that other courts have interpreted this language to confer discretionary authority. See Chambers v. Family Health Plan Corp., 100 F.3d 818, 825 (10th Cir.1996) (plan language stating “medical [or] surgical ... procedures ... which in the judgment of [the insurer] are experimental” expressly gave insurer discretion to determine whether to deny a claimant insurance benefits for an “experimental” procedure); Loyola Univ. of Chicago v. Humana Ins. Co., No. 89 C 7855, 1992 WL 80522, at *2 (N.D.Ill. April 14, 1992), aff'd, 996 F.2d 895 (7th Cir.1993). PacificSouree also interprets this court’s prior ruling as recognizing that it retains discretion on this narrow issue. See Opinion and Order (docket # 27), p. 11.

Contrary to PacificSource’s interpretation, this court’s prior ruling did not find that PacificSouree retains the discretion to decide whether the exclusion for experimental and investigational procedures is satisfied. Rather, it unambiguously stated that this language was not sufficient to notify a claimant that the Plan granted discretionary authority to Pacific-Source to determine claims. Absent this broad grant of discretion, the standard of review in the Ninth Circuit is de novo, even where the Plan contains discretionary language as to one element of the Plan. “[A] plan will not sufficiently confer discre tion sufficient to invoke review for abuse of discretion just because it includes a discretionary element. Rather, the power to apply that element must also be ‘unambiguously retained’ by the administrator.” Sandy v. Reliance Std. Life Ins. Co., 222 F.3d 1202, 1204 (9th Cir.2000) (citation omitted).

b. Generally Accepted Medical Practice

PacifieSource first argues that ABA therapy is experimental or investigational, as those terms are defined by the Plan, because it is not the generally accepted standard of care for autism in Oregon or anywhere else. In making this determination, PacifieSource relied exclusively on the opinion of its Chief Medical Officer, Steven D. Marks, M.D., and offers his declaration explaining his rationale for finding that ABA therapy falls within this exclusion. McHenry objects to the admission of this declaration on the grounds that it is outside the administrative record.

This court has discretion to allow additional evidence not before the plan administrator, but should exercise this discretion “only when circumstances clearly establish that additional evidence is necessary to conduct an adequate de novo review of the benefit decision.” Mongeluzo, 46 F.3d at 944 (citation omitted). One such circumstance is where a claim requires “consideration of complex medical questions or issues regarding the credibility of medical experts.” Opeta v. Nw. Airlines Pension Plan, 484 F.3d 1211, 1217 (9th Cir.2007) (citation and internal quotation marks omitted).

Dr. Marks is the sole expert relied upon by PacifieSource for arguing that ABA therapy falls within one of the Plan’s exceptions. He formulated PacifieSource’s autism policy, including its rejection of ABA therapy as a covered benefit. His declaration is primarily a summary of his reasoning for choosing to reject ABA therapy based upon the Plan’s exclusions. Considering the complexity of the medical issues in this case, McHenry’s objection is overruled.

Dr. Marks states that in the course of developing the PacifieSource policy to reject coverage for ABA therapy for autism (SR 25-29), he “read many articles and textbook chapters, along with doing some intensive internet searches to better understand the current state of treatment for autism.” Marks Deck, (docket # 50), ¶ 4. After considering all of these materials, he concluded that the “consensus of all that I read was that there was and is no cure for autism.” Id. Rather, “each treatment modality had its supporters and its detractors!,] • • • there is no ‘gold standard’ for the treatment of autism, and there is much debate in the literature regarding the efficacy of any one approach, including ABA.” Id. From his review of the literature and his own experience as a practitioner, “it became clear that ABA was not a well-proven or evidence-based standard of medical care, nor was it a standard of coverage within the industry.” Id., ¶ 5. 4

McHenry attacks Dr. Marks opinion on multiple fronts. First, she deems it irrelevant since he is not an expert in treating autism or other PDDs. Second, McHenry counters it with the opinion of Karen Grant, Psy.D. SR 110-12. Unlike Dr. Marks, Dr. Grant actively practices and does research in the field of autism treat ment. She opines that based on “33 years of research!,] ... ABA therapy is not only an empirically supported and validated treatment, but ... is also a long standing treatment for individuals with autism,” and cites to numerous articles to support her conclusion. SR 110-11. Third, McHenry cites a raft of scientific articles to contradict the notion that ABA therapy is experimental or investigational. Fourth, McHenry points to numerous government and state agencies which have concluded that “ABA-based procedures represent best practices for individuals with autism. ...” SR 968. 5 Fifth, McHenry notes that Dr. Marks has not been consistent in his position. Early on in the handling of J.M.’s claim, he indicated a favorable opinion of ABA therapy, stating “ideally I’d like to see these kids get into an ABA-type program that we could contract for on a case rate basis.” SR 14. Finally, McHenry submits a recent external review obtained by the Oregon Insurance Division which concluded that ABA therapy was medically necessary for the treatment of autism and that denying ABA therapy was not consistent with national standards of care. SR 117-18.

Based upon a thorough examination of the record, this court concludes that the weight of the evidence demonstrates that ABA therapy is firmly supported by decades of research and application and is a well-established treatment modality of autism and other PDDs. It is not an experimental or investigational procedure. Dr. Grant’s opinion corresponds with this court’s overall impression of the scientific consensus surrounding ABA therapy after reviewing each of the studies in the record. Moreover, because she is an expert in the field, Dr. Grant’s opinion is much more persuasive than that of Dr. Marks. From a review of the numerous articles and other material in the record, this court finds no basis for Dr. Marks’s opinion that “ABA was not a well-proven or evidence— based standard of medical care, nor was it a standard of coverage within the industry.” Indeed, just the opposite is the case.

This court’s view of the science is shared by multiple government agencies and professional organizations. For example, in 1999, the Department of Health and Human Services (“DHS”) issued a report on the state of mental health and mental health treatment in the United States. One of its findings was that “[t]hirty years of research demonstrated the efficacy of applied behavioral methods in reducing inappropriate behavior and in increasing communication, learning, and appropriate social behavior.” SR 975 (DHS, Mental Health: A Report of the Surgeon General, c. 3., p. 163 (1999), available at http://www. surgeongeneral.gov/library/ mentalhealth/ pdfs/c3.pdf) (last accessed Jan. 5, 2010). The National Institute of Mental Health (“NIMH”) has similarly concluded that “[a]mong the many methods available for treatment and education of people with autism, applied behavior analysis (ABA) has become widely accepted as an effective treatment.” SR 1106 (NIMH, Autism Spectrum Disorders: Pervasive Develop mental Disorders, Doc. No. NIH 08-551 (2008), available at http://www.nimh.nih. gov/health/publications/autism/nimhautism spectrum.pdf) (last accessed Jan. 5, 2010). Professional organizations concluding that ABA therapy is an appropriate treatment for autism include the American Association on Mental Retardation, American Psychological Association, and the Association for Science in the Treatment of Autism. SR 968.

That a given mental disorder has no absolute cure is not a basis for rejecting treatments which purport to alleviate or ameliorate its symptoms. Many treatments purport only to alleviate symptoms or increase the quality or length of life of those suffering from a chronic, incurable disease. Furthermore, the fact that ABA therapy is not effective for every autistic child is not a reasonable basis for concluding that it is experimental or investigational. It is possible for a treatment to be both well-established and of limited efficacy in curing a neurological or mental disorder. Likewise, a scientific debate on the degree of improvement provided by a treatment or the instances in which it is the most effective does not show that the treatment is experimental or investigatory. The great majority of the studies in the record indicate that ABA therapy is not only supported by decades of research, but is one of the only autism treatments which has consistently shown measurable success in improving the lives of autistic children. 6 These studies, and the other sources cited above, demonstrate that ABA therapy has become one of the standard treatment options for autistic children throughout the nation. Notably, other than Dr. Marks’s summary opinion, PaeificSource has pointed to no authority in the record that has labeled ABA therapy an experimental or investigational treatment for autistic children or that declare it to be not a standard of care in Oregon or anywhere else.

Here PaeificSource has submitted only one piece of evidence in support of its conclusion, namely the opinion of its own Chief Medical Officer. In light of the wealth of conflicting scientific research supporting ABA therapy, it was not reasonable for PacifieSource to rely on Dr. Marks’s opinion alone. As a result, this court concludes that ABA therapy is not experimental or investigational in nature and that PacifieSource lacked a reasonable basis reaching the opposite conclusion,

c. Approved by the Centers for Medicare and Medicaid Services

Second, PacifieSource argues that ABA therapy is not approved for reimbursement by the Centers for Medicare and Medicaid Services (“CMS”). It relies upon a letter sent from the Oregon Department of Human Services (“ODHS”) to McHenry’s attorney explaining that it “does not currently recognize BCBA as a specific provider type,” but “therapists, with a specialty of BCBA, can be enrolled with the Department as an approved County Mental Health Program (CMHP) provider” and may “bill any appropriate covered procedure codes, including autism.” Shaw Decl. (docket #43), ¶2 & Ex. A, p. 1 (emphasis in original). 7 The letter plainly does not state that ABA is not reimbursable, but states only that BCBAs as a provider type are not recognized by the ODHS or CMS. Notably, the letter provides a specific method by which ABA therapy could be successfully billed. Thus, it does not provide a reasonable basis for concluding that ABA therapy is not approved for billing by the CMS. 8 Thus, PacifieSource has failed to show that ABA therapy falls within the exclusion for experimental or investigative treatments.

2. Educational Services

PacifieSource argues that ABA therapy, even if not experimental or investigatory, is excluded as “educational or correctional services or sheltered living provided by a school or halfway house.” SR 1790.

As support, PacifieSource points to Dr. Grant’s statement that “ABA intervention for children on the autism spectrum have been shown over time to be highly effective in teaching and generalizing skills for these children in all areas of difficulty.” SR 110 (emphasis added). Additionally, some of the articles cited by McHenry use language seemingly indicative of educational or social training. See, e.g., SR 967 (“ABA-based approaches for educating children with autism and related disorders have been extensively researched and empirically supported.”) (emphasis added); SR 1106 (“Among the many methods available for treatment and education of people with autism, applied behavior analysis (ABA) has become widely accepted as an effective treatment.”) (emphasis added). Even advocates of ABA therapy describe it in terms that suggest it is educationally based. For example, Hoyt’s website states that ABA therapy “[i]instruction focuses on teaching Core Learning Skills, Verbal Behavior, and social/play skills in natural and structured learning environments.” SR 188 (emphasis added). Aso, of the six categories of “treatment options” identified by the Autism Society for America (“ASA”), the ASA placed ABA therapy under the “Educational” category. SR 1089.

According to Dr. Marks, these sources agree with the literature he reviewed on ABA which “frequently referred to the persons receiving the therapy as ‘learners’; the plans for working with the child as ‘curricula’; referenced ‘teacher/instructors,’ and ‘teacher/learner’ ratios; and talked about teaching various skills in ‘structured learning environments.’ ” Marks Deck, ¶ 6. He concludes that “applied behavioral analysis was more akin to remedial education and ‘generalization’ skill techniques, and not clinical treatment per se.” Id, ¶ 7. As a result, in his view, ABA therapy is properly classified along side special education classes or individualized education plans utilized to assist children with learning disabilities.

However, the full sentence of the exclusion reads as follows: “This plan does not cover educational or correctional services or sheltered living provided by a school or halfway house, except outpatient services received while temporarily living in a shelter!.]” PaeifieSource reads the clause “provided by a school or halfway house” as modifying only “sheltered living.” However, there is no comma separating “educational or correctional services” and “or sheltered living.” As a result, the clause “provided by a school or halfway house” may be read as not only modifying “sheltered living,” but also as modifying “educational or correctional services.” Given this ambiguity, the language must be construed against PaeifieSource and in favor of McHenry. McClure v. Life Ins. Co. of N. Am., 84 F.3d 1129, 1134 (9th Cir.1996) (“ERISA insurance policies are governed by the rule that ambiguous language is construed against the insurer and in favor of the insured”). Construing the language most favorably to McHenry, even if ABA treatment were “educational,” it is excluded only if it is “provided by a school or halfway house.” J.M.’s services were not provided by a school or halfway house, but by an employee of a private company that provides rehabilitative services for autistic children. Thus, this exclusion does not apply.

3. Academic and Social Skills Training

Finally, PaeifieSource relies on the Plan’s exclusion for “academic skills training ... and social skills training.” SR 1790. While acknowledging that ABA therapy may benefit an autistic child’s academic and social skills, McHenry counters that its primary focus is on modifying behaviors pertinent to every area of that child’s life.

As discussed above, autistic children may exhibit many types of problem behavior detrimental to social or academic progression. A list assembled by one article includes: aerophagy/swallowing, aggression, bruxism/teethgrinding, coprophagy/feees eating, dawdling, destruction, depression, disruption/tantrum, drooling, elective mutism, elopement (run), feces smearing, fears, food refusal, food theft, genital stimulation, hallucinating, hyperactive behavior, hyperventilation, inappropriate vocalizations, insomnia, noncompliance, obesity, obsessive compulsive disorder, pica, public disrobing, rapid eating, rectal digging, rumination, seizure behavior, self-injurious behavior, stereotypy, tongue protrusion, and vomiting. SR 1235 (Robert H. Horner, et al, Problem Behavior Interventions for Young Children with Autism: A Research Synthesis, 32 J. Autism and Developmental Disorders 423, 431 (October 2002)).

It is reasonable to assume that a child exhibiting some of these behaviors would face serious obstacles to academic and social development. Autism’s noted adverse impact on the ability of a child to form social connections or to express empathy or even awareness of another would have similar severe impacts in these areas. Indeed, the impairments caused by autism are acutely social in nature and the diagnostic criteria for autism require some “qualitative impairment in social interaction” in order to affirm a positive diagnosis. DSM-IV-TR at 70-71. Given the inherently social nature of the behavioral impairments caused by autism and the negative impacts of some of these behaviors on a child’s academic development, it is no surprise that ABA therapy seeks to modify this behavior.

While ABA therapy may have beneficial effects on an autistic child’s social and academic skills, its defining characteristic is application of techniques to modify behavior in every area of an autistic child’s life. In this regard, a sports analogy is instructive. While participation in sports can benefit a students academic and social skills, no one would classify sports as academic or social skills training. Similarly, the incidental benefits in these areas resulting from ABA therapy, while real, do not dictate that it be classified as either as academic or social skills training. Rather, it is more properly classified as behavioral modification.

PacificSouree’s contrary interpretation would sweep many other covered benefits into this exception to which it clearly does not apply. Nearly all types of psychological treatment (counseling, psychotherapy, etc.) could be classified as academic or social skills training. These types of treatments, like ABA therapy, undoubtedly have benefits on a person’s ability to succeed in education and help to teach proper skills and behaviors for social interactions. However, they would presumably not fall within those exclusions.

The focus of ABA therapy on discrete behaviors affecting all facets of living sets it apart. Researchers have found ABA to be effective in reducing problem behaviors, SR 1233 (Horner, supra, at 429), and in improving a child’s ability to function in multiple areas including “intellectual, social, emotional, and adaptive functioning.” SR 1252 (Svein Eikeseth, et al, Outcome for Children with Autism Who Began Intensive Behavioral Treatment Between Ages I and 7, 31 Behavior Modification 264, 265 (2007). While aimed at improving social and academic functioning, it does this by specifically addressing behavioral deficits possessed by autistic children that interfere with every area of their life, not by educating kids on social norms or teaching study skills or other tools specific to academic success. To find for Pacific-Source on this issue would be to improperly stress the benefits of ABA therapy in only two out of many areas of functioning.

According to the weight of the evidence, ABA therapy is not primarily academic or social skills training, but is .behavioral training. Accordingly, it is not subject to the exclusions under the Plan for academic or social skills training.

C. Eligible Provider

Although ABA therapy is medically necessary to treat J.M.’s autism, does not fall within any exclusion, and thus is a covered benefit under the 2007 Plan, McHenry is not entitled to reimbursement unless it is provided by an eligible provider. See SR 1772-74, 1778-79. The 2007 Plan defines eligible providers for mental health treatment as follows:

2. Provider Eligibility. A provider is eligible for reimbursement if:

a. The provider is approved by the Department of Human Services;

b. The provider is accredited for the particular level of care for which reimbursement is being requested by the Joint Commission on Accreditation of Hospitals or the Commission on Accreditation of Rehabilitation Facilities; or

c. The patient is staying overnight at the facility and is involved in a structured program at least eight hours per day, five days per week; and

d. The provider is providing a covered benefit under this policy; and

e. The provider meets the credentialing requirements of Pacific-Source.

SR 1778.

The 2007 Plan further defines “provider” as “a person who meets the credentialing requirements of PacificSource, is otherwise eligible to receive reimbursement under the policy, and is ...; v. An individual behavioral health or medical professional authorized for reimbursement under Oregon law.” Id.

The Member Benefits Handbook (or Summary Plan Description (“SPD”)) contains a slightly different description of eligible providers of mental health services as persons or facilities:

that meet the credentialing requirements of PacificSource, if credentialing is required, are otherwise eligible to receive reimbursement for coverage under the policy and are either a health care facility, a residential program or facility, a day or partial hospitalization program, an outpatient service, or an individual behavioral health or medical professional authorized for reimbursement under Oregon law.

SR 1837.

Both the 2007 Plan 9 and SPD 10 provide a list of eligible providers. These lists are materially the same with BCBAs notably absent from both. That absence is immaterial because the SPD and Plan state only that “eligible providers include” and not “eligible providers are limited to ” or similar exclusive language. See Ariz. State Bd. For Charter Sch. v. U.S. Dep’t of Educ., 464 F.3d 1003, 1007 (9th Cir.2006) (“In both legal and common usage, the word ‘including’ is ordinarily defined as a term of illustration, signifying that what follows is an example of the preceding principle.”), citing Black’s Law Dictionary, 777-78 (8th ed.2004) (“[t]he participle including typically indicates a partial list”). By way of contrast, the 2006 Plan included a restrictive clause, stating that “[o]nly the following providers ... are eligible for reimbursement under this policy.” SR 1722. PacificSource’s decision to eliminate the restrictive clause in the 2006 Plan and replace it with a word commonly understood to proceed only a partial list is strong evidence that it did not intend the list of eligible providers in the 2007 Plan to be exhaustive.

Even so, PacificSource argues that Hoyt cannot be included with the other listed eligible provider types because she lacks the one attribute common to all others listed, namely, state licensing. While that may be true, the 2007 Plan nowhere explicitly requires state licensure as a precondition to provider eligibility. Thus, if Hoyt were to meet the explicit criteria set forth in the 2007 Plan, this nonexclusive list would not disqualify her even though she may not share one of the common features.

Turning to those explicit criteria, the court first recognizes not only significant overlap between the 2007 Plan’s definitions of “provider” and “eligible provider” and the SPD’s definition of “eligible provider,” but also important distinctions. Combining the two terms used in the 2007 Plan to remove redundancies, Hoyt must: (1) be approved by ODHS; (2) meet Pacific-Source’s credentialing requirements; (3) be authorized for reimbursement under Oregon law; and (4) provide a covered benefit or be “otherwise eligible” to receive reimbursement for coverage under the policy. The criteria in the SPD differ from these four elements in two important ways: first, the SPD only requires credentialing “if credentialing is required;” and second, there is no requirement that Hoyt be approved by ODHS. McHenry argues that the SPD controls.

To resolve a disagreement between plan documents, the court must adopt the language most favorable to the claimant. See Bergt v. Retirement Plan for Pilots Employed by MarkAir, Inc., 293 F.3d 1139, 1145 (9th Cir.2002). Accordingly, the court finds that the more favorable elements in the SPD control. In view of the evidence, as discussed below, it seems likely that being approved by the ODHS is the method of being authorized for reimbursement under Oregon law. However, in the event of a distinction that neither party has pointed out, the court follows the terms of the SPD and removes approval by ODHS as one of the criteria.

Therefore, because Hoyt was providing a covered benefit under the Plan, as discussed above, to be eligible for reimbursement, McHenry must prove: (1) either that Hoyt met PacificSource’s credentialing requirement or that PacificSource did not require her to be credentialed, and (2) that Hoyt was authorized for reimbursement under Oregon law.

1. Credentialing

It is undisputed that Hoyt has not been credentialed by PacificSource and does not meet its credentialing requirements. Instead, McHenry argues that no credentialing is required for Hoyt because she was a nonparticipating provider or a network not available provider. 11 Neither the 2007 Plan nor the SPD defines PacificSource’s credentialing requirements or describes the process for becoming credentialed. These requirements are explained in two other documents, namely, the Physician and Provider Manual (“Provider Manual”) (SR 735) and the Provider Network Management Credentialing Manual (“Credentialing Manual”) (SR 870). As described by § 4.2 of the Provider Manual, the credentialing process “includes meticulous verification of the education, experience, judgment, competence, and licensure of all healthcare providers.” SR 755. The process is described in outline form in the Provider Manual and in greater detail in the Credentialing Manual.

Significantly, the Provider Manual states that if, after the credentialing process is completed, “the Credentialing Committee does not approve the provider, the provider may be considered a ‘nonparticipating provider’ and claims may be processed at the nonparticipating benefit level.” SR 756. Based on this language, McHenry argues that only those providers who wish to be participating providers must pass the certification process. For those who cannot, the Provider Manual expressly provides for the option of reimbursing them at the non-participating provider rate.

Nothing in the Credentialing Manual contradicts the Provider Manual. Indeed, the general statement of policy on the first page of the Credentialing Manual reads: “PacificSource makes every effort to contract with qualified participating practitioners by using appropriate credentialing standards.” SR 870. This language confirms McHenry’s argument that credentialing is related to issues of contracting with approved providers. The remainder of the Credentialing Manual describes in detail the requirements necessary to become and remain a participating provider through the credentialing process. It also requires that “[a]ll participating practitioners will be recredentialed at a minimum of every three years (36 months).” SR 876. The Credentialing Manual provides no similar recredentialing requirements for nonparticipating providers.

PacificSource argues against McHenry’s interpretation by pointing to Section 4.2.4 of the Provider Manual which provides a limited exception from credentialing for “providers who practice exclusively within the inpatient setting and who provide care for the health plans’ members only as a result of members being directed to the hospital or other inpatient setting.” SR 757. According to PacificSource, this is the only class of providers who need not be credentialed.

The record reveals little else to resolve this issue. The 2007 Plan defines a nonparticipating provider as “a provider of covered medical services or supplies that does not directly or indirectly hold a provider contract or agreement with Pacific-Source.” SR 1752. This merely returns the reader to the definition of “provider” in the 2007 Plan requiring the person to be “credentialed.” On the other hand, the SPD indicates that credentialing may not always be required and the Provider Manual states that a non-credentialed person may be reimbursed at the nonparticipating provider rate. While the Provider Manual contains only one explicit exception to the credentialing requirement, it also explicitly contemplates reimbursing a person for services provided by a practitioner who fails to meet PacificSource’s credentialing requirements.

Given these conflicting provisions and the lack of a clear indication that all providers must be credentialed, the 2007 Plan is, at best, ambiguous on this issue. Given this ambiguity, the court must adopt the interpretation most favorable to McHenry. McClure, 84 F.3d at 1134. Consequently, the court finds that Hoyt need not be credentialed with PacificSource to be considered an eligible provider.

2. Authorized for Reimbursement Under Oregon Law

Both the Plan and the SPD require an eligible provider to be authorized for reimbursement under Oregon law. The parties agree that ORS 743A.168(5) provides the applicable standards:

(5) A provider is eligible for reimbursement under this section if:

(a) The provider is approved by the Department of Human Services;

(b) The provider is accredited for the particular level of care for which reimbursement is being requested by the Joint Commission on Accreditation of Hospitals or the Commission on Accreditation of Rehabilitation Facilities;

(c) The patient is staying overnight at the facility and is involved in a structured program at least eight hours per day, five days per week; or

(d) The provider is providing a covered benefit under the policy.

The four separate requirements are phrased in the disjunctive, meaning McHenry need satisfy only one to be “eligible for reimbursement.” The final requirement renders a provider eligible for reimbursement simply by “providing a covered benefit under the policy.” McHenry argues that because Hoyt was providing such a benefit (ABA therapy), she is eligible for reimbursement under Oregon law and thus satisfies the final requirement to be an eligible provider under the 2007 Plan.

But in order to be eligible for reimbursement, ORS 743A.168(5), like the 2007 Plan, requires that the person first be a “provider.” In turn, a “provider” must have “met the credentialing requirements of a group health insurer,” be “otherwise eligible to receive reimbursement for coverage under the policy” and be “[a]n individual behavioral health or medical professional authorized for reimbursement under Oregon law.” ORS 743A.168(l)(e) These elements are identical to the 2007 Plan’s definition of provider. The final element is the critical one here, namely that Hoyt has been, or could be, authorized for reimbursement.

Here McHenry’s evidence consists only of the ODHS letter responding to her attorney’s inquiry as to whether “[BCBAs] who treat children with [ASDs] are approved as providers by the Department of Human Services.” Shaw Deck, ¶ 2, & Ex. A, p. 1. ODHS responded that the department “does not currently recognize BCBA as a specific provider type” but that “therapists, with a specialty of BCBA, can be enrolled with the Department as an approved County Mental Health Program (CMHP) provider.” Id. (emphasis in original). Once enrolled as a CMHP provider, the therapist would be “able to bill any appropriate covered procedure codes, including autism, which is a covered diagnosis for the Oregon Health Plan[.]” Id.

This letter fails to establish that Hoyt is “authorized for reimbursement.” The letter merely poses a hypothetical situation in which a provider could bill the ODHS for ABA therapy. It does not establish that Hoyt fits within that hypothetical. Under its terms, to be authorized for reimbursement, a provider must be, at minimum, (1) a therapist with a specialty in BCBA and (2) enrolled with the ODHS as an approved CMHP provider. Even assuming that the ODHS would classify Hoyt as a therapist with a specialization in BCBA, there is no evidence that she has been, or could be, enrolled as an approved CMHP provider. 12 Nothing in the record defines this class of providers or describes the process to become one. There is no indication that Hoyt has ever attempted to become one, or that she is even capable of doing so. In short, the record fails to establish that Hoyt, a professional with a BCBA certification and nothing more, is authorized for reimbursement under Oregon law. Indeed, in the absence of any basis for authorization other than ODHS approval, the record affirmatively forecloses the possibility as ODHS “does not recognize BCBA as a specific provider type.” Id.

Because nothing in the record demonstrates that Hoyt is authorized for reimbursement for Oregon law, McHenry has failed to prove that Hoyt satisfies the definition of eligible provider under the 2007 Plan.

D. Alternate Bases for Recovery 1. Limited Coverage for Ineligible Mental Health Providers

McHenry offers two additional arguments for finding in her favor. The first relies upon the PacificSource internal policy titled “Administrative Procedure: Request for Ineligible Mental Health Providers.” SR 725. This internal policy permits PacificSource to extend coverage for six visits to an otherwise ineligible provider where there exists “[license equivalency,” “[njetwork accessibility” issues, or other special circumstances. Id. The six visits are intended to provide for “transitional care to an eligible provider[,]” but “[i]f compelling reasons and special circumstances are demonstrated, the Medical Director may approve additional visits.” Id. Any benefits approved for an ineligible provider “are subject to nonparticipating provider benefit rates for approved services.” Id.

McHenry argues that in light of the overwhelming evidence in the record demonstrating the necessity and efficacy of ABA therapy in this case and the utter absence of any other participating providers available in Clackamas County to provide ABA therapy (see SR 253-55, 1158), this policy should provide benefits even if Hoyt is not an eligible provider.

Even if she is correct, McHenry has not demonstrated the basis on which this court could enforce this policy. The policy appears to be a wholly discretionary internal procedure for handling claims which, in PacificSource’s judgment, merit limited special consideration despite the lack of coverage under the terms of the 2007 Plan. It does not appear in the 2007 Plan or the SPD and is not otherwise incorporated into the Plan. In an action pursuant to 29 USC § 1132(a)(1), the plaintiff is only entitled to pursue or clarify benefits or rights due him “under the terms of the plan.” There is no basis for this court to expand those terms to include discretionary internal policies adopted by PacificSource.

2. Illusory Contract

Finally, McHenry argues that if J.M. is not entitled to ABA therapy under the 2007 Plan, then its purported coverage for autism is illusory. In construing a contract, “an interpretation which gives a reasonable, lawful, and effective meaning to all the terms is preferred to an interpretation which leaves a part unreasonable, unlawful, or of no effect.” RESTATEMENT (SECOND) OF CONTRACTS § 203(a) (1981), quoted in U.S. v. Franco-Lopez, 312 F.3d 984, 991 (9th Cir.2002). Thus, “the provisions of an ERISA plan should be construed so as to render none nugatory and to avoid illusory promises.” Carr v. First Nationwide Bank., 816 F.Supp. 1476, 1493 (N.D.Cal.1993) (citations omitted).

According to McHenry, ABA therapy is the “gold standard” of autism treatment, such that to exclude ABA therapy is to not treat autism. Therefore, she argues, if the Plan does not cover ABA therapy, then its autism coverage is purely illusory. It is equally illusory, then, to find that the 2007 Plan covers ABA therapy, but to construe its provider eligibility requirements to eliminate the only providers of ABA therapy-

In support of her argument, McHenry cites K.F. ex rel. Fry v. Regence Blueshield, No. C08-0890RSL, 2008 WL 4330901, at *4 (W.D.Wash. Sept. 19, 2008), where the court confronted a similar situation. In Fry, an ERISA-governed medical benefits plan provided home health care for medically necessary inpatient care. The plaintiff sought payment for hourly nursing services to provide that care, However, the plan expressly excluded payment for hourly nursing services. The court concluded that interpreting the plan to exclude in-home nursing would render its promise of substituted services illusory in most circumstances because one of the primary reasons for inpatient care is round-the-clock nursing services. More importantly, the court found that the exclusion for hourly nursing services did not clearly apply to the substituted service provision. Under the doctrine of reasonable expectations,

[a]n insurer wishing to avoid liability on a policy purporting to give general or comprehensive coverage must make exclusionary clauses conspicuous, plain, and clear, placing them in such a fashion as to make obvious their relationship to other policy terms, and must bring such provisions to the attention of the insured.

Id at *4, quoting Saltarelli v. Bob Baker Group Med. Trust, 35 F.3d 382, 386 (9th Cir.1994).

By violating this doctrine, the court held that the exclusion did not apply.

Unlike the exclusion at issue in Fry, the eligible provider term in the 2007 Plan is a clear condition of coverage on which McHenry bears the burden of proof. To be an eligible provider, McHenry must prove that Hoyt was authorized for reimbursement under Oregon law. As discussed above, the evidence submitted by McHenry fails to meet that burden of proof. Therefore, the reasonable expectations doctrine is inapplicable to bar the exclusion that eliminates coverage here.

Moreover, there is insufficient evidence in the record to conclude that eliminating coverage for BCBAs would eliminate all coverage of ABA therapy under the 2007 Plan. The ODHS letter explicitly posits a scenario in which a practitioner providing ABA therapy would be authorized for reimbursement under Oregon law. Unfortunately, neither the letter nor anything else in the record establishes whether such a practitioner exists. In 2007, McHenry’s husband contacted all of the participating mental health care providers in Clackamas County and found that none of them provides ABA therapy. SR 253-55, 1158. However, this evidence does not establish that no ABA therapy practitioners are available who would meet the eligibility requirements of the 2007 Plan.

To the extent that other providers of ABA therapy are available to McHenry, or that Hoyt could become authorized for reimbursement herself by following the procedure outlined in the ODHS letter but has failed to do so, the lack of coverage is due to McHenry choosing a provider who is not covered by the Plan. That Hoyt is not authorized for reimbursement under Oregon law is solely a product of Oregon law, not an illusory contract of insurance. In that case, McHenry’s remedy is with the Oregon State Legislature or the ODHS.

If the record established that no other possible providers of ABA therapy can be found within a reasonable geographic area, then the potential of illusory coverage would be much stronger. However, the record does not affirmatively establish that fact. Absent such evidence, the court is reticent to override the eligible provider provisions in the 2007 Plan as creating illusory coverage for autism. The specific provisions at issue are adopted wholesale out of Oregon’s insurance code and, thus, reflect not only the bargain struck between McHenry and PacificSource, but also Oregon’s public policy.

The requirement that Hoyt be authorized for reimbursement under Oregon law is not an unreasonable condition in the 2007 Plan. The purpose of the requirement appears to be to ensure that providers are subject to a state-sanctioned governing body which is able to set standards and exercise control over its members. Lacking such oversight of providers of ABA therapy, PacificSource would have no way to assure that the services being provided to its members are legitimate or uniform.

The court recognizes the hardship that its ruling may impose on McHenry and her family. However, ERISA only authorizes this court to grant benefits as provided for in the plan. The services provided by Hoyt are not covered under the 2007 Plan. Therefore, the court must deny McHenry’s motion and grant PacificSource’s cross-motion.

FINDINGS OF FACT

1.ABA therapy is medically necessary to treat J.M.’s autism.

2. PacificSource has failed to establish that ABA therapy is an investigational or experimental treatment as those terms are defined by the 2007 Plan.

3. PacificSource has failed to establish that ABA therapy is educational as that term is defined by the 2007 Plan.

4. PacificSource has failed to establish that ABA therapy is academic or social skills training as those terms are defined by the 2007 Plan.

5. McHenry has failed to establish that Hoyt is authorized to receive reimbursement under Oregon law.

CONCLUSIONS OF LAW

1. ABA therapy does not fall within any exclusion under the 2007 Plan and is therefore a covered benefit.

2. Hoyt is not an eligible provider under the 2007 Plan.

3. Under the terms of the 2007 Plan, McHenry is not entitled to reimbursement for the services provided by Hoyt.

ORDER

McHenry’s Motion for Summary Judgment (construed as a motion for judgment on the record) (docket #41) is DENIED and defendants’ Cross-motion for Summary Judgment (construed as a cross-motion for judgment on the record) (docket # 47) is GRANTED.

DATED this 5th day of January, 2010.

1 . “SR” refers to the stipulated record filed by the parties on May 22, 2009 (docket # 46).

2 . The stipulated record contains many articles from scientific and academic journals, government publications, websites, and other sources which McHenry submitted during the course of the administrative appeals process. These articles are cited by the page(s) on which they appear in the stipulated record and, if published, to the appropriate journal or publication. For government publications or other articles, a parallel citation to the website at which the article is available is given for the reader’s convenience to the extent practicable.

3 . ORS 743.556 was renumbered as ORS 743A.168 in 2007. The newer citation is used for ease of reference.

4 . A partial list of the sources Dr. Marks' relied upon in reaching his conclusion are appended to the PacifieSource Health Service Procedure: Autism — Draft II, the development of which Dr. Marks’ oversaw. Marks Aff., ¶ 8; SR 28-29. Dr. Marks represents that these sources included some articles that supported ABA therapy and some articles calling into question the validity of the studies used by supporters of ABA.

5 . An online article from the Kennedy Krieger Institute lists many of these entities including: National Institute of Mental Health, National Institute of Child Health and Human Development, The National Academies Press, American Association on Mental Retardation, American Psychological Association, Association for Science in the Treatment of Autism, the Surgeon General of the United States, New York State Department of Health, California State Department of Developmental Services, Florida State Department of Children and Families, and Maine Administrators of Services for Children with Disabilities. SR 968 (Louis P. Hagopian & Eric W. Boelter, Applied Behavioral Analysis: Overview and Summary of Scientific Support, available at http://www.kennedykrieger.org/kki_misc.jsp? pid=4761) (last accessed Jan. 5, 2010).

6 . See, e.g., SR 894 (Francis, supra, at 495 (finding "[t]he literature shows that intensive behavioural therapy clearly benefits children with autism and yields a high degree of parental satisfaction; however, the original effectiveness claim was overstated and its cost-effectiveness, in terms of time, effort, and money, has not been adequately assessed”)); SR 900 (Scott O. Lilienfeld, Scientifically Unsupported and Supported Interventions for Childhood Psychopathology: A summary, 115 Pediatrics 761, 762 (2005) ("The most efficacious psychosocial treatment for autism is applied behavior analysis.... In controlled within-subject studies, applied behavior analysis has demonstrated positive effects on autistic children's social and intellectual behaviors, although almost all of these children are left with serious deficits in adaptive functioning.”); SR 913-15 (Howlin, supra, pp. 111-13 (finding that of all the treatments for autism, early behavioral intervention enjoys the most scientific support although there remain unanswered questions about its total efficacy and proper methodology)); SR 926 (Barbaresi, supra, at 1171 (“Decades worth of scientific research provide clear and convincing support for the technique referred to as [ABA].”)); SR 932 (Filipek, supra, at 208 ("Behavioral, as opposed to pharmacologic, treatment is the hallmark of effective intervention for everyone with autism.”)); SR 1323-34 (Dillenburger, supra, at 120-21 (noting that "[ejxtensive research over 30 years shows that early intensive behavioural intervention can lead to significant gains in cognitive, social, emotional, and motor functioning that can be generalised to other situations and maintained in the long term” and that "[a] review of over 500 studies shows that ABA consistently offers positive outcomes in terms of educating children with ASD and enhancing life skills”)); SR 1471 (Remington, supra, at 418 (noting that an "increasing body of empirical research suggests that early, intensive, structured intervention, based on principles of applied behavioral analysis, is effective in remediating the intellectual, linguistic, and adaptive deficits associated with autism” and reporting that a two-year study conducted by the authors further confirmed this research)).

7 . Whether a treatment is approved for reimbursement by the CMS presumably is the basis for the ODHS approving it under the Oregon Health Plan. Neither party has addressed this issue.

8 . Cf. Parents League for Effective Autism Servs. v. Jones-Kelley, 565 F.Supp.2d 905, 915-16 (S.D.Ohio 2008) (granting TRO after finding that plaintiffs had a strong likelihood of success on their claim that ABA therapy was compensable under federal medicaid law), aff’d in unpublished opinion, 339 Fed.Appx. 542 (6th Cir.2009).

9 . The 2007 Plan provides that:

Eligible providers include:

a. A program licensed, approved, established, maintained, contracted with, or operated by the Mental Health Division for Alcoholism;

b. A program licensed, approved, established, maintained, contracted with, or operated by the Mental Health Division for Drug Addiction;

c. A program licensed, approved, established, maintained, contracted with, or operated by the Mental Health Division for Mental or Emotional Disturbance;

d. a. A medical or osteopathic physician licensed by the State Board of Medical Examiners;

e. A psychologist (Ph.D.) licensed by the State Board of Psychologists’ Examiners;

f. A nurse practitioner registered by the State Board of Nursing;

g. A clinical social worker (LCSW) licensed by the State Board of Clinical Social Workers;

h. A Licensed Professional Counselor (LPC) licensed by the State Board of Licensed Professional Counselors and Therapists;

i. A Licensed Marriage and Family Therapist (LMFT) licensed by the State Board of Licensed Professional Counselors and Therapists; and

j. A hospital or other healthcare facility licensed for inpatient or residential care and treatment of mental health conditions and/or chemical dependency.

SR 1779.

10 . Omitting the institutional providers, the SPD provides that:

"Eligible providers include: Licensed medical or osteopathic physicians (M. D. or D. O.), including psychiatrists, licensed psychologists (Ph. D.) and psychology associates, registered nurse practitioners (N. P.), licensed clinical social workers (L. C. S. W.), licensed professional counselors (L. P. C.), and licensed marriage and family therapists (L. M. F. T.).

11 . A network not available provider is a nonparticipating provider located in an area where the member does not have reasonable access to a participating provider. SR 1771, 1823. The designation affects only the reimbursement rate; there appears to be no distinction between the two in terms of credentialing requirements.

12 . ORS 430.610-695 provides for the creation and oversight of CMHPs, and OAR 309-14-0020 provides specific requirements for the establishment and management of CMHPs within communities. In particular, organizations seeking to be contractually affiliated with the local mental health authority for the purpose of enrolling with the CMHP must apply to the CMHP for a certificate of approval. OAR 309-12-0160(2). There is no evidence in the stipulated record that either Hoyt or her employer, Building Bridges, contracted with the relevant CMHP to provide services for autistic children or received the requisite certificate of approval. The stipulated record also fails to reveal that Hoyt or Building Bridges applied for a certificate of approval as a qualifying "non-inpatient provider” under OAR 309-12-0160(3) for services provided in accordance with ORS 743A.168 or sought a variance from the administrative requirements pursuant to OAR 309-39-0580.