Wit v. United Behavioral Health

Full title: DAVID WIT, et al., Plaintiffs, v. UNITED BEHAVIORAL HEALTH, Defendant…

Court: UNITED STATES DISTRICT COURT NORTHERN DISTRICT OF CALIFORNIA

Date published: Feb 28, 2019

Case No. 14-cv-02346-JCS (N.D. Cal. Feb. 28, 2019)

Opinion

Case No. 14-cv-02346-JCS

02-28-2019

DAVID WIT, et al., Plaintiffs, v. UNITED BEHAVIORAL HEALTH, Defendant. GARY ALEXANDER, et al., Plaintiffs, v. UNITED BEHAVIORAL HEALTH, Defendant.

JOSEPH C. SPERO Chief Magistrate Judge

Related Case No. 14-cv-05337 JCS

FINDINGS OF FACT AND CONCLUSIONS OF LAW

REDACTED

I. INTRODUCTION

Defendant United Behavioral Health ("UBH"), which also operates as OptumHealth Behavioral Solutions, administers mental health and substance use disorder benefits for commercial welfare benefit plans. In that capacity, it has developed Level of Care Guidelines and Coverage Determination Guidelines (collectively, "Guidelines") that it uses for making coverage determinations. Plaintiffs in these related class actions assert claims under the Employee Retirement Income Security Act of 1974 ("ERISA"), 29 U.S.C. § 1001 et seq., alleging that they were improperly denied benefits for treatment of mental health and substance use disorders because UBH's Guidelines do not comply with the terms of their insurance plans and/or state law. The Court conducted a 10-day bench trial and now makes the following findings of fact and conclusions of law pursuant to Federal Rule of Civil Procedure 52(a).¹ The parties have consented to the jurisdiction of the undersigned magistrate judge pursuant to 28 U.S.C. § 636(c).

1 Any findings of fact that constitute conclusions of law shall be deemed to have been found by the Court as a matter of law. Likewise, any conclusions of law that constitute findings of fact shall be deemed to have been found by the Court as a matter of fact.

II. FINDINGS OF FACT

A. The Parties

1. Defendant UBH administers insurance benefits for behavioral health services, including diagnosis and treatment of mental health conditions and substance use disorders. Trial Ex. 880-0004 (Stipulations of Fact) ¶¶ 1, 2. In this role, UBH administers requests for coverage on behalf of members of health benefit plans governed by ERISA, including the health benefit plans of the class members in these actions (collectively, the "Plans"). Id. ¶ 3.

2. Named Plaintiffs in this case are as follows: David and Natasha Wit, Brian Muir, Brandt Pfeifer, Lori Flanzraich, Cecilia Holdnak, Linda Tillitt, Gary Alexander, Corinna Klein, David Haffner and Michael Driscoll. Each of the named Plaintiffs was at all relevant times a beneficiary of an ERISA-governed health benefit plan for which UBH acted as a claims administrator. Id. ¶ 4.²

2 The Court notes that the Stipulations of Fact omit Plaintiffs Linda Tillitt and Michael Driscoll from the list of named Plaintiffs in the preamble. This appears to be an inadvertent omission. UBH concedes that Tillitt and Driscoll were beneficiaries of ERISA-governed health benefit plans and that they received clinical non-coverage determinations from UBH. See UBH's Post-Trial Proposed Findings of Fact and Conclusions of Law ¶¶ 9-11.

3. David and Natasha Wit: At all times relevant to UBH's liability, David Wit was a participant in the "Insperity Group Health Plan" (the "Wit Plan"), a healthcare policy issued by UnitedHealthcare Insurance Company. Trial Ex. 245 (Wit Plan). Mr. Wit's daughter, Natasha Wit, was a beneficiary of the Wit Plan. Trial Ex. 246-002. The Wits sought coverage under the Wit Plan for Natasha's residential treatment at Monte Nido Vista.³ Trial Ex. 246-0002. UBH issued a Clinical Non-Coverage Determination on May 3, 2013 denying coverage for Natasha's residential treatment from April 30, 2013 forward. Trial Ex. 246-0002 to -0007. On the same day, UBH issued a written notification of the adverse benefit determination, citing its 2013 Level of Care Guidelines as the basis for the denial, stating: "It is my determination that the member's treatment does not meet the medical necessity criteria for residential mental health treatment per UBH Level of Care Guidelines for Residential Mental Health treatment . . . ." Trial Ex. 246-0002. The Wits appealed UBH's adverse benefit determination. Trial Ex. 246-0008. UBH denied the appeal on May 3, 2013, again citing its Level of Care Guidelines for Residential Mental Health Treatment, and informed the Wits, "[t]his is the Final Adverse Determination of your internal appeal. All internal appeals through UBH have been exhausted." Trial Ex. 246-0009.

3 UBH's letter denying coverage refers to this facility as "Montenido Lake Vista Treatment Center," see Trial Ex. 246-0002, whereas Plaintiffs refer to the facility as Monte Nido Vista. See Wit v. United Behavioral Health, Case No. 14-cv-2346 JCS (hereinafter, "Wit"), First Amended Class Action Complaint ("FAC") ¶ 43. The Court assumes that both refer to the same facility and uses the name provided by Plaintiffs.

4. Brian Muir: At all times relevant to UBH's liability, Brian Muir was a beneficiary of a group health plan sponsored by Deloitte LLP (the "Muir Plan"). Trial Ex. 239 (Muir Plan). The plan administrator is Deloitte LLP. Trial Ex. 239-0088. On March 1, 2013, Muir sought coverage under the Muir Plan for residential treatment at Sierra Tucson. Trial Ex. 240-0002. On March 7, 2013, UBH issued a Clinical Non-Coverage Determination denying all coverage for Muir's residential rehabilitation treatment from March 1, 2013 forward, citing as the basis for the denial the UBH Coverage Determination Guideline for Residential Rehabilitation for Substance Use Disorders in effect as of March 2013. Trial Ex. 240-0002 to -0003. Muir's provider filed an urgent appeal of UBH's adverse benefit determination, which UBH denied on March 7, 2013, again citing UBH's Coverage Determination Guideline for Residential Rehabilitation for Substance Use Disorders. Trial Ex. 240-0004 to -0006. UBH informed Muir, "[t]his is the Final Adverse Determination of your internal appeal. All internal appeals through United Behavioral Health (UBH) have been exhausted." Trial Ex. 240-0004 to -0006.

5. Brandt Pfeifer: At all times relevant to UBH's liability, Brandt Pfeifer was a participant in the "Continental Offices Limited" plan (the "Pfeifer Plan"), a group healthcare policy issued and underwritten by United Healthcare of Illinois. Trial Ex. 241 (Pfeifer Plan). Pfeifer's late wife, Lauralee Pfeifer, was a beneficiary of the Pfeifer Plan. Trial Ex. 242-0002. On October 26, 2013, the Pfeifers sought coverage under the Pfeifer Plan for residential rehabilitation treatment of Lauralee at Passages-Malibu ("Passages"), a residential treatment facility in Malibu, California. Trial Ex. 242-0002; Wit FAC ¶ 121. On November 1, 2013, UBH issued a Clinical Non-Coverage Determination denying all coverage for Lauralee's residential rehabilitation treatment, from the date of her admission forward, citing the 2013 UBH Level of Care Guidelines. Trial Ex. 242-0002 to -0003. The Pfeifers filed an urgent appeal of UBH's adverse benefit determination, which UBH denied on November 1, 2013, again citing UBH's 2013 Level of Care Guidelines. Trial Ex. 242-0004 to -0006. UBH informed the Pfeifers, "[t]his is the Final Adverse Determination of your internal appeal. All internal appeals through UBH have been exhausted." Trial Ex. 242-0004 to -0006.

6. Lori Flanzraich: At all times relevant to UBH's liability, Lori Flanzraich and her daughter, Casey, were beneficiaries of the "Flanzraich Group Health Plan" (the "Flanzraich Plan"), a healthcare policy underwritten by UBH's affiliate, Oxford Health Insurance, Inc. Trial Ex. 231 (Flanzraich Plan). On December 7, 2012, Lori Flanzraich requested coverage under the Flanzraich Plan for Casey's residential treatment at Solacium New Haven Treatment Center ("New Haven"). Trial Ex. 232-0002; Wit FAC ¶¶ 164-165. UBH issued a Clinical Non-Coverage Determination on February 18, 2013, denying all coverage for Casey's residential treatment from December 7, 2012 forward. Trial Ex. 232-0002 to -0011. On February 18, 2013, UBH sent the Flanzraichs a written notification of its adverse benefit determination concerning Casey's residential treatment. Trial Ex. 232-0002. The notification cited UBH's Level of Care Guidelines as the basis for the denial, stating that the determination was "[b]ased on the clinical information and UBH Level of Care Guidelines for Mental Health Residential Care . . . ." Trial Ex. 232-0002. The Flanzraichs appealed UBH's adverse benefit determination. Trial Ex. 232-0024. UBH denied the appeal on August 23, 2013, again citing UBH's Level of Care Guidelines for Mental Health Residential Care. Id. UBH notified the Flanzraichs that "[t]his is the Final Determination of your internal Appeal. All internal appeals through UBH have been exhausted." Trial Ex. 232-0025.

7. Cecilia Holdnak: At all times relevant to UBH's liability, Cecilia Holdnak was a participant in a group healthcare plan sponsored by American Express Company (the "Holdnak Plan"). Trial Ex. 235 (Holdnak Plan). Cecilia Holdnak's daughter "Emily"⁴ was a beneficiary of the Holdnak Plan. Trial Ex. 236-0002. On December 13, 2013, Emily sought coverage under the Holdnak Plan for her residential treatment at New Haven. Trial Ex. 236-0018. Although UBH initially authorized coverage for Emily's treatment, on January 4, 2014 it denied further coverage from that date forward on the ground that the treatment was "custodial" and therefore excluded from the Holdnak Plan. Id. UBH reversed that denial following an urgent appeal in which Emily's treating psychiatrist opined that treating Emily in a less restrictive setting would not be safe. Trial Ex. 236-0009 to -0015. On January 31, 2014, UBH issued another Clinical Non-Coverage Determination, denying coverage for Emily's residential treatment from that date forward, citing UBH's Coverage Determination Guideline for Residential Treatment Center, Major Depressive Disorder and Dysthymic Disorder, in effect as of that date. Trial Ex. 236-0018 to -0024. Emily's provider urgently appealed this second adverse benefit determination. Trial Ex. 236-0025. On February 1, 2014, UBH denied the appeal, citing UBH's Coverage Determination Guideline for the Residential Treatment of Major Depression. Trial Ex. 236-0025 to -0031. On March 25, 2014, the Holdnaks filed a second-level appeal. Trial Ex. 236-0032 to -0046. UBH denied that appeal on April 7, 2014. UBH's denial again cited UBH's Coverage Determination Guideline for Residential Treatment Center, Major Depressive Disorder and Dysthymic Disorder. Trial Ex. 236-0047 to -0053.

4 Pursuant to the parties' stipulation, the Court uses the pseudonym "Emily" for Cecilia Holdnak's daughter, who was a minor at the time Cecilia Holdnak became a plaintiff in the case.

8. Linda Tillitt: At all times relevant to UBH's liability, Linda Tillitt was a participant in the "Lockton, Inc. Welfare Benefit Plan" (the "Tillitt Plan"), a group healthcare plan sponsored by Lockton, Inc. Trial Ex. 243 (Tillitt Plan). Linda Tillitt's late son, Maxwell Tillitt ("Max"), was a beneficiary of the Tillitt Plan. Trial Ex. 244-0002. On June 18, 2015, the Tillitts requested coverage under the Tillitt Plan for Max's residential treatment at Beauterre Recovery Institute ("Beauterre"), a residential treatment facility in Owatonna, Minnesota. Trial Ex. 244-0002; Intervenor Complaint ¶ 48. On July 9, 2015, UBH issued a Clinical Non-Coverage Determination denying any further coverage for Max's residential treatment from that date forward, citing UBH's Residential Coverage Determination Guidelines for Substance-Related Disorder. Trial Ex. 244-0002. Max's provider submitted an urgent appeal of UBH's denial of coverage, which UBH denied on July 13, 2015, this time citing UBH's 2015 Level of Care Guidelines for Substance Use Disorder Residential Treatment Rehabilitation. Trial Ex. 244-0009. UBH notified Max that "[t]his is the Final Adverse Determination of your internal appeal. All internal appeals through United Behavioral Health (UBH) have been exhausted." Trial Ex. 244-0009 to -0016.

9. Gary Alexander: At all times relevant to UBH's liability, Gary Alexander was a participant in the "Granite Construction Health Plan" (the "Alexander Plan"), a group healthcare policy issued and underwritten by United Healthcare Insurance Company. Trial Ex. 225 (Alexander Plan). The plan administrator is Granite Construction. Trial Ex. 225-0157. Jordan Alexander, Gary Alexander's son, was a beneficiary of the Alexander Plan. Trial Ex. 226-0002. In August of 2013, the Alexanders sought coverage under the Alexander Plan for Jordan's residential treatment at Lifeline for Youth, in North Salt Lake, Utah. Trial Ex. 226-0002; Alexander v. United Behavioral Health, Case No. 14-cv-5337 JCS (hereinafter, "Alexander"), Complaint ¶ 63. On September 16, 2013, UBH sent the Alexanders a written notification of its adverse benefit determination, citing as the basis for the denial UBH's Coverage Determination Guideline for Substance Use Disorder IOP Treatment. Trial Ex. 226-0008 to -0010. Jordan's provider appealed UBH's denial on September 16, 2013 and UBH denied the appeal on the same day, again citing UBH's Coverage Determination Guideline for Substance Use Disorder IOP Treatment. Trial Ex. 226-0011 to -0013. UBH notified the Alexanders that "[t]his is the Final Adverse Determination of your internal appeal. All internal appeals through UBH have been exhausted." Trial Ex. 226-0011 to -0013.

10. Corinna Klein: At all times relevant to UBH's liability, Corinna Klein was a beneficiary of the "Legal Aid Society Group Health Plan" (the "Klein Plan"), a group healthcare policy issued and underwritten by Oxford Health Plans, Inc. Trial Ex. 237 (Klein Plan). Klein sought coverage under the Klein Plan for outpatient mental health treatment, which her psychiatrist prescribed at a frequency of two to three times per week. Trial Ex. 238-0008 to -0016. On September 22, 2014, UBH issued an adverse benefit determination, prospectively limiting Klein's coverage for outpatient mental health treatment to one session per week. Trial Ex. 238-0008 to -0016. On September 22, 2014, UBH sent Klein a written notification of its adverse benefit determination, citing as a basis for the denial UBH's 2014 Level of Care Guidelines, stating: "Based on our UBH Level of Care Guideline for Mental Health Outpatient Level of Care, it is my determination that no further authorization can be provided for multiple weekly therapy visits . . . ." Trial Ex. 238-0008 to -0016. On October 21, 2014, Klein's psychiatrist faxed an urgent appeal of UBH's adverse benefit determination. Trial Ex. 238-0008 to -0016. UBH never responded to the appeal.

11. David Haffner: At all times relevant to UBH's liability, David Haffner was a participant in the "Science Systems and Applications, Inc. Health and Medical Plan" (the "Haffner Plan"), a group healthcare policy issued and underwritten by United Healthcare Insurance Company. Trial Ex. 233 (Haffner Plan). In 2011, Haffner requested coverage under the Haffner Plan for twice-weekly, 45-minute outpatient psychotherapy sessions (with medical evaluation and management) with Michael S. Diamond, M.D. in Chevy Chase, Maryland. Trial Ex. 234-0002; Alexander Complaint ¶ 104. On December 5, 2011, UBH issued a Clinical Non-Coverage Determination prospectively limiting Haffner's coverage for outpatient mental health treatment to one session per month. Trial Ex. 234-0002 to -0008. On December 5, 2011, UBH sent Haffner a written notification of its adverse benefit determination. Trial Ex. 234-0002 to -0008. The written notification stated the rationale for UBH's decision to deny benefits but did not cite the relevant Guideline. Trial Ex. 234-0002 to -0008. Haffner appealed UBH's denial on April 16, 2012. Trial Ex. 234-0009 to -0011. UBH denied the appeal on May 17, 2012, citing UBH's Coverage Determination Guidelines for Personality Disorders, Outpatient Treatment of Obsessive Compulsive Disorder, and Outpatient Treatment of Bipolar Disorder. Trial Ex. 234-0012 to -0014. The letter further notified Haffner that "[t]his is the Final Determination of your internal appeal. All internal appeals through UBH have been exhausted." Trial Ex. 234-0012 to -0014.

12. Michael Driscoll: At all times relevant to UBH's liability, Michael Driscoll was a participant in the "George Washington University Plan" (the "Driscoll Plan"), a group healthcare policy. Trial Ex. 227 (Driscoll Plan). Driscoll's daughter, "Sara,"⁵ was a beneficiary of the Driscoll Plan. Trial Ex. 229-0002. On September 10, 2013, the Driscolls sought coverage under the Driscoll Plan for IOP treatment of Sara's substance use disorder at The Canyon at Santa Monica ("The Canyon"). Trial Ex. 229-0007. UBH issued a Clinical Non-Coverage Determination denying coverage for Sara's IOP treatment, in its entirety, and her providers appealed the decision. Trial Ex. 229-0007 to -0008. On March 26, 2014, UBH sent the Driscolls a written notification of its decision to uphold the adverse benefit determination, citing the UBH Coverage Determination Guideline for Treatment of Substance Use Disorders in effect as of March 2014. Trial Ex. 229-0007 to -0008. The Driscolls appealed UBH's adverse benefit determination a second time and UBH denied the appeal on June 2, 2014, again citing UBH's Coverage Determination Guideline for Treatment of Substance Use Disorders. Trial Ex. 229-0009 to -0012. UBH informed the Driscolls, "[t]his is the Final Adverse Determination of your internal appeal. All internal appeals through UBH have been exhausted." Trial Ex. 229-0009 to -0012.

5 Pursuant to the parties' stipulation, the Court uses the pseudonym "Sara" for Michael Driscoll's daughter, who was a minor at the time Michael Driscoll became a plaintiff in the case.

B. The Classes

13. The Court certified the following classes for trial:

• Wit Guideline Class: Any member of a health benefit plan governed by ERISA whose request for coverage of residential treatment services for a mental illness or substance use disorder was denied by UBH, in whole or in part, between May 22, 2011 and Junel, 2017, based upon UBH's Level of Care Guidelines or UBH's Coverage Determination Guidelines. The Wit Guideline Class excludes members of the Wit State Mandate Class, as defined below.

• The Wit State Mandate Class: Any member of a fully-insured health benefit plan governed by both ERISA and the state law of Connecticut, Illinois, Rhode Island, or Texas, whose request for coverage of residential treatment services for a substance use disorder was denied by UBH, in whole or in part, within the Class period, based upon UBH's Level of Care Guidelines or UBH's Coverage Determination Guidelines, and not

upon the level-of-care criteria mandated by the applicable state law. With respect to plans governed by Texas law, the Wit State Mandate Class includes only denials of requests for coverage of substance use disorder services that were sought or received in Texas. The Class period for the Wit State Mandate Class includes denials governed by Texas law that occurred between May 22, 2011 and June 1, 2017, denials governed by Illinois law that occurred between August 18, 2011 and June 1, 2017, denials governed by Connecticut law that occurred between October 1, 2013 and June 1, 2017, and denials governed by Rhode Island law that occurred between July 10, 2015 and June 1, 2017.

• The Alexander Guideline Class: Any member of a health benefit plan governed by ERISA whose request for coverage of outpatient or intensive outpatient services for a mental illness or substance use disorder was denied by UBH, in whole or in part, between December 4, 2011 and June 1, 2017, based upon UBH's Level of Care Guidelines or UBH's Coverage Determination Guidelines. The Alexander Guideline Class excludes any member of a fully insured plan governed by both ERISA and the state law of Connecticut, Illinois, Rhode Island or Texas, whose request for coverage of intensive outpatient treatment or outpatient treatment was related to a substance use disorder, except that the Alexander Guideline Class includes members of plans governed by the state law of Texas who were denied coverage of substance use disorder services sought or provided outside of Texas.

14. During discovery, the parties agreed that rather than producing the Plan term documents and administrative records for all class members, UBH would produce those documents only for the named Plaintiffs and a small, random sample of Class Members (hereinafter, the "Claim Sample"). Trial Ex. 897-0001 (Joint Stipulation Concerning Sampling Methodology). UBH stipulated at trial that for the purposes of this case, the Claim Sample is a "representative sample of the entire class." Trial Tr. 1890:1-15.

15. At trial, the Court admitted the following evidence concerning the Claim Sample members:

• The applicable document reflecting the terms of each Claim Sample member's plan (i.e.,

the Certificate of Coverage or Summary Plan Description for each plan), see Trial Tr. 678:10-679:7 (listing plan term exhibits admitted into evidence); and

• The following charts summarizing what the parties consider to be the relevant provisions of each Claim Sample member's plan: Trial Ex. 892 (Plaintiffs' Summary Exhibit A: Plan Terms); Trial Ex. 893 (Plaintiffs' Summary Exhibit B: Plan Groupings); Trial Ex. 1653 (UBH's Summary Exhibit: Plans); Trial Ex. 1654 (UBH's Summary Exhibit: Custodial Care Definition).

C. The Claims

16. Plaintiffs assert two claims: 1) breach of fiduciary duty (the "Breach of Fiduciary Duty Claim") and 2) arbitrary and capricious denial of benefits (the "Denial of Benefits Claim"). Plaintiffs assert the Breach of Fiduciary Duty Claim under 29 U.S.C. § 1132(a)(1)(B) (Count I in all of the operative complaints) and, to the extent the injunctive relief Plaintiffs seek is unavailable under that section, they assert the claim under 29 U.S.C. § 1132(a)(3)(A) (Count III in all of the operative complaints). Similarly, Plaintiffs assert the Denial of Benefits Claim under 29 U.S.C. § 1132(a)(1)(B) (Count II in all of the operative complaints) and under 29 U.S.C. § 1132(a)(3)(B) (Count IV in all of the operative complaints).

17. The Breach of Fiduciary Duty Claim is based on the theory that UBH is an ERISA fiduciary under 29 U.S.C. § 1104(a) and owed fiduciary duties to the class members, including the duties to administer the class members' health benefit plans "solely in the interest of the participants and beneficiaries," 29 U.S.C. § 1104(a)(1), "with . . . care, skill, prudence, and diligence," 29 U.S.C. § 1104(a)(1)(B), and "in accordance with the documents and instruments governing the plans," 29 U.S.C. § 1104(a)(1)(D). According to Plaintiffs, UBH breached these duties by: 1) developing guidelines for making coverage determinations that are far more restrictive than those that are generally accepted even though Plaintiffs' health insurance plans provide for coverage of treatment that is consistent with generally accepted standards of care; and 2) prioritizing cost savings over members' interests.

18. The Denial of Benefits Claim is based on the theory that UBH improperly adjudicated and denied Plaintiffs' requests for coverage by using its overly restrictive Guidelines to make coverage determinations. According to Plaintiffs, UBH's reliance on the Guidelines was arbitrary and capricious because: 1) Plaintiffs' health insurance plans provided for coverage consistent with generally accepted standards of care; and 2) as to the Wit State Mandate Class, the Class members' health insurance plans were subject to state laws that explicitly mandate the use of clinical criteria issued by the American Society of Addiction Medicine ("ASAM") or the Texas Department of Insurance ("TDI").

19. Plaintiffs stipulated at the class certification stage of the case that they do not ask the Court to make determinations as to whether individual class members were actually entitled to benefits (which might have required the Court to consider a multitude of individualized circumstances relating to the medical necessity for coverage and the specific terms of the member's plan). Rather, they assert only facial challenges to the Guidelines.

D. Credibility Findings ⁶

6 In addition to these general credibility findings as to key witnesses, the Court makes specific credibility findings as to particular testimony offered by these and other witnesses throughout its Findings of Fact.

20. Plaintiffs retained two experts, Dr. Marc Fishman and Dr. Eric Plakun, who offered testimony at trial addressing, inter alia, generally accepted standards of care related to mental health and substance use disorder treatment and whether the UBH Guidelines meet those standards.

21. Dr. Fishman is a psychiatrist who specializes in addiction psychiatry and addiction medicine, with subspecialties in the treatment of adolescents and young adults, and the treatment of opioid use disorders and use of medication. Trial Tr. 62:1-23 (Fishman). After graduating from Columbia medical school he completed a residency in general psychiatry at Johns Hopkins Hospital in 1992. Trial Tr. 62:5-8 (Fishman); Trial Ex. 670-0002 (CV). He worked briefly as a full-time professor of psychiatry at Johns Hopkins, then in 1993 moved to a part-time faculty position there, which he still holds, when he became medical director of Mountain Manor Treatment Center. Trial Ex. 670-0002 (CV). Since 1998 he has also served as the medical director of the Maryland Treatment Center, a network of community treatment providers for addictions and co-occurring conditions. Trial Tr. 62:10-13 (Fishman); Trial Ex. 670-0002. A particular focus of Dr. Fishman's practice and research has been on levels of care, level of care guidelines, and treatment matching strategies to ensure patients receive treatment in the appropriate and most effective level of care. Tr. 62:24-63:2 (Fishman). In 1997, he was appointed to the steering committee for the ASAM Criteria and since that time has served as a co-author of the ASAM Criteria. Trial Tr. 67:1-9 (Fishman); Trial Ex. 670-003 to -004 (CV). In addition to being a member of the steering committee, Dr. Fishman has also headed ASAM's Work Group on Adolescent Patient Placement Criteria since 1997 and headed ASAM's Workgroup on Patient Placement Criteria Supplement on Pharmacotherapies for Alcohol Dependence between 2006 and 2011.

22. Dr. Fishman offered testimony based on his extensive experience as an addiction medicine specialist focused on treatment of substance use disorders and co-occurring conditions with regard to adults, children, and adolescents. In addition, Dr. Fishman offered testimony on mental health treatment of adults, children, and adolescents based on his many years of experience in general psychiatry. Dr. Fishman's testimony was credible in all respects. The Court found that Dr. Fishman's decades-long involvement in and intimate familiarity with the development of the ASAM Criteria made him a particularly persuasive witness with respect to the ways in which UBH's Guidelines are more restrictive than generally accepted standards of care.

23. Dr. Plakun is a board-certified psychiatrist. Trial Tr. 468:11 (Plakun). He graduated from Columbia medical school and in 1978, after completing a psychiatry residency at Dartmouth, entered a four-year fellowship in psychoanalytic studies at the Austen Riggs Center. Trial Tr. 469:1-10. The Austen Riggs Center is a residential treatment facility that also provides a "hospital-based continuum of care," and is consistently recognized as one of the top ten psychiatric hospitals in the country. Trial Tr. 468:15; 470:4-8 (Plakun). Dr. Plakun served for thirty-five years as the Director of Admissions for the Austen Riggs Center. Trial Tr. 471:21-24 (Plakun). In that capacity, he evaluated thousands of patients to determine whether residential treatment was appropriate or if instead a recommendation for a higher or lower level of care should be made. Trial Tr. 473:3-474:8 (Plakun). In addition, since the early 1990s, Dr. Plakun has been a "treatment team leader" at Austen Riggs and in that role has been responsible for making level-of-care decisions about which of the different programs within the Riggs continuum of care a patient should be placed in. Trial Tr. 471:13-17. Dr. Plakun is currently the Associate Medical Director at Austen Riggs. Trial Tr. 468:12-13. Dr. Plakun also served for twenty-one years as a member of the clinical faculty at Harvard Medical School, is a Distinguished Life Fellow of the American Psychiatric Association, and has edited two books, including one on residential treatment of "treatment resistant" patients that addresses his research on predictors of outcomes as to such individuals. Trial Tr. 476:3-7 (Plakun).

24. Dr. Plakun offered testimony focused on treatment of mental health conditions and co-occurring disorders in adults. The Court found Dr. Plakun's testimony to be generally credible.

25. UBH's experts, on the other hand, had serious credibility problems. The Court found that with respect to a significant portion of their testimony each of them was evasive - and even deceptive - in their answers when confronted with contrary evidence. Therefore, the Court discounts the testimony of UBH's expert witnesses as described further below.

26. UBH offered the testimony of one retained expert, Dr. Thomas Simpatico. Dr. Simpatico went to medical school at Rush Medical College, in Chicago, Illinois, and completed a psychiatry residency at the University of Chicago. Trial Tr. 1142:2-6 (Simpatico). He has been practicing psychiatry since 1985 and specializes in systems of care and standards of care, among other things. Trial Tr. 1142:16-19 (Simpatico). Prior to moving to Vermont in 2004, Dr. Simpatico worked in various administrative roles, including medical director, related to the provision of mental health care at community mental health centers and state hospitals. Trial Tr. 1143:14-1144:8 (Simpatico). He has been a professor of psychiatry at the University of Vermont since 2004. Trial Tr. 1142:20-1143:4. He also worked as the medical director of the Vermont State Hospital from approximately 2004 to 2009 and for the past seven years he has served as medical director of Pathways Vermont. Trial Tr. 1143:13-16; 1144:9-13 (Simpatico). In addition, Dr. Simpatico worked for approximately four and a half years as the chief medical officer of the Vermont Medicaid Authority. Trial Tr. 1144:14-25 (Simpatico).

27. Dr. Simpatico offered testimony about generally accepted standards of care with respect to mental health and substance use disorder treatment. His testimony on that subject was generally credible. He also offered testimony that UBH's Guidelines are consistent with generally accepted standards of care. That testimony was not credible. At numerous points in his testimony, Dr. Simpatico overlooked language in the Guidelines that was inconsistent with generally accepted standards of care. For example, when asked how he would interpret a Guideline requiring "clear and compelling evidence that continued treatment at this level of care is required to prevent acute deterioration or exacerbation that would then require a higher level of care," Dr. Simpatico testified that "clear and compelling" meant "reasonably likely," acknowledging that "clear and compelling" is not a phrase that is typically used in medical or behavioral health guidelines. Trial Tr. 1237:13-1238:6 (Simpatico). When pressed by the Court, Dr. Simpatico insisted that "clear and compelling" and "reasonably likely" were "equivalent," Trial Tr. 1239:16-20 (Simpatico), before finally conceding that the literal meaning of these words set a more stringent standard than his interpretation and that the words "clear and compelling" set an "impossible metric." Trial Tr. 1238:9-1240:24, 1242:8-9 (Simpatico). At that point, Dr. Simpatico explained that "any practitioner worth his salt" would not rely on the Guidelines themselves but instead, would go straight to the underlying documents that set forth generally accepted standards of care, such as the APA Clinical Practice Guidelines, the ASAM Criteria or the LOCUS (discussed below). Trial Tr. 1241:13-1242:10. He reasoned that such an approach was appropriate because the Guidelines instruct that practitioners are to adhere to generally accepted standards of care, asking rhetorically, how else would a doctor making a medical necessity determination "reconcile the discrepancy" between the Guidelines and the source documents for the Guidelines. Trial Tr. 1242:1-3, 1242:21-24 (Simpatico).

28. Dr. Simpatico's opinions about the Guidelines were premised on the assumption that practitioners making medical necessity determinations for UBH are authorized to ignore the plain language of the Guidelines when it is inconsistent with generally accepted standards of care. The evidence presented at trial does not support that assumption. While the Guidelines allow for some exercise of clinical judgment, they are the criteria against which UBH Peer Reviewers make clinical coverage determinations, and they are mandatory. Trial Tr. 732:20-733:3 (Triana). Because there is no evidence in the record that the words in the Guidelines can be ignored by the Peer Reviewers when they are in conflict with generally accepted standards of care - or that they are, in fact, used that way - the Court finds that Dr. Simpatico's testimony on the question of whether the Guidelines are consistent with generally accepted standards of care was not credible.

29. UBH designated Dr. Lorenzo Triana as its corporate representative under Rule 30(b)(6) of the Federal Rules of Civil Procedure and as a non-retained in-house expert witness. Trial Tr. 697:18-20 (Triana). Dr. Triana has been UBH's Senior Vice president of Behavioral Medical Operations since 2010, and all senior medical directors and clinical operations report directly to him. Trial Tr. 698:20-699:23 (Triana). The senior medical directors and clinical operations are responsible for making and supervising clinical coverage decisions. Trial Tr. 699: 7-12 (Triana). Dr. Triana chaired UBH's Behavioral Policy and Analytics Committee ("BPAC"), the committee responsible for approving the Guidelines, between 2011 and 2016. Trial Tr. 703:3-16 (Triana); Trial Ex. 482-0002 (BPAC minutes showing members). When BPAC was replaced by the Utilization Management Committee ("UMC") in 2016, Triana served as chair of the UMC. Trial Tr. 698:7-11 (Triana); Trial Ex. 552-002 (August 9, 2016 UMC minutes listing Dr. Triana as chair). Dr. Triana was also a member of the Level of Care Guidelines Workgroup, which also included Mr. Niewenhous and Drs. Triana, Martorana, Bonfield and Brock. Trial Tr. 1697:2-5 (Triana).

30. While some of Dr. Triana's testimony was credible, his testimony that UBH does not consider benefit expense (sometimes referred to as "benex" or "Ben Ex") when it develops the Guidelines was not credible in light of evidence and testimony introduced at trial, discussed below, showing that financial considerations have played a significant role in the development of the Guidelines throughout the relevant class periods.

31. Dr. Andrew Martorana is a board-certified psychiatrist. Trial Tr. 923:12 (Martorana). In 1985, after graduating from the University of Illinois Medical School, he completed a four-year combined internship and psychiatric residency at the University of Illinois hospitals. Trial Tr. 923:3-9 (Martorana). He then engaged in private practice for 17 years, treating patients for both mental health and substance use disorders. Trial Tr. 924:14-16 (Martorana). He has been employed by UBH since 2002 and currently holds the position of Senior Behavioral Medical Director. Trial Tr. 922:21-24 (Martorana). In that position, he reports directly to Dr. Triana. Trial Tr. 699:7-12 (Triana). His responsibilities include supervision and training of UBH Care Advocacy clinicians and "quality improvement." Trial Tr. 925:5-24 (Martorana). He was a member of the BPAC from 2013 to 2016 and has been a member of the UMC since its creation, in 2016. Trial Tr. 927:19-20, 928:21-22 (Martorana). He has also been a member of the Level of Care Guidelines Workgroup. Trial Tr. 1697:2-5 (Triana).

32. Although Dr. Martorana's testimony was credible on some issues, his testimony about the meaning of the Guidelines was not always credible because in several instances he ignored the plain meaning of the words used in the Guidelines. See, e.g., Trial Tr. 974:23-976:13 (Martorana testimony that the words "safely managed" in the Guidelines mean the same thing as "effectively treated"); Trial Tr. 1054:12-17 (Martorana testimony that "Why Now" factors referenced in the Guidelines call for an assessment of the "whole person" or the patient's entire multi-dimensional history). Further, Dr. Martorana's testimony that clinicians were trained to apply the Guidelines in a manner that was inconsistent with their plain meaning was not supported by other evidence introduced at trial. See, e.g., Trial Tr. 978:11-12 (Martorana).

33. Mr. Gerard Niewenhous was trained as a social worker and has been employed by UBH since 2003. Trial Tr. 1732:7-10 (Triana); Trial Tr. 297:4-5 (Niewenhous). He was responsible for maintaining the Level of Care Guidelines from 2003 to the middle of 2016 and for drafting the Coverage Determination Guidelines from 2010 to the middle of 2015. Trial Tr. 297:4-9, 297:12-15 (Niewenhous). He offered extensive testimony addressing the process UBH used to draft and update the Guidelines, factors that were considered in creating them, and the meaning of the words used in the Guidelines. While Mr. Niewenhous's testimony was credible on some issues, his testimony that the Guidelines were developed solely to reflect generally accepted standards of care was not credible. As discussed further below, internal UBH communications involving Mr. Niewenhous make it crystal clear that the primary focus of the Guideline development process, in which Mr. Niewenhous played a critical role, was the implementation of a "utilization management" model that keeps benefit expenses down by placing a heavy emphasis on crisis stabilization and an insufficient emphasis on the effective treatment of co-occurring and chronic conditions.

34. Dr. Theodore Allchin is a board-certified child and adolescent psychiatrist. Trial Tr. 1354:20-22 (Allchin). He received his medical degree from Ohio State University in 1982 and subsequently completed an internship, general psychiatry residency, and a child psychiatry fellowship at the University of Chicago. Trial Tr. 1353:18-23 (Allchin). From 1987 to 2009, Dr. Allchin had a private practice that focused mainly on children and adolescents. Trial Tr. 1355:20-25. He began working part-time at UBH in 1988, splitting his time between private practice and his work at UBH until 2009, when he ended his private practice. Trial Tr. 1357:8-13. At UBH, Dr. Allchin's title is Associate Medical Director. Trial Tr. 1358:12 (Allchin). In that capacity, he performs peer reviews, conducts case consultations with providers and does "rounds" with UBH care advocates, as well as serving on a national credentialing committee. Trial Tr. 1358:15-1359:4 (Allchin). Dr. Allchin's testimony was only partially credible. As discussed further below, his testimony that UBH's Guidelines are consistent with generally accepted standards of care with respect to the treatment of children and adolescents, which he based primarily on the "clinical best practices" in the Guidelines, was not persuasive in light of his admission that the unique factors that relate to the placement of children and adolescents are absent from the coverage criteria in the Guidelines. See Trial Tr. 1377:13-20 (Allchin) (testifying that the clinical best practices section contains "sufficient detail to tease out aspects that are developmentally related" to make up for the lack of coverage criteria tailored to young people).

35. Dr. Danesh Alam is a board-certified psychiatrist. Trial Tr. 1568:25-1569:4 (Alam). He received his medical degree in India and completed his psychiatry training at the University of Illinois at Chicago, where he remains on faculty. Trial Tr. 1568:19-23 (Alam). Dr. Alam has served as president of the Illinois chapter of ASAM and has been on "a couple of committees" of ASAM at the national level. Trial Tr. 1570:6-13. He is employed by UBH and holds the position of Behavioral Medical Director. Trial Tr. 1571:23-25. In that capacity he supervises Care Advocacy staff and makes medical necessity determinations. Trial Tr. 1572:1-6. Dr. Alam testified on the question of whether UBH's Guidelines are consistent with generally accepted standards of care. The Court finds that Dr. Alam's testimony on this subject was not credible. In particular, the Court finds that Dr. Alam's testimony on the subject of whether the Guidelines cover certain lower levels of residential treatment set forth in the ASAM Criteria, and his testimony about Mr. Shulman's conclusions on this subject, was evasive and at times untruthful. His testimony at trial also revealed that he had misrepresented material facts in his expert report when he stated that UBH contracts with "few, if any" providers of lower-intensity residential treatment, namely, at the 3.3 and 3.5 levels under ASAM; at trial, in contrast, he conceded that UBH does contract with such providers. Trial Tr. 1575:10-21 (Alam); 1642:21-1644:10 (Alam). Dr. Alam also repeatedly offered interpretations of the Guidelines that were inconsistent with their plain meaning and dismissed changes to the Guidelines proposed by Mr. Shulman as "just changing words." Trial Tr. 1651:3-8. The Court places no weight on the testimony offered by Dr. Alam that UBH Guidelines are consistent with generally accepted standards of care.

E. Overview of the Guidelines

36. UBH has created a set of clinical policies and guidelines, which include but are not limited to its Level of Care Guidelines ("LOCGs") and its Coverage Determination Guidelines ("CDGs"). Trial Ex. 880-009 (Stipulation of Facts) ¶ 6. In this case, Plaintiffs challenge only UBH's LOCGs and CDGs. See Trial Ex. 880-006 ¶ 19 & Ex. A (chart listing "all Level of Care Guidelines and certain Coverage Determination Guidelines in effect from May 22, 2011 through the present" and which Plaintiffs have stipulated "contains a complete list of all guidelines at issue in these related actions").

37. UBH's own internal auditing system, which measures "Inter-Rater Reliability" ("IRR"), reflects that the Guidelines are applied consistently, which is an important goal at UBH. See Trial Tr. 735:5-739:23 (Triana testimony that for 2011 through 2016 the IRR rate met or exceeded the 90% goal set by UBH, showing that the Guidelines are applied consistently by Peer Reviewers). Where the IRR audit reveals "areas of discrepancy," clinical leaders are expected to take "corrective action." Trial Ex. 259 (2014 Utilization Management Program Description). Based on this evidence, the Court finds that the testimony of some UBH witnesses that Peer Reviewers can deviate from the Guidelines based on their clinical judgment was not credible. See, e.g., Trial Tr. 949:20-22 (Martorana) (testifying that Peer Reviewers can depart from the Guidelines if their clinical judgment "takes them there"); Trial Tr. 1404:25-1405:2 (Allchin) (testifying that he had issued coverage determinations that were inconsistent with the Guidelines and had not required authorization to do so). Rather, the Court finds that UBH employees apply the Guidelines as written, that is, their exercise of clinical judgment is constrained by the criteria for coverage set forth in the Guidelines, which are mandatory.

38. The LOCGs are organized according to the situs of the care at issue (e.g., outpatient vs. residential treatment) whereas most of the CDGs are organized by diagnosis. Trial Tr. 939:4-10 (Martorana). UBH also issues CDGs governing custodial care that apply to any diagnosis. See Trial Exs. 10, 47, 84, 108, 148, 195, 221. The LOCGs are used to make coverage determinations for plans that contain a medical necessity requirement while the CDGs are used to make coverage determinations in cases involving plans that do not contain a medical necessity requirement. Trial Tr. 940:1-3 (Martorana). Whether a claim is denied under an LOCG or a CDG, the denial is considered a clinical denial rather than an administrative denial, that is, a denial that is the result of the exercise of clinical judgment by a practitioner acting on UBH's behalf. Trial Tr. 717:6-19 (Triana); Trial Ex. 259-12 (2014 Optum Utilization Management Program Description) (defining "clinical denial" as "[a] nonauthorization that involves clinical decision" and "administrative denial" as "[a] nonauthorization that is based upon the member's benefit coverage and does not require clinical decision-making").

1. The Level of Care Guidelines

39. UBH's Level of Care Guidelines are used to make coverage determinations under the health benefit plans it administers, and in particular, to establish criteria consistent with generally accepted standards for determining the appropriate level of care. Trial Tr. 1876:22-25 (UBH admission that "the generally accepted standards of care in terms of level of treatment are defined by UBH in its Level of Care Guidelines"); Trial Tr. 298:13-15 (testimony of Mr. Niewenhous that the LOCGs are "supposed to reflect generally accepted standards of care"). The LOCGs are also intended to standardize coverage determinations with respect to the appropriate level of care. Trial Ex. 1-0002 (2011 Level of Care Guidelines ("2011 Guidelines")) (stating that LOCGS are "intended to standardize care advocacy decisions regarding the most appropriate and available level of care needed to support a member's path to recovery"); Trial Ex. 2-0002 (2012 Level of Care Guidelines ("2012 Guidelines")) (same); Trial Ex. 3-0002 (2013 Level of Care Guidelines ("2013 Guidelines")) (same); Trial Ex. 4-0002 (2014 Level of Care Guidelines ("2014 Guidelines")) (LOCGs are "used to standardize coverage determinations"); Trial Ex. 5-0004 (2015 Level of Care Guidelines ("2015 Guidelines")) (same); Trial Ex. 6-0004 (2016 Level of Care Guidelines, Approved January 2016 ("2016 Guidelines (January)")) (same); Trial Ex. 7-0004 (2016 Level of Care Guidelines, Approved January 2016 with Revisions in June 2016 ("2016 Guidelines (June)")) (same); Trial Ex. 8-0002 (2017 Level of Care Guidelines ("2017 Guidelines")) (same). UBH's Guidelines state that they are "objective," "evidence-based" and "derived from generally accepted standards of behavioral practice." Trial Ex. 1-0002 (2011 Guidelines); Trial Ex. 2-0002 (2012 Guidelines); Trial Ex. 3-0002 (2013 Guidelines); Trial Ex. 4-0002 (2014 Guidelines).

40. UBH regularly reevaluates its LOCGs and reissued them at least annually between 2011 and 2017. See Trial Ex. 880-0006 (Stipulations of Fact) ¶ 19; Trial Exs. 1-8 (all versions of the LOCGs in effect throughout the Class Period). Each version of the LOCGs at issue in this case contained an Introduction, a set of "Common Criteria" that applied to coverage at all levels of care, and additional criteria applicable to particular levels of care in the context of both mental health conditions and substance use disorders. See generally Trial Exs. 1-8. The three levels of care that are at issue in this case are: 1) residential treatment, or "RTC;" 2) intensive outpatient treatment, or "IOP;" and 3) outpatient treatment. For each of these levels of care, there is a separate set of criteria for mental health conditions and substance use disorders.

41. The introductory section for every year's LOCGs contains "Guiding Principles"—a statement describing UBH's approach to member care. See Trial Ex. 1-0002 to -0003 (2011 Guidelines); Trial Ex. 2-0002 to -0003 (2012 Guidelines); Trial Ex. 3-0003 to -0004 (2013 Guidelines); Trial Ex. 4-0003 to -0004 (2014 Guidelines); Trial Ex. 5-0004 to -0005 (2015 Guidelines); Trial Ex. 6-0004 to -0005 (2016 Guidelines (January)); Trial Ex. 7-0004 to -0005 (2016 Guidelines (June)); Trial Ex. 8-0002 to -0003 (2017). From 2011 through 2013, the LOCGs set forth four "Guiding Principles": (1) care should promote the member's recovery; (2) care should be accessible; (3) care should be appropriate; and (4) care should be effective. See Trial Exs. 1-0002 to -0003 (2011 Guidelines); Trial Ex. 2-0002 to -0003 (2012 Guidelines); Trial Ex. 3-0003 to -0004 (2013 Guidelines). Since 2014, the LOCGs' "Guiding Principles" have been based on three "pillars": "Care Advocacy," "Service System Solutions," and "Information Management and Technology." See Trial Ex. 4-0003 (2014 Guidelines); Trial Ex. 5-0004 (2015 Guidelines); Trial Ex. 6-0004 (2016 Guidelines (January)); Trial Ex. 7-0004 (2016 Guidelines (June)); Trial Ex. 8-0003 (2017 Guidelines). The Guiding Principles explain that these three pillars "enable the system of care to become more engaging, effective, and affordable." Id. They further explain that "[e]ngagement, evidence-based practices, as well as recovery, resiliency, and wellbeing are integral to each of the pillars." Id.

42. The Common Criteria section contains Level of Care Criteria, that is, general requirements for coverage that apply to all levels of care for making admission, continued coverage and discharge determinations, as well as "best practices" that providers are required to follow in making recommendations about the appropriate level of care. Starting in 2014, these "best practices" were set forth in a separate section of the Common Criteria; before that they were integrated into the Common Criteria. In all versions, the "best practices" are focused on the information treating practitioners should gather in order to diagnose the plan member and create an appropriate treatment plan. See Trial Exs. 1-8; Trial Tr. 980:3-24 (Martorana) (testifying that the "best practices" are the "standard" UBH "hold[s] a competent and qualified clinician to," requiring that the practitioner conduct "a thorough and complete assessment," take "[a]ll this information . . . into consideration in terms of diagnosis and treatment," and use it to develop a "treatment plan that addresses the problems that are at hand, in an appropriate way and [that is] evidence based.").

43. For all versions of the LOCGs that are at issue in this case, every provision of the Common Criteria had to be satisfied in order to obtain coverage at any level of care. This is apparent in the 2014 through 2017 versions of the Common Criteria on their face, as many of the listed requirements are separated by the word "AND," in all capital letters and typically underlined. See Trial Ex. 4-0007 to -0010 (2014 Guidelines); Trial Ex. 5-0008 to -0010 (2015 Guidelines); Trial Ex. 6-0009 to -0011 (2016 Guidelines (January)); Trial Ex. 7-0009 to -0011 (2016 Guidelines (June)); Trial Ex. 8-0006 to -0007 (2017 Guidelines). Although earlier versions of the Common Criteria (in the 2011-2013 Guidelines) did not separate the provisions in the numbered list with the word "AND," they also required that all of the provisions had to be met in order for a service to qualify for coverage, as counsel for UBH conceded at trial. See Trial Tr. 285:16-287:17 (colloquy between counsel and the Court in which UBH counsel conceded that earlier versions "worked the same way" even though they did not contain the word "AND" between the provisions).

44. In addition to satisfying all of the requirements of the Common Criteria, a request for coverage must also meet the requirements contained in the specific LOCG for the applicable level of care.

2. CDGs

45. UBH began developing its CDGs in 2010 as part of its implementation of the Mental Health Parity and Addiction Equity Act (the "Parity Act"), 29 U.S.C. § 1185a. Trial Tr. 1708:22-25 (Triana). UBH updates its CDGs on an annual basis. See Trial Ex. 880-0006 (Stipulations of Fact) ¶ 19 & Ex. A thereto. Like the LOCGs, the CDGs are supposed to reflect generally accepted standards of care. Trial Tr. 298:13-15 (Niewenhous).

46. Most of UBH's CDGs are diagnosis-specific, meaning that each one contains detailed criteria relating to the treatment of a particular mental health condition or substance use disorder. See, e.g., Trial Ex. 214 (2017 CDG for Substance-Related and Addictive Disorders); Trial Ex. 222 (2017 CDG for Bipolar and Related Disorders). UBH's CDGs governing custodial care, however, apply to inpatient or residential treatment for any diagnosis. See Trial Exs. 10, 47, 84, 108, 148, 195, 221 ("Custodial Care CDGs").

47. Except for the Custodial Care CDGs, Plaintiffs challenge the CDGs only to the extent that they incorporate the Level of Care Guidelines. See generally Trial Ex. 880-0009 to - 0020 (stipulated chart listing all challenged Guidelines, by effective date).⁷ Plaintiffs challenge UBH's Custodial Care CDGs not only on the basis that they incorporate the LOCGs but also on independent grounds, as discussed below. See Trial Exs. 10, 47, 84, 108, 148, 195, 221 (UBH's Custodial Care CDGs).

7 At trial, Plaintiffs presented evidence that the diagnosis-specific CDGs listed in Trial Ex. 880 incorporate UBH's LOCGs. That issue will be decided at a later stage of the case, when the Court addresses remedies.

F. The Claims Administration Process

48. When a member or provider submits a request for coverage to UBH, a "Care Advocate" is assigned to (1) determine whether there is an administrative (i.e., non-clinical) basis to deny the request, such as a contractual exclusion for a particular form of treatment or a certain condition, and (2) make an initial determination whether the prescribed treatment, at the proposed level of care, meets criteria in the applicable Guideline. See Trial Ex. 259-0017 (2014 Utilization Management Program Description ("UMPD")); Trial Tr. 721:9-722:6 (Triana). Care Advocates may deny a request on administrative grounds or grant a request on clinical grounds; but if they conclude based on the applicable Guidelines and the information they have collected about the member that the requested service should be denied for clinical reasons they must pass the request on to a Peer Reviewer, who is a physician or doctoral-level psychologist authorized by UBH to make a Clinical Non-Coverage Determination. Trial Tr. 722:7-12 (Triana); Trial Ex. 880-003 to -004 (Stipulations of Fact) Definitions, ¶ 6.

49. A Peer Reviewer's job is to decide, for each request for coverage, whether the prescribed treatment meets the criteria set forth in the Guidelines. Trial Tr. 725:18-726:11 (Triana); Trial Tr. 1102:17-19 (Martorana); see also Trial Exs. 256-0018, 257-0020, 258-0018, 259-0019, 260-0010, 261-0012, 262-0013 (Utilization Management Program Descriptions); Trial Tr. 309:15-18 ("UBH bases coverage determinations on the Level of Care . . . Guidelines, the Coverage Determination Guidelines . . . , and/or the psychological and neurological testing guidelines.") (Niewenhous quoting Trial Ex. 735-0026). Typically, Peer Reviewers spend approximately thirty minutes talking to the physician who has requested the treatment and writing up their conclusions. Trial Tr. 1101:8-1102:13 (Martorana). The Peer Reviewer may spend additional time reviewing the information collected by the Care Advocate and the Care Advocate's recommendations. Trial Tr. 1101:10-1101:13 (Martorana).

50. If the Peer Reviewer makes a Clinical Non-Coverage Determination, UBH provides written notification of the determination to the member and the provider. Trial Ex. 880-004 (Stipulations of Facts) ¶ 8. The "[w]ritten notification of a denial" must include "[t]he rationale for the denial," which must "cite the Level of Care Guidelines, the Coverage Determination Guidelines, the Psychological and Neuropsychological Testing Guidelines, or other clinical guidelines required by contract or regulation, as appropriate, on which the denial was based . . . ." Trial Ex. 259-0020 (2014 UMPD). As a matter of UBH policy, UBH's denial letters must summarize all the reasons for denial. Trial Tr. 792:19-24 (Triana).

51. In this case, the denial letters (or in a few cases, the case notes) reflect that each class member's denial was based on UBH's determination that the member failed to meet the criteria in UBH's Guidelines. See Trial Ex. 896 (Class List stipulation); Trial Ex. 894 (denial letter and case note excerpts for Claim Sample).

G. The Plans

52. The specific terms and conditions of coverage for mental health and substance use disorder treatment administered by UBH are set forth in the plan term documents for each Plan, including but not limited to the Certificate of Coverage and/or Summary Plan Description. Trial Ex. 880-004 (Stipulations of Fact) ¶ 5. The Plans fall into two general categories: 1) fully insured plans,⁸ where UBH pays the benefits for the services it approves out of the fees it receives from the plans; and 2) self-funded plans,⁹ where UBH charges an administrative fee only, and the plan pays the benefits UBH approves. Trial Ex. 711-0003 to -0004 (Stipulation Concerning Per-Member Per-Month Rates) ¶¶ M, N.

8 The parties and witnesses also sometimes referred to the fully insured plans as "fully funded" or "risk" plans. The Court understands these terms to be interchangeable.

9 The self-funded plans were sometimes referred to as "administrative services only" (or "ASO") plans. Again, no distinction was drawn between these terms, which the Court understands to be interchangeable.

53. Every class member's health benefit plan includes, as one condition of coverage, a requirement that the requested treatment must be consistent with generally accepted standards of care. See Trial Ex. 892 (Plaintiffs' summary of plan terms for Claim Sample); Trial Tr. 674:5-675:7 (Duh). This requirement is conveyed in a variety of ways, with some plans providing that coverage is available only for services that are consistent with generally accepted standards of care and others excluding services that are not. Id. The exact phrasing of this requirement varies somewhat from plan to plan. Id. These minor variations do not reflect any substantive difference between the plans with respect to the requirement that covered services must be consistent with generally accepted standards of care. On the other hand, Plaintiffs do not dispute that a service that is consistent with generally accepted standards of care may, nonetheless, be excluded from coverage under a particular class member's plan. See Trial Tr. 685:24-686:1 (Duh) (testifying that she was not opining that all treatments that are consistent with generally accepted standards of care are covered under the class members' benefit plans).

54. All of the class members' health benefit plans grant discretion to UBH, as the claims administrator, to interpret plan terms, limitations and exclusions in determining whether a requested service is covered. Trial Tr. 36:04-10 (UBH admission that the plans "grant UBH the discretion to interpret the plans and manage the behavioral health benefits under those plans"); Trial Tr. 38:05-07 (UBH admission that "the health benefit plans give UBH the discretion to interpret the plans, administer the benefits, and decide if the treatment is medically necessary"); Trial Tr. 908:24-909:2 (Dehlin) (testifying that the responsibility of the claims administrator is to "apply the terms of the plan"); see also Trial Ex. 1653 (UBH plan summary for Claim Sample). The Guidelines UBH promulgates are an exercise of the discretion the Plans delegated to UBH as the claims administrator. Trial Ex. 880-004 (Stipulations of Fact) ¶¶ 3, 5.

55. Some of the class members' plans expressly reference the Guidelines used by UBH to administer claims for coverage. Trial Tr. 854:15-20 (Dehlin). For example, the Alexander Plan excludes "services which are not consistent with [UBH's] level of care guidelines or best practices as modified from time to time." Trial Ex. 225-107. These references do not convert the Guidelines into Plan terms. The Guidelines themselves do not purport to be plan terms but rather, are described in the introduction of all of the relevant versions as objective criteria for making standardized decisions about coverage. See Trial Ex. 1-0002 (2011 Guidelines); Trial Ex. 2-0002 (2012 Guidelines); Trial Ex. 3-0002 (2013 Guidelines); Trial Ex. 4-0002 (2014 Guidelines); Trial Ex. 5-0004 (2015 Guidelines); Trial Ex. 6-0007 (2016 Guidelines (January)); Trial Ex. 7-0007 (2016 Guidelines (June)); and Trial Ex. 8-0004 (2017 Guidelines). Similarly, UBH has consistently treated the Guidelines as being distinct from the Plans in its Utilization Management Program Descriptions ("UMPD"). For example, the 2013 UMPD describes the LOCGs as "clinically-based indicators developed to assist Care Advocacy personnel with making benefit decisions about appropriate levels of care for individual members," Trial Ex. 258-0012, and states that "[t]he role of the Peer Reviewer is to exercise clinical judgment in reviewing the relevant information and to review the case against the pertinent Level of Care Guidelines, . . . [and] the member's benefit plan . . . ." Trial Ex. 258-0018. Similar language is used in the UMPDs for all other years in the class period. See Trial Ex. 259-0019 (2014 UMPD); Trial Ex. 260-0010 (2015 UMPD); Trial Ex. 1186-0010 (2016 UMPD); Trial Ex. 262-0013 (2017 UMPD); Ex. 257-0020 (2012 UMPD template).

56. The Court's conclusion that the Guidelines are not Plan terms is further supported by the evidence showing that they are developed internally by UBH without input from Plan sponsors. In addition, no evidence was offered to show that when UBH revises the Guidelines it complies with the requirements contained in class members' Plans for amending those Plans. See Reply Brief, Ex. A (summarizing relevant provisions of plans of Claim Sample members with respect to amendment and identifying relevant trial exhibit numbers).¹⁰ Furthermore, the Court concludes that the language quoted above, referring to UBH's Guidelines "as modified from time to time," is not a delegation to UBH by Plan sponsors of authority to amend the Plans without the express approval of Plan sponsors. Such unfettered discretion to modify the terms of the Plans without notice to Plan participants and beneficiaries flies in the face of ERISA, which requires that participants and beneficiaries must be provided with a summary plan description that is "written in a manner calculated to be understood by the average plan participant," and that is "sufficiently accurate and comprehensive to reasonably apprise such participants and beneficiaries of their rights and obligations under the plan." 29 U.S.C. § 1022(a). In light of this requirement, the Court does not construe the references to UBH Guidelines in class members' Plans as transforming the Guidelines into Plan terms or giving UBH the authority to change the terms of class members' Plans without the approval of Plan sponsors.

10 The Court overrules UBH's objection to Exhibit A to the Reply Brief. UBH contends this exhibit is a "belated" summary exhibit under Rule 1006 of the Federal Rules of Evidence, which permits the use of a "summary, chart, or calculation to prove the content of voluminous writings, recordings, or photographs that cannot be conveniently examined in court." Because it was not offered or discussed at trial, UBH contends, it is improper. UBH is incorrect. Plaintiffs do not use Exhibit A to prove the content of anything. All of the plans that are listed in the exhibit were introduced into evidence at trial. The chart merely summarizes the relevant provisions of the plans in support of an argument made in response to UBH's assertion in its post-trial brief that the Guidelines are plan terms rather than a tool to interpret plan terms.

H. Whether the UBH Guidelines Adhere to Generally Accepted Standards of Care

1. Sources of Generally Accepted Standards of Care

57. In the context of this case, generally accepted standards of care are the standards that have achieved widespread acceptance among behavioral health professionals. There is no single source of generally accepted standards of care. Rather, they can be gleaned from multiple sources, including peer-reviewed studies in academic journals, consensus guidelines from professional organizations, and guidelines and materials distributed by government agencies. Trial Tr. 958:3-9 (Martorana). In this case, expert witnesses for both Plaintiffs and UBH offered opinions about generally accepted standards of care. While they relied on a variety of sources in support of their opinions, the resources that all of the experts agreed reflect generally accepted standards of care include the following: 1) the American Society of Addiction Medicine Criteria ("ASAM Criteria"); 2) the American Association of Community Psychiatrist's ("AACP") Level of Care Utilization System ("LOCUS"); 3) the Child and Adolescent Level of Care Utilization System ("CALOCUS") developed by AACP and the American Academy of Child and Adolescent Psychiatry ("AACAP"), and the Child and Adolescent Service Intensity Instrument ("CASII"), which was developed by AACAP in 2001 as a refinement of CALOCUS; and 4) the Medicare benefit policy manual issued by the Centers for Medicare and Medicaid Services ("CMS Manual"). Other sources that the parties' witnesses relied upon and which the Court finds reflect generally accepted standards of care are: 1) the APA Practice Guidelines for the Treatment of Patients with Substance Use Disorders, Second Edition (Trial Ex. 634); 2) the APA Practice Guidelines for the Treatment of Patients with Major Depressive Disorder (Trial Ex. 639); and 3) AACAP's Principles of Care for Treatment of Children and Adolescents with Mental Illnesses in Residential Treatment Centers (Trial Ex. 693).

58. ASAM Criteria: The American Society of Addiction Medicine ("ASAM") is a society of physicians and other professionals who specialize in the treatment of substance use disorders. Trial Tr. 65:4-6 (Fishman). ASAM has published three editions of the ASAM Criteria. The parties in this case relied on the Second Edition - Revised, published in 2001, and the Third Edition, published in 2013. See Trial Ex. 642 (ASAM PPC-2R: ASAM Patient Placement Criteria for the Treatment of Substance-Related Disorders (Second Edition - Revised)); Trial Ex. 662 (The ASAM Criteria: Treatment Criteria for Addictive, Substance-Related and Co-Occurring Conditions) (Third Edition)). The ASAM Criteria are the most widely accepted articulation of the generally accepted standards of care for how to conduct a comprehensive multidimensional assessment of a patient with substance related disorder, translate that into patient treatment needs and match those needs to the appropriate level of care. Trial Tr. 69:20-24 (Fishman); Trial Tr. 1575:25-1576:2 (Alam) ("the ASAM Criteria are consistent with generally accepted standards of care."); Trial Tr. 957:22-958:9, 1112:5-16 (Martorana); Trial Tr. 1375:21-25 (Allchin). In many states, including Rhode Island and Illinois, state-funded providers are required to use ASAM Criteria for placement of patients with substance related disorders. See Trial Ex. 548-00069 to -00070; see also Trial Ex. 673-0004 (2011 article by Martorana and Alam entitled "Addiction Treatment: Level of Care Determination" (hereinafter, "Alam/Martorana Article"), stating that "[a]bout 30 U.S. states require the use of at least some aspects of the ASAM [C]riteria.").

The ASAM Criteria set forth the "Six Dimensions of Multidimensional Assessment," establishing six "unique dimensions, which represent different life areas that together impact any and all assessment, service planning, and level of care placement decisions." Trial Ex. 662-0064. ASAM uses this multidimensional approach to "create a holistic, biopsychosocial assessment of an individual to be used for service planning and treatment across all services and levels of care." Id. The ASAM Criteria also describe a continuum of care, using a numbering system ranging from 0.5 (Early Intervention) to 4.0 (Medically Managed Intensive Inpatient Services). Trial Ex. 662-0127 to -0128. The core of the ASAM Criteria is in Chapters 4, 5 and 7, which together set forth the "decision rules" for using the six dimensions to determine which level of care a patient should be treated in, and for how long. See Trial Tr. 69:23 (Fishman).

59. LOCUS: The LOCUS, Trial Ex. 653, was developed in the 1990s to articulate generally accepted standards for level of care placement for mental health treatment of adults. Trial Tr. 499:24-500:25 (Plakun); Trial Tr. 501:2-503:10 (Plakun). It has been updated several times, including in 2009. Trial Ex. 653-0001; Trial Tr. 500:15-19 (Plakun). The parties agree that LOCUS reflects generally accepted standards of care. Trial Tr. 503:7-10 (Plakun); Trial Tr. 1241:25-1242:10, 1338:18-20 (Simpatico). LOCUS uses "six evaluation parameters or dimensions: 1) Risk of Harm; 2) Functional Status; 3) Medical, Addictive and Psychiatric Co-Morbidity; 4) Recovery Environment; 5) Treatment and Recovery History; and 6) Engagement and Recovery Status." Trial Ex. 653-004. It also defines six "levels of care" in the service continuum, where "each level describes a flexible or variable combination of specific service types," and sets forth patient placement criteria for each level. Trial Ex. 653-0005.

60. CALOCUS/CASII: The Child and Adolescent Level of Care Utilization System, Trial Ex. 644, which was renamed the Child and Adolescent Service Intensity Instrument in 2001, is based on LOCUS but is "adapted to reflect a developmental perspective, family focus, and inclusion of the comprehensive array of services in systems that serve children and adolescents." Trial Ex. 645-0005. CASII is aimed at children between 6 and 18 years old. Id. CASII was most recently updated in 2014. Trial Ex. 645-0001. There is no dispute that CALOCUS and CASII reflect generally accepted standards of care for determining the most appropriate level of care for children and adolescents. Trial Tr. 180:9-13 (Fishman); Trial Tr. 1453:2-5, 1455:4-6 (Allchin).

61. CMS Manual: The Centers for Medicare and Medicaid Services administer the Medicare program. Coverage decisions under Medicare must comply with the CMS Manual, the functional equivalent of the "health plan" for purposes of Medicare. The CMS Manual includes provisions on what constitutes a "reasonable expectation of improvement" (see, e.g., Trial Ex. 656-0026 to -0027), standards for determining frequency and duration of services (see, e.g., Trial Ex. 656-0028), and definitions of active treatment (see, e.g., Trial Ex. 735-0104 to -0105, Trial Ex. 655-0006 to -0008) and custodial care (see, e.g., Trial Ex. 735-0088 to -0089, Trial Ex. 654-0029). There is no dispute that the standards on improvement, custodial care, and active treatment set forth in the CMS Manual are consistent with generally accepted standards of care. See, e.g., Trial Ex. 281-0002 (UBH's "Hierarchy of Clinical Evidence"); Trial Tr. 310:4-6, 311:12-20 (Niewenhous); Trial Tr. 111:14-17 (Fishman); Trial Tr. 499:10-16 (Plakun).

2. Continuum of Service Intensity in Behavioral Health Care: Overview of Levels

62. In the area of mental health and substance use disorder treatment, there is a continuum of intensity at which services are delivered. In the most extreme situations, where a patient poses an imminent risk of serious harm to self or others, a provider will recommend inpatient hospitalization (referred to as Level Four under the ASAM Criteria). See Trial Ex. 662-0305 (ASAM Criteria). The focus of treatment at this level of service is crisis stabilization, that is, to address the acute crisis so that the patient can be moved to a lower level of care where the patient "can get back to doing the work that needs to happen over time" to address the "drivers of the recurrent risk of crisis." Trial Tr. 487:8-21 (Plakun).

63. The next level of intensity below inpatient hospitalization is residential treatment. Residential treatment is for individuals who do not pose an imminent risk of serious harm to self or others (i.e., who do not need inpatient hospitalization), but rather, "because of specific functional limitations, need safe and stable living environments and 24-hour care." Trial Ex. 662-0240 (ASAM Criteria) (describing generally ASAM Level 3 programs); see also Trial Ex. 634-0011 (APA Practice Guidelines for Treatment of Patients with Substance Use Disorders) ("Residential treatment is indicated for patients who do not meet the clinical criteria for hospitalization but whose lives and social interactions have come to focus predominantly on substance use, who lack sufficient social and vocational skills, and who lack substance-free social supports to maintain abstinence in an outpatient setting."); Trial Ex. 693-0011 (AACAP's Principles of Care for Treatment of Children and Adolescents with Mental Illnesses in Residential Treatment Centers) ("Residential care should be considered for those children and adolescents who present with prolonged and chronic symptoms that have not responded to acute, short-term hospitalization."). At this level of care, treatment is not limited to addressing acute symptoms to achieve crisis stabilization; instead, it is designed to provide patients with an "opportunity to engage underlying chronic, recurrent, comorbid issues" so that they are able to "turn a corner" and move to a lower level of service intensity. Trial Tr. 489:7-14 (Plakun).

64. Residential treatment takes different forms. As reflected most explicitly in the ASAM Criteria, there are sub-levels of residential treatment, "on a continuum ranging from the least intensive residential services [level 3.1] to the most intensive medically monitored intensive inpatient services [level 3.7]." Trial Ex. 662-0240 (ASAM Criteria). Level 3.7 programs "provide a planned and structured regimen of 24-hour professionally directed evaluation, observation, medical monitoring, and addiction treatment in an inpatient setting." Trial Ex. 662-0290 (ASAM Criteria). Levels 3.1 through 3.5 are "clinically managed," which means that "on-site physician services are not required" but patients still "are in need of interventions directed by appropriately trained and credentialed addiction treatment staff." Trial Ex. 662-0241 (ASAM Criteria).

65. Level 3.3 describes residential treatment programs designed specifically for persons with cognitive limitations, such as individuals with traumatic brain injury, developmental disabilities, and/or dementia. Trial Ex. 662-0255 to -0256 (ASAM Criteria). "Typically, [such patients] need a slower pace of treatment because of mental health problems or reduced cognitive functioning (Dimension 3), or because of the chronicity of their illness (Dimensions 4 and 5)." Trial Ex. 662-0256 (ASAM Criteria).

66. Level 3.1 requires at least five hours of individual, group and/or family therapy per week; it is "not intended to describe or include sober houses, boarding houses, or group homes where treatment services are not provided." Trial Ex. 662-0244 to -0245 (ASAM Criteria). "The length of stay in a clinically managed Level 3.1 program tends to be longer than in the more intensive residential levels of care [because] [l]onger exposure to monitoring, supervision, and low-intensity treatment interventions is necessary for patients to practice basic living skills and to master the application of coping and recovery skills." Trial Ex. 662-0244 to -0245 (ASAM Criteria).

67. The next level of service intensity below residential treatment is partial hospitalization ("PHP"). While partial hospitalization does not involve the 24-hour structure of residential treatment (and in that sense, is a lower level of care), it differs from residential treatment (and is more like inpatient hospitalization) in that it is an acute, crisis-focused level of care. Trial Tr. 488:13-17 (Plakun) ("[PHP is] generally focused on crisis stabilization, crisis intervention, in a way that's similar to the way inpatient hospitals are and usually limited in duration with an eye, again, toward stabilizing the crisis and returning someone to a lower level of care."); see also Trial Ex. 656-0031 (CMS Manual) ("Patients admitted to a PHP generally have an acute onset or decompensation of a covered Axis I mental disorder."). PHP treatment provides approximately 20 hours per week of treatment services. Trial Tr. 488: 5-11 (Plakun). Plaintiffs in this case do not challenge UBH's Guidelines for placement at the PHP level of care.

68. Below PHP, the next level of service intensity is intensive outpatient ("IOP") treatment. IOP is typically a structured program involving 9 hours per week of outpatient treatment (or 6 hours for children). See, e.g., Trial Ex. 5-0030 (UBH Guidelines) (describing IOP level of care). It is "a program in which you have added services [to routine outpatient treatment] to try to make it possible for someone to deal with the underlying comorbidities, recurrent problems, histories of early and later adversity, trauma, all the complexity that is actually in reality part of what mental disorders are about." Tr. 486:10-14 (Plakun). Intensive outpatient treatment, while more intensive than routine outpatient treatment, is "not at all limited to crisis stabilization." Trial Tr. 486:9-10 (Plakun).

69. The lowest level of service intensity is outpatient treatment, such as once- or twice-a-week psychotherapy. Some patients may be prescribed outpatient treatment only once, or for a short duration, but its purpose is just as commonly to treat chronic conditions. See, e.g., Trial Tr. 580:23-24 (Plakun) ("Someone might be seeking outpatient treatment for chronic reasons rather than acute reasons."); Trial Ex. 662-207 (ASAM Criteria) (Level 1 outpatient services often are provided indefinitely to patients with chronic conditions). In order for treatment to be effective at this level of care, the patient must be able not only to effectively "use the sessions," that is, to "manage them, bear what emotions get brought up in the course of them," and "understand instructions," but also to "function adaptively" between sessions. Trial Tr. 481:6-12 (Plakun). When there is "trouble in one or both of those domains," providers may "add services" in order to "help someone's capacity to use the sessions better and to manage adaptively between the[m]," such as "having sessions more frequently" or adding medications, skills training, group sessions, and/or substance abuse treatment. Trial Tr. 481:13-22 (Plakun).

3. Generally Accepted Standards of Care Relevant to the Guidelines Challenged in this Action

70. At trial, the parties offered extensive testimony on the generally accepted standards of care that apply to patient placement in the context of behavioral health treatment. The Court finds, by a preponderance of the evidence, that the following standards are generally accepted in the field of mental health and substance use disorder treatment and placement.

a. It is a generally accepted standard of care that effective treatment requires treatment of the individual's underlying condition and is not limited to alleviation of the individual's current symptoms

71. Many mental health and substance use disorders are long-term and chronic. See Trial Tr. 1599:24-1600:1 (Alam); Trial Tr. 486:1-5, 492:1-14 (Plakun); Trial Ex. 634-0127 (APA practice guideline stating that "[a]lthough there is considerable heterogeneity among patients with substance use disorders, the disease course is often chronic, lasting for years"); Trial Ex. 548-0010 (Optum draft document entitled Optum Point of View on Substance Use Disorder (SUD) stating that "Substance Use Disorder (SUD) is a chronic, complex condition that is subject to reoccurrence of symptoms (relapse)".). While current symptoms are typically related to a patient's chronic condition, see Trial Tr. 972:10-11 (Martorana), it is generally accepted in the behavioral health community that effective treatment of individuals with mental health or substance use disorders is not limited to the alleviation of the current symptoms. Rather, effective treatment requires treatment of the chronic underlying condition as well. Thus, ASAM recommends that practitioners develop an "individualized plan [that is] based on a comprehensive biopsychosocial assessment of the patient" and explains that "[a]ddiction treatment services have as their goal not simply stabilizing the patient's condition but altering the course of the patient's disease toward wellness." Trial Ex. 662-0025 (ASAM Criteria). Likewise, Dr. Plakun testified that mental health treatment that only manages crises is not effective, as "you wind up in a recipe that is sadly all too familiar in the world these days; that is, of people going in and out of hospital, rotating back and forth between trying to make outpatient treatment work, failing in it, having chronic ongoing crises that need to be managed, winding up in an inpatient unit." Trial Tr. 492:1-9 (Plakun). Analogizing a chronic mental health condition to a pot of water over a flame, Dr. Plakun testified, "[i]t's optimal to try to find a way to turn the flame down and not simply feed the recurrent loop of crisis." Trial Tr. 492:13-14 (Plakun); see also Trial Tr. 490:8-14 (Plakun) ("You cannot really assess an individual's needs in terms of a treatment plan, including level of care, unless you get a pretty comprehensive picture not only of what's the . . . presenting symptom right now, but also how does that connect to the part of the iceberg that's not sticking up out of the water. What's this person's story? What are they struggling with?"); Trial Tr. 701:19-21 (Triana) ("[O]ngoing mental illness is not necessarily cured when an acute episode is stabilized.").

b. It is a generally accepted standard of care that effective treatment requires treatment of co-occuring behavioral health disorders and/or medical conditions in a coordinated manner that considers the interactions of the disorders and conditions and their implications for determining the appropriate level of care

72. Many individuals with a behavioral health diagnosis have multiple, co-occurring disorders. Trial Tr. 484:6-17 (Plakun) (testifying that patients who seek treatment for a particular mental health diagnosis often have "chronic, comorbid, recurrent underlying issues"). Co-occurring disorders can interact in a "reciprocal way" that makes each of them "worse." Trial Tr. 81:9-17 (Fishman); see also Trial Tr. 610:24-611:14 (Plakun). Because co-occurring disorders can aggravate each other, treating any of them effectively requires a comprehensive, coordinated approach to all conditions. Trial Tr. 81:18-22 (Fishman); Trial Tr. 525:16-20 (Plakun) ("[T]he whole focus of the treatment . . . is to focus on treating, not simply managing, but engaging and treating the co-occurring behavioral health issues."); Trial Ex. 673-0006 (Alam/Martorana Article) ("Addicted or drug-abusing individuals with co-occurring mental disorders should have both disorders treated in an integrated way."). Similarly, the presence of a co-occurring medical condition is an aggravating factor that may necessitate a more intensive level of care for the patient to be effectively treated. See, e.g., Trial Tr. 108:3-5, 108:22-24, 139:15-17, 227:13-20 (Fishman).

The ASAM Criteria and the LOCUS both reflect the importance of a comprehensive approach to treating co-occurring disorders in determining the appropriate level of care. For example, ASAM Dimension 2 "assesses the need for physical health services, including whether there are needs for acute stabilization and/or ongoing disease management for a chronic physical health condition." Trial Ex. 662-0066 (ASAM Criteria); Trial Tr. 77:25-78:3 (Fishman). ASAM Dimension 3 "assesses the need for mental health services," and "specifically references mental health conditions, including trauma-related issues and conditions such as posttraumatic stress, cognitive conditions and developmental disorders, and substance related mental health conditions." Trial Ex. 662-0066 (ASAM Criteria); see also Trial Ex. 662-0046 (ASAM Criteria) (noting that when "two or more disorders co-occur and are concurrent, they all need to be addressed simultaneously as 'primary' conditions in order to provide the most effective integrated and holistic care").

Similarly, LOCUS Dimension III recognizes the importance of taking a comprehensive approach to co-occurring disorders in order to effectively treat a patient and to determine the appropriate level of care. Dimension III "measures potential complications in the course of illness related to co-existing medical illness, substance use disorder, or psychiatric disorders, in addition to the condition first identified or most readily apparent (here referred to as the presenting disorder)." Trial Ex. 653-0011 (LOCUS). LOCUS further recognizes that "[c]o-existing disorders may prolong the course of illness in some cases, or may necessitate availability of more intensive or more closely monitored services in other cases." Trial Ex. 653-0011 (LOCUS).

c. It is a generally accepted standard of care that patients should receive treatment for mental health and substance use disorders at the least intensive and restrictive level of care that is safe and effective

73. In order to treat patients with mental health or substance use disorders effectively, it is important for providers to "match" them to the appropriate level of care. See Trial Tr. 7:14-76: 22 (Fishman); Trial Ex. 673-004 (Alam/Martorana Article) ("Choosing the appropriate level of care is important."). The evidence presented at trial supports the conclusion that under generally accepted standards, the driving factors in determining the appropriate treatment level should be safety and effectiveness; however, where the clinician determines that more than one service level will meet both of these requirements, the least intensive and/or restrictive setting should be selected. See Trial Ex. 662-0132 (ASAM Criteria) ("The paramount objective [of treatment] should be safety and effectiveness"); Trial Ex. 639-16 (APA Practice Guideline for the Treatment of Patients with Major Depressive Disorder) (stating that "[t]he psychiatrist should determine the least restrictive setting for treatment that will be most likely not only to address the patient's safety, but also to promote improvement in the patient's condition"); Trial Ex. 1507-17 (CMS Manual) (stating that "[i]n general, patients should be treated in the least intensive and restrictive setting which meets the needs of their illness"); Trial Ex. 634-22 (APA Practice Guideline for the Treatment of Patients With Substance Use Disorders) (stating that "[i]ndividuals should be treated in the least restrictive setting that is likely to prove safe and effective"); Trial Ex. 662-374 (ASAM Criteria) (noting that "[r]eferral to the 'least intensive level of care that is effective' or to the 'least restrictive environment for care' is generally the norm for members of the general public who seek addiction treatment"); Trial Ex. 673-004 (Alam/Martorana Article) (stating that "[t]he ideal level of care is one that is least intensive, that addresses all the treatment needs, and that provides the individual the best opportunity to develop sobriety"); Trial Tr. 213:9-15 (Fishman) (testifying that placement decisions are typically driven by what is most effective but where two levels of care are identically effective, which "rarely occur[s]," the less restrictive level should be chosen).

The evidence at trial did not support the conclusion that under generally accepted standards of care, there is a balancing of effectiveness against the restrictiveness or intensity factor; in other words, the fact that a lower level of care is less restrictive or intensive does not justify selecting that level if it is also expected to be less effective. Placement in a less restrictive environment is appropriate only if it is likely to be safe and just as effective as treatment at a higher level of care in addressing a patient's overall condition, including underlying and co-occurring conditions.

d. It is a generally accepted standard of care that when there is ambiguity as to the appropriate level of care, the practitioner should err on the side of caution by placing the patient in a higher level of care

74. Research has demonstrated that patients with mental health and substance use disorders who receive treatment at a lower level of care than is clinically appropriate face worse outcomes than those who are treated at the appropriate level of care. See Trial Tr. 74:14-75:13 (Fishman) (describing research findings regarding adverse consequences of mismatching to a lower level of care in the area of substance use disorder treatment); Trial Ex. 673-004 (Alam/Martorana Article) (noting that improper placement at less intensive level of care for substance use disorder may result in relapse). On the other hand, there is no research that establishes that placement at a higher level of care than is appropriate results in an increase in adverse outcomes. Trial Ex. 673-004 (Alam/Martorana Article) (stating that "[t]here is no research evidence to the existence of a consequence to choosing a more intensive level of care than necessary"); Trial Tr. 1674:9-11 (Alam) (testifying that "there's no research saying if you choose a higher level of care, whether it's bad for you"). Consequently, it is a generally accepted standard of care that where there is uncertainty as to the likely effectiveness of different proposed levels of care, practitioners treating patients for mental health and substance use disorders should exercise caution by selecting the higher level of service intensity. See Trial Ex. 653-0007 (LOCUS) (stating that when there is "ambiguity" with respect to the appropriate level of care practitioners should assign the "highest score in which it is more likely than not that [at] least one criterion has been met should generally be assigned" so that "errors [regarding the appropriate level of service] will be made on the side of caution"); Trial Ex. 662-0132 (ASAM Criteria) ("In general, when the criteria designate a treatment placement that is not available, a strategy must be crafted that gives the patient the needed services in another placement or combination of placements. The paramount objective should be safety and effectiveness, which usually requires opting for a program of greater intensity than the placement criteria indicate."); Trial Ex. 656-0026 (CMS Manual) ("Services are noncovered only where the evidence clearly establishes that the criteria are not met.") (emphasis added); Trial Ex. 645-0021 (CASII) (providing that "[t]he clinician should select the highest rating level in each dimension that most accurately identifies the child or adolescent's condition"); Trial Tr. 213:22-214:1 (Fishman) ("In general, the approach is, if the most effective level of care is not available or there's a gray area between two levels of care, one should take the conservative position and round up, as it were, or go to the next highest level of care.").

e. It is a generally accepted standard of care that effective treatment of mental health and substance use disorders includes services needed to maintain functioning or prevent deterioration

75. While effective treatment may result in improvement in the patient's level of functioning, it is well-established that effective treatment also includes treatment aimed at preventing relapse or deterioration of the patient's condition and maintaining the patient's level of functioning. Thus, for example, the CMS Manual provides that services satisfy the "reasonable expectation of improvement" requirement for Medicare coverage "[w]here there is a reasonable expectation that if treatment services were withdrawn the patient's condition would deteriorate, relapse further, or require hospitalization." Trial Ex. 656-0026 (CMS Manual). The CMS Manual explains, "[f]or many . . . psychiatric patients, particularly those with long-term, chronic conditions, control of symptoms and maintenance of a functional level to avoid further deterioration or hospitalization is an acceptable expectation of improvement." Trial Ex. 656-0026 (CMS Manual); see also Trial Ex. 653-0009 (LOCUS) ("[p]ersons with ongoing, longstanding deficits who do not experience any acute changes in their status" automatically given a rating of three for LOCUS "functional status" dimension); Trial Tr. 110:24-111:23 (Fishman) (testifying that these concepts reflect generally accepted standards of care); Trial Tr. 561:1-562:11 (Plakun) (same).

Similarly, ASAM cautions that "[t]reatment successes such as a period of abstinence or improvement in function sometimes are misinterpreted as indicating that treatment is completed." Trial Ex. 662-020 (ASAM Criteria). Instead, treatment of substance use disorders should continue so long as there is a risk of relapse. See Trial Tr. 131:5-18 (Fishman) ("Q: If the acute symptoms [that] have made somebody with substance use disorder seek treatment [no longer require treatment], does that mean that that person is cured and no longer requires treatment of any kind, Doctor? A: . . . [N]othing could be further from the truth for many patients who are succeeding in ongoing, enduring, low-intensity treatment like outpatient treatment. It is the treatment itself and its enduring nature that is keeping them in good stead, and we would be remiss to discontinue it to wait for them to relapse to need further treatment.").

f. It is a generally accepted standard of care that the appropriate duration of treatment for behavioral health disorders is based on the individual needs of the patient; there is no specific limit on the duration of such treatment

76. As the CMS Manual explains, "[t]here are no specific limits on the length of time that services may be covered." Trial Ex. 656-0028 (CMS Manual). Rather, in determining whether to continue treatment, practitioners consider such factors as "the nature of the illness, prior history, the goals of treatment, and the patient's response." Trial Ex. 656-0028 (CMS Manual); see also Trial Ex. 673-0005 (Alam/Martorana Article) ("The appropriate duration for individuals depends on their problems and needs."); Trial Ex. 662-0325 (ASAM Criteria).

g. It is a generally accepted standard of care that the unique needs of children and adolescents must be taken into account when making level of care decisions involving their treatment for mental health or substance use disorders

77. One of the primary differences between adults, on the one hand, and children and adolescents, on the other, is that children and adolescents are not fully "developed," in the psychiatric sense. See Trial Tr. 495:19-25 (Plakun) (testifying that a person does not become an adult until the age of 25). Clinicians recognize that a child or adolescent's level of development is an important consideration in making level of care determinations. See Trial Tr. 495:16-18 (Plakun) ("[F]or many people, particularly a group we call emerging adults, it's extraordinarily important to pay attention to developmental considerations."); Trial Tr. 101:4-13 (Fishman) ("adolescents have a different set of needs [than adults], they have different assets and vulnerabilities."); Trial Tr. 1383:2-1384:15, 1385:11-21 (Allchin) (addressing treatment needs specific to children and adolescents). The ASAM Criteria, for example, recognize that "[t]o be most effective" in treating adolescents, practitioners "must adapt their methods and strategies to respond to adolescents' emotional, behavioral, and cognitive vulnerabilities and strengths, as well as a developmental perspective that evolves dynamically." Trial Ex. 662-0070 (ASAM Criteria).

78. One of the ways practitioners take into account the developmental level of a child or adolescent in making treatment decisions is by relaxing the threshold requirements for admission and continued service at a given level of care. See Trial Tr. 151:19-22 (Fishman) ("For any given level of care, the entry criteria, that is, the decision rules for matching treatment severity and needs to level of care, are more inclusive, more permissive for adolescents."); see also Trial Tr. 152:7-9 (Fishman) ("[I]n a variety of ways, we tend to think that youth would need higher levels of care for longer durations with lower barriers to access than adults."). Thus, under the ASAM Criteria, placement at a given level of care might be appropriate for a child or adolescent with a lower level of severity in Dimension 1 than would be required to warrant the same placement for an adult. Trial Tr. 151:19-25 (Fishman). Similarly, the ASAM Criteria apply a more lenient standard to children and adolescents by not requiring a showing in as many dimensions as is required for adults to warrant the same level of care. For example, for an adult to meet criteria for level 3.1 residential treatment, the patient must "meet[] specifications in each of the six dimensions," whereas an adolescent need only "meet[] specifications in at least two of the six dimensions." Trial Ex. 662-0249 to -0053 (ASAM Criteria). As a corollary of these more lenient standards, children and adolescents are likely to need longer duration of treatment than adults. Trial Tr. 101:4-13 (Fishman) ("[A] dole scents have a different set of needs. . . Most often they will need longer duration of treatment than adults."); see also Trial Ex. 645-0042 (CASII) ("It may be desirable . . . for a child or adolescent to remain at a higher level of service intensity to preclude relapse and unnecessary disruption of care, and to promote lasting stability."); Trial Tr. 1463:15-21 (Allchin) ("[I]t might be appropriate to require some level of improvement for an adult within some period of time that might not be appropriate for a child").

h. It is a generally accepted standard of care that the determination of the appropriate level of care for patients with mental health and/or substance use disorders should be made on the basis of a multidimensional assessment that takes into account a wide variety of information about the patient

79. "Individuals with mental and substance use disorders can be viewed as suffering from biopsychosocial illnesses that, to varying degrees, have biological and medical, psychological and psychiatric, and sociocultural origins and clinical features." Trial Ex. 662-0075 (ASAM Criteria). Consequently, except in acute situations that require hospitalization, where safety alone may necessitate the highest level of care, decisions about the level of care at which a patient should receive treatment should be made based upon a "holistic, biopsychosocial assessment" that involves consideration of multiple dimensions. Trial Tr. 490:25-491:2 (Plakun) (testifying that it is a generally accepted standard of care to select a level of care where the acute crisis and the chronic and comorbid behavioral health conditions can be safely and effectively treated but that in the case of inpatient treatment "you might be forced to go with very limited information about a crisis").

80. Under the ASAM Criteria, for example, the six dimensions that clinicians should consider are as follows: 1) Acute Intoxication and/or Withdrawal Potential; 2) Biomedical Conditions and Complications; 3) Emotional, Behavioral, or Cognitive Conditions and Complications; 4) Readiness to Change; 5) Relapse, Continued Use, or Continued Problem Potential; and 6) Recovery/Living Environment. Trial Ex. 662-0064 (ASAM Criteria). These criteria are not rigid requirements for making level of care determinations. Instead, each of the six dimensions is "assessed independently and receives its own risk rating." Trial Ex. 662-0076 (ASAM Criteria). These scores are then combined, so that lower scores in one dimension can be offset by higher scores in another. Further, ASAM instructs that "cross-dimensional interactions" should be considered, as these may increase or decrease the level of risk. Trial Ex. 662-0080 ("Being aware of cross-dimensional interactions, and the potential increase or decrease in overall risk they pose, can have a great effect on service planning and placement decisions").

81. LOCUS has a similar set of dimensions, instructing clinicians to consider: (1) Risk of Harm, (2) Functional Status, (3) Medical, Addictive, and Psychiatric Co-Morbidity, (4) Recovery Environment, (5) Treatment and Recovery History, and (6) Engagement and Recovery Status. Trial Ex. 653-0008 to -0018 (LOCUS). As is the case under the ASAM Criteria, placements using LOCUS are based on consideration of all of the dimensions, and a low score in one dimension may be offset by a higher score in another, with the result that different combinations of factors within the LOCUS dimensions may point toward the same placement determination. See Trial Ex. 653-0028 to -0029 (LOCUS Level of Care Determination Decision Tree); see also Trial Tr. 84:2-7, 84:2-7 (Fishman) ("[T]he numbering of [the dimensions] and the ordering of them and the names used isn't the critical thing . . . . [H]owever you order them, however you name them, however you enumerate or catalog them, the content of each of these is essential to being able to do a comprehensive assessment, a comprehensive enumeration of treatment needs, and then using that as the basis for a level of care placement matching."); Trial Tr. 490:2-14, 491:3-14 (Plakun) (a "comprehensive, multifaceted assessment from multiple domains . . . is what mental healthcare is about").

4. Whether UBH Guidelines are Consistent with Generally Accepted Standards of Care

a. Whether UBH Guidelines deviate from generally accepted standards of care by placing excessive emphasis on acuity and crisis stabilization

82. Having reviewed all of the versions of the Guidelines that Plaintiffs challenge in this case and considered the testimony of the witnesses addressing the meaning of the Guidelines, the Court finds, by a preponderance of the evidence, that in every version of the Guidelines in the class period, and at every level of care that is at issue in this case, there is an excessive emphasis on addressing acute symptoms¹¹ and stabilizing crises while ignoring the effective treatment of members' underlying conditions. While the particular form this focus on acuity takes varies somewhat between the versions, in each version of the Guidelines at issue in this case the defect is pervasive and results in a significantly narrower scope of coverage than is consistent with generally accepted standards of care.¹²

11 The Court does not consider the dictionary definitions offered by Plaintiff in their reply brief and therefore does not rule on UBH's objections to those definitions.

12 The specific provisions of the Guidelines that reflect a focus on the treatment of acute symptoms that is inconsistent with generally accepted standards of care are identified by Plaintiffs in the Consolidated Claims Chart, Docket No. 404-2 ("Claims Chart"), with the short form "Acuity" in the "Why Flawed" column of the chart. For the reasons set forth herein, and based on the specific testimony cited in the Claims Chart, the Court finds that each of these provisions is inconsistent with generally accepted standards of care requiring effective treatment of both acute and chronic conditions.

i. Meaning of "acute" and related terms used in the Guidelines

83. As a preliminary matter, the Court addresses the meaning of the word "acute" for the purposes of this case. Based on the evidence and testimony introduced at trial, the Court concludes that in the context of the treatment of mental health and substance use disorders, this word generally refers to both the timing and severity of a patient's condition or symptoms. See Trial Tr. 80:10-13 (Fishman) (testifying that ASAM Dimension 1 is about "acute intoxication," that is, whether "the person [is] currently experiencing or under the influence of substances") (emphasis added); Trial Tr. 269:20-24 (Fishman) (describing "acute" problems as those that are "different from baseline"); Trial Tr. 1005:24-1006:1 (Martorana) (testifying that "acute changes" are those that are "immediate, generally short-lived, and have some impact as to why [the patient] needs to be in this level of care"); Trial Tr. 1083:15-16 (Martorana) ("acute symptoms" are "symptoms that have arisen relatively short term as opposed to long-lasting chronic symptoms"). Likewise, in the Guidelines the word "acute" is used to focus on the immediate crisis, that is, symptoms associated with rapid onset that are typically of short duration and that cause the patient to seek treatment at that time.

84. Testimony by UBH witnesses that the word "acute" and phrases such as "acute changes" refer not only to the recent changes that brought the member to treatment but also the chronic or comorbid conditions that may not be immediately apparent was not credible in light of other evidence and testimony in the record. See, e.g., Trial Tr. 1599:15-20 (Alam) ("When you talk about acute symptoms, you're really referring to the acute changes of a chronic condition. . . . So when you're talking about acute changes, you're referring to the acute changes that are contributed because of the underlying chronic condition."). For example, when asked why the word "acute" had been removed from the 2017 Common Criteria but had been left in the Guideline for residential treatment, Dr. Martorana testified:

we took it out of the other ones and we left it in here because we recognize that residential treatment is a 24-hour level of care for someone who requires a higher, more intensive level of care. So we want to understand what happened, what changed that -- what was the new change that happened that needs to be addressed that puts them into a 24-hour setting.

Trial Tr. 1006:21-1007:2 (Martorana) (emphasis added).

It is also apparent from denial letters that were sent to Claim Sample members that coverage of services to treat "acute" symptoms under the Guidelines was about crisis stabilization rather than treatment of the member's underlying condition . See, e.g., Trial Ex. 236 (denial letter stating that coverage was denied because "[t]he crisis which led to [the member's] admission to acute facility based care has resolved"); Trial Ex. 238 (denial letter stating that coverage for multiple weekly therapy sessions was denied because "[t]he use of multiple weekly therapy sessions typically is limited to acute exacerbations of illness, or in the context of a clinically urgent situation"); Trial Ex. 1350 (same); Trial Ex. 1373 (denial letter stating that coverage was denied because the member was "not exhibiting risk factors that require acute stabilization").

Indeed, testimony by Mr. Niewenhous, who was the UBH employee primarily responsible for development of the Guidelines during most of the class period, reflects that this focus on crisis stabilization is a fundamental tenet of the "Acute Care Utilization Management Model" upon which the Guidelines are based. In particular, Mr. Niewenhous testified at trial that a 2015 Powerpoint presentation that he created describing UBH's "Acute Care UM Model," referred to the fact that in UBH's "commercial business[,] the services focus on the reasons why somebody came into treatment at that point." Trial Tr. 303:4-305:3 (Niewenhous) (emphasis added) (testifying about Trial Ex. 512-0007). Similarly, in a 2016 email, Mr. Niewenhous stated that "[o]ur guidelines are used to authorize services. Presumption is that services are acute." Trial Ex. 522-0002. Mr. Niewenhous goes on to note in the email that "services for severely and persistently ill members that are intended to endure[] don't play to an acute care UR model." Id.

85. Numerous other words and phrases are used in the Guidelines to refer to the acute symptoms that cause a member to seek treatment, including "presenting symptoms," "presenting problems," "presenting condition," and factors "leading to" or "precipitating" admission. See, e.g., Trial Tr. 99:1- 4, 269:20-24 (Fishman) (testifying with respect to the phrase "presenting problems" in the Guidelines that "even though the word 'acute' isn't used, it focuses a user on thinking about the kinds of changes that are likely to be acute as different from baseline . . . ."). Dr. Martorana's testimony that "presenting problems" includes the "totality" of the member's condition, including chronic and co-morbid conditions, see Trial Tr. 983:1-8 (Martorana), was not credible for the reasons discussed above.

ii. Presenting Problems Requirement

86. One of the requirements that reflects UBH's overemphasis on acuity is the requirement contained in all challenged versions of the Guidelines that in order to obtain coverage upon admission, there must be a reasonable expectation that services will improve the member's "presenting problems" within a reasonable period of time. See Trial Ex. 1-0005 (2011 Level of Care Guidelines) Common Criteria ¶ 6 (providing, in part, that "[t]here must be a reasonable expectation that essential and appropriate services will improve the member's presenting problems within a reasonable period of time"); Trial Ex. 2-0007 (2012 Level of Care Guidelines) Common Criteria ¶ 6 (same); Trial Ex. 3-0008 (2013 Level of Care Guidelines) Common Criteria ¶ 7 (same); Trial Ex. 4-0009 (2014 Level of Guidelines) Common Criteria and Best Practices for All Levels of Care (requiring "a reasonable expectation that services will improve the member's presenting problems within a reasonable period of time"); Trial Ex. 5-0008 (2015 Level of Care Guidelines) Common Criteria and Clinical Best Practices for All Levels of Care 1.8 (requiring "a reasonable expectation that services will improve the member's presenting problems within a reasonable period of time"); Trial Ex. 6-0010 (2016 Level of Care Guidelines) Common Criteria and Clinical Best Practices for All Levels of Care ¶ 1.8 (same); Trial Ex. 7-0010 (2016 Level of Care Guidelines (June)) Common Criteria for All Levels of Care ¶ 1.8 (same); Trial Ex. 8-0007 (2017 Level of Care Guidelines) Common Admission Criteria for All Levels of Care (requiring "a reasonable expectation that service(s) will improve the member's presenting problems within a reasonable period of time").

87. The plain language of the "presenting problems" requirement focuses on the immediate, acute symptoms that brought the member to treatment rather than the broader question that should be considered under generally accepted standards of care, namely, whether the services being considered will be effective in treating not only the current symptoms but also the individual's underlying condition. This emphasis on crisis stabilization is further reinforced by the "reasonable period of time" requirement in the Guidelines quoted above, which suggests that treatment of long-term, chronic conditions beyond what is necessary to treat the presenting symptoms is not covered by this requirement.

88. This interpretation of the "presenting problems" requirement finds further support in the contemporaneous evidence, which reflects that UBH knowingly and purposefully drafted its Guidelines to limit coverage to acute signs and symptoms. In particular, in June 2010, the BPAC issued a request to the Coverage Determination Committee ("CDC") to "consider adding [to the CDG for Custodial Care] a condition to the definition of 'active treatment' that care should be in the least intensive level of care." Trial Ex. 307-0002. In response, the CDC adopted the following action item at a meeting on July 1, 2010: "Add clarification that reasonable expectation of improvement in the patient's condition is improvement in the patient's acute condition." Id. (emphasis in original). Mr. Niewenhous was instructed to "edit the CDG" accordingly, id., which he did. See Trial Ex. 10-0003 (August 2010 CDG for Custodial Care stating that "Improvement of the patient's condition is indicated by the reduction or control of the acute symptoms that necessitated hospitalization or residential treatment."); Trial Tr. 340:9-341:18 (Niewenhous) (testimony that Niewhous made the requested change in the Guidelines).

89. By including in the Common Criteria the "presenting problems" requirement cited above, UBH's Guidelines restrict coverage to treatment necessary to alleviate the patient's most immediate symptoms. This is because each criterion in the Common Criteria must be satisfied in order for services to be covered, as discussed above.

90. The focus on acuity associated with the "presenting problems" requirement was made particularly explicit in the 2012-2016 versions of the Guidelines, when a sentence was added to these provisions of the Common Criteria spelling out that "improvement" meant "reduction or control of the acute symptoms that necessitated treatment in a level of care." See Trial Ex. 2-0007 (2012 Level of Care Guidelines) Common Criteria ¶ 6; Trial Ex. 3-0008 (2013 Level of Care Guidelines) Common Criteria ¶ 7; Trial Ex. 4-0009 (2014 Level of Care Guidelines) sub-bullet beginning "[i]mprovement of"; Trial Ex. 5-0009 (2015 Level of Care Guidelines) Common Criteria and Clinical Best Practices for All Levels of Care ¶ 1.8.1; Trial Ex. 6-0010 (2016 Level of Care Guidelines) Common Criteria and Clinical Best Practices ¶ 1.8.1; Trial Ex. 7-0010 (2016 Level of Care Guidelines (June)) Common Criteria and Clinical Best Practices ¶ 1.8.1. Although UBH removed the word "acute" from this provision in the 2017 version, the Guidelines continued to require a "reasonable expectation" that services will "reduc[e] or control . . . the signs and symptoms that necessitated treatment in a level of care," thus changing the words used while preserving the meaning of the Guidelines with respect to the "presenting problems" requirement from the earlier versions. See Trial Ex. 8-0007 (2017 Level of Care Guidelines). Based on the evidence presented at trial, the Court finds that all versions of the Guidelines imposed the same "presenting problems" requirement, regardless of whether they used the term "acute" to describe it, and that this requirement is not consistent with generally accepted standards of care for the reasons stated above.

iii. Introduction of "why now" to the Guidelines

91. Starting in 2014, UBH went even further in limiting covered services to those aimed at the treatment of acute symptoms by adopting the concept of "why now," which comes from "crisis intervention literature." See Trial Ex. 1659-0006 to -0007 (Bonfield Dep.) at 206:10-15. The addition of "why now" to the Guidelines was the idea of former UBH Chief Medical Officer Dr. William Bonfield, who is a licensed psychiatrist. Trial Ex. 1659-0001, -0005 (Bonfield Dep.) at 10:07-10:18, 181:01-181:04. According to Dr. Bonfield, the concept was first developed by the medical director of a managed care company called Biodyne. Trial Ex. 1659-0006 to -0007 (Bonfield Dep.) at 205:21-206:05. In the 2014 version of the Guidelines, UBH defined "why now" as the "acute changes in the member's signs and symptoms and/or psychosocial and environmental factors leading to admission." Trial Ex. 4-0007 (2014 Guidelines) (Admission sub-bullet beginning "[t]he member's current"). The same definition is used in the 2015 and 2016 versions of the Guidelines. See Trial Ex. 5-0008 (2015 Guidelines); Trial Ex. 6-0009 (2016 Guidelines); Trial Ex. 7 (2016 Guidelines (June)). Thus, the Court finds that the meaning of "why now" as used in the Guidelines is unambiguous and refers to the recent severe changes in the member's signs and symptoms and/or psychosocial and environmental factors.

92. Dr. Bonfield testified that "why now" is aimed at addressing the "root cause" of a patient's problems, suggesting that the concept encompasses not only "acute changes" but also the patient's underlying chronic condition. See Trial Ex. 1659-003 to -004 (Bonfield Dep.) at 176:20-177:22. Witnesses Martorana, Allchin and Robinson-Beale offered similar testimony. See Trial Tr. 1054:12-17 (Martorana) (testifying that the "why now" factors "really want to focus people more on thinking about the whole person and everything they're bringing to the point of request for this level of care, the 'why now'"); Trial Tr. 1422:19-1423:2 (Allchin) (testifying that an individualized treatment plan that is focused on addressing the "why now" factors should address "any past issues . . . whether they're acute or chronic in nature"); Trial Tr. 1561:5-19 (Robinson-Beale) (testifying the "why now" is about taking "a more holistic approach to patient care" that is "about evaluating the entire patient" "and not just the symptoms"). This testimony was not credible for the reasons stated below.

93. First, the definition of "why now" in the Guidelines, discussed above, contradicts this testimony. That definition makes clear that the focus of "why now" is the member's recent severe changes and that it does not encompass factors related to the member's chronic condition that are not directly tied to those acute changes.

94. Second, UBH included other provisions in the Guidelines that referred explicitly to the types of factors UBH's witnesses testified (unconvincingly) were subsumed in the "why now" factors, treating them as distinct from "why now." For example, in the "best practices" section of the Guidelines for the years that included "why now," UBH specified that a treating provider should evaluate not only the "why now" factors, but also a host of other factors including the member's chief complaint, psychiatric and medical history, psychosocial and environmental problems (as distinct from acute changes in those issues), risk factors, and readiness for change. See Trial Ex. 4-0007 to -0009 (2014 Guidelines) (column headed "Evaluation & Treatment Planning" under "Clinical Best Practices"); see also Trial Ex. 5-0010 to -0011 (2015 Guidelines) ¶ 4.1.2; Trial Ex. 6-0011 to -0012 (2016 Guidelines) ¶ 4.1.2; Trial Ex. 7-0011 to -0012 (2016 Guidelines (June)) ¶ 4.1.2.

95. Third, although Dr. Bonfield testified that at some point he reviewed "the crisis intervention literature" from which the "why now" concept was borrowed, he was unable to remember any specific sources that addressed the concept, much less any that supported his explanation of its meaning. Trial Ex. 1659-0006 to -0007 (Bonfield Dep.) at 206:10-207:11. Further, Dr. Bonfield did not recall if he had reviewed any journals or academic publications addressing "why now." Nor did UBH offer into evidence any crisis intervention literature that supported the testimony of its witnesses with respect to the meaning of "why now."

96. In the 2014, 2015 and 2016 versions of the Guidelines, coverage upon admission required not only a finding that the patient could not be "treated in a less intensive setting," but also that the reason the patient required a higher level of care was the "why now" factors - i.e., the member's "acute changes." Trial Ex. 4-0007 (2014 Guidelines) (admission requires that "[t]he member's current condition cannot be safely, efficiently and effectively assessed and/or treated in a less intensive setting due to acute changes in the member's signs and symptoms and/or psychosocial and environmental factors (i.e., the 'why now' factors leading to admission)."); see also Trial Ex. 5-0008 (2015 Guidelines) ¶ 1.4 (same); Trial Ex. 6-0009 (2016 Guidelines) ¶ 1.4 (same); Trial Ex. 7-0009 (2016 Guidelines (June)) ¶ 1.4 (same). In the 2015 and 2016 versions, UBH added another admission requirement that "[a]ssessment and/or treatment of acute changes in the member's signs and symptoms and/or psychosocial and environmental factors (i.e., the 'why now' factors leading to admission) require the intensity of services provided in the proposed level of care." Trial Ex. 5-0008 (2015 Guidelines) ¶ 1.5; Trial Ex. 6-0009 (2016 Guidelines) ¶ 1.5; Trial Ex. 7-0009 (2016 Guidelines (June)) ¶ 1.5. These requirements are not consistent with generally accepted standards of care because they are overly focused on treatment of acute symptoms. In particular, under these provisions a member is denied coverage - even if the other criteria are met - if the reason the patient requires the prescribed level of care and "cannot" be treated in a lower level of care is anything other than "acute changes in the member's signs and symptoms and/or psychosocial and environmental factors." But as discussed above, neither "acute symptoms" nor "acute changes" should be a mandatory prerequisite for coverage of outpatient, intensive outpatient or residential treatment.

97. UBH removed references to the "why now" factors from the 2017 Guidelines, which were revised after the Court certified the classes in this case. See Trial Ex. 8 (2017 Guidelines). Nonetheless, the 2017 Guidelines Common Admission Criteria continued to require that "treatment of the factors leading to admission require the intensity of services provided in the proposed level of care," preserving the focus on crisis stabilization embodied in the "why now" concept even though that phrase was no longer used. See Trial Ex. 8-0006 to -0007 (2017 Guidelines) (bullet point beginning "The member's current condition").

iv. Coverage Ends When Acute Crisis Has Passed

98. The overemphasis on treatment of acute symptoms is found not only in the admission criteria of the challenged Guidelines but also in the continued service and discharge criteria that apply to all levels of care. Under these Guidelines, coverage of services at a given level of care may be terminated if the member either does not meet the continued service criteria or does meet the discharge criteria.

99. As an initial matter, in all challenged versions of the Guidelines members were required to show that they continued to meet the admission criteria for the applicable level of care in order to qualify for coverage of continued services at that level of care. See Trial Ex. 1-0078 (2011 Guidelines) Continued Service Criteria ¶ 1; Trial Ex. 2-0082 (2012 Guidelines) Continued Service Criteria ¶ 1; Trial Ex. 3-0089 (2013 Guidelines) Continued Service Criteria ¶ 1; Trial Ex. 4-0007 (2014 Guidelines) first bullet point in "Continued Service" column under "Level of Care Criteria"; Trial Ex. 5-0009 (2015 Guidelines) Continued Service Criteria ¶ 2.1; Trial Ex. 6-0010 (2016 Guidelines) Continued Service Criteria ¶ 2.1; Trial Ex. 7-0010 (2016 Guidelines (June)) Continued Service Criteria ¶ 2.1; Trial Ex. 8-0007 (2017 Guidelines) first bullet point in Common Continued Service Criteria for All Levels of Care. This means that just as a showing of acute symptoms is necessary for admission to a level of care, the patient must continue to suffer from those acute symptoms for coverage to continue at that level of care.

100. Other Common Criteria applicable to continued service and discharge also make clear that coverage will end when the member's symptoms are no longer acute. In the 2011 through 2013 versions of the Guidelines, this rule was reflected in the criteria stating that "[t]he goal of treatment is to improve the member's presenting symptoms to the point that treatment in the current level of care is no longer required" and further requiring that the member must be seeking "active treatment of a behavioral health condition." Trial Ex. 1-0006 (2011 Guidelines) Common Criteria ¶¶ 7-8 (emphasis added); Trial Ex. 2-0007 (2012 Guidelines) Common Criteria ¶¶ 7-8 (same); Trial Ex. 3-0008 (2013 Guidelines) Common Criteria ¶¶ 8-9 (same).

101. Similarly, in the 2014 through 2016 versions of the Guidelines - the versions that contain express references to "why now" - the continued service criteria required that in order for coverage to continue the patient must be receiving "active treatment," which required, inter alia, that the treatment plan be "focused on addressing the 'why now' factors." Trial Ex. 4-0007 (2014 Guidelines) first bullet point in "Continued Service" column under "Level of Care Criteria"; Trial Ex. 5-0009 (2015 Guidelines) Continued Service Criteria ¶ 2.1; Trial Ex. 6-0010 (2016 Guidelines) Continued Service Criteria ¶ 2.1; Trial Ex. 7-0010 (2016 Guidelines (June)) Continued Service Criteria ¶ 2.1. The discharge criteria for the Guidelines in these years further reinforces the rule that treatment services will not be covered once the immediate crisis has passed, providing that "[t]he continued stay criteria are no longer met" when "[t]he 'why now' factors which led to admission have been addressed to the extent that the member can be safely transitioned to a less intensive level of care, or no longer requires care." Trial Ex. 5-0009 (2015 Guidelines) ¶ 3.1; Trial Ex. 6-0010 (2016 Guidelines) ¶ 3.1; Trial Ex. 7-0010 (2016 Guidelines (June)) ¶ 3.1.

102. Even in the 2017 Guidelines, after the words "why now" had been removed, the Common Continued Service Criteria required "active treatment" and further explained that such treatment required, inter alia, that the treatment had to be "focused on the factors leading to admission." Trial Ex. 8-0007 (2017 Guidelines) first bullet point of Common Continued Service Criteria for All Levels of Care.

103. Nor does a denial of coverage at one level of care automatically lead to authorization of coverage at a lower level of care under the Guidelines. Rather, with respect to all challenged versions of the Guidelines, the member must qualify again under the admissions criteria for the lower level of care. See Trial Tr. 1104:14-1104:16, 1424:14-1424:19 (Martorana). Where coverage at a particular level of care has been denied or terminated on the ground that the member's acute symptoms have been alleviated, services even at a lower level of care may not be covered because of the focus on acute symptoms in the admissions criteria for all levels of care.

104. UBH witnesses Dr. Allchin testified that UBH's continued service and discharge criteria incorporate the admission criteria for the lower level of care - that is, that coverage at a higher level of care will not be discontinued unless the member satisfies the admissions criteria at a lower level of care. See Trial Tr. 1425:13-1426: 5 (Allchin) (testimony that "whenever we're making a discharge evaluation, it's what is the next level of care and how is that being utilized and why would that be safe and effective"). That testimony was not credible because it finds no support in the Guidelines. To the contrary, nothing in the Common Criteria provides that services at a particular level of care may not be terminated where there is no lower level of care that will be both safe and effective. Indeed, the discharge criteria for the 2014 through 2017 versions of the Guidelines require only that the member can be "safely" transitioned to a lower level of care. See Trial Ex. 4-0007 (2014 Guidelines) first bullet point in "Discharge" column under "Level of Care Criteria"; Trial Ex. 5-0009 (2015 Guidelines) Discharge Criteria ¶ 3.1.1; Trial Ex. 6-0010 (2016 Guidelines) Discharge Criteria ¶ 3.1.1; Trial Ex. 7-0010 (2016 Guidelines (June)) Discharge Criteria ¶ 3.1.1. Although the 2011-2013 versions of the Guidelines (which do not contain explicit "discharge criteria") include in the Common Criteria a requirement (found in paragraph 5) that a member's condition "cannot be effectively and safely treated in a lower level of care," see Trial Ex. 1-0005 (2011 Guidelines) Common Criteria ¶ 5 (emphasis added); Trial Ex. 2-0006 (2012 Guidelines) Common Criteria ¶ 5; Trial Ex. 3-0008 (2013 Guidelines) Common Criteria ¶ 5, that requirement must be read in conjunction with the other Common Criteria as all of them must be met to obtain coverage. Given that the Common Criteria in all of these versions of the Guidelines also contain a requirement that treatment must be "to improve the member's presenting symptoms to the point that treatment in the current level of care is no longer required" (discussed above), the Court does not find that paragraph 5 of these versions provides for the sort of feed-back loop described by Dr. Allchin. Consequently, under UBH's Guidelines patients may be denied coverage at a higher level of care because their acute symptoms have been addressed and it is safe to move them to a lower level of care even though treatment at a lower level of care may not be effective or even covered.

v. Other provisions of the Guidelines that address chronic conditions do not mitigate overemphasis on acuity

105. UBH points to the Clinical Best Practices section of the Common Criteria to show that UBH takes into account factors related to members' chronic conditions in making coverage determinations. The Clinical Best Practices of the Common Criteria instructs health care providers to collect information on a wide variety of topics, many of which relate to the member's underlying condition, in developing a treatment plan. See, e.g., Trial Ex. 5-0010 to -0011 (2015 Guidelines) Clinical Best Practices Section 4.1.2 (instructing provider to collect information on topics including "[t]he history of the presenting illness," "[t]he history of behavioral health services," the member's "medical history" and "developmental history," "current and historical life information" such as "[a]ge," "[g]ender, sexual orientation," "[c]ulture" and "[s]piritual beliefs," "[e]ducational history" and so on); Trial Ex. 6-0011 (2016 Guidelines) Clinical Best Practices Section 4.1.2 (same); Trial Ex. 7-0011 (2016 Guidelines (June)) Clinical Best Practices Section 4.1.2 (same); Trial Ex. 8-000 (2017 Guidelines) Clinical Best Practices second black bullet point (same). However, although the experts who testified at trial agreed that much of the information contained in this section is relevant to a member's chronic underlying condition, see e.g., Trial Tr. 189:16-190:14 (Fishman) (testifying that the topics listed in the Best Practices section are relevant to chronicity), the Guidelines often do not allow this information to be taken into account in the actual determination of coverage, which is based on consideration of the more limited factors related to the treatment of the member's acute symptoms.

106. UBH also cites the Guiding Principles in all of the challenged versions of the Guidelines, which use language that suggests a focus on the member's overall well-being rather than on simply managing crises. See Trial Ex. 1-0002 (2011 Guidelines) (treatment should support "broader recovery goals"); Trial Ex. 2-0002 (2012 Guidelines) (same); Trial Ex. 3-0003 (2013 Guidelines) (treatment should "support broader recovery/resiliency goals"); Trial Ex. 4-0003 (treatment should "support the member's broader recovery, resiliency and wellbeing goals"); Trial Ex. 5-0004 (2014 Guidelines) ("recovery, resiliency, and well-being are integral to" UBH's "core competencies"); Trial Ex. 6-0004 (2016 Guidelines) (same); Trial Ex. 7-0004 (2016 Guidelines (June)) (same); Trial Ex. 8-0002 (2017 Guidelines) (Guidelines "support members' recovery, resiliency, and wellbeing"). Once again, however, while these statements of principle are consistent with generally accepted standards of care, they are not incorporated into the specific Guidelines that establish rules for making coverage determinations. For the reasons discussed above, those Guidelines embody a much narrower focus aimed primarily at alleviating acute symptoms and managing crises while ignoring the question of whether treatment is likely to be effective in addressing the member's underlying condition or, in UBH's words, supporting the member's "broader recovery, resiliency and wellbeing."

b. Whether UBH Guidelines deviate from generally accepted standards of care by failing to address the effective treatment of co-occurring conditions

107. As discussed above, co-occurring conditions may require that a patient be placed at a higher level of care so that all of the patient's conditions can be effectively treated. In all of the challenged versions, UBH's Guidelines instruct practitioners to consider co-occurring physical and behavioral health conditions in developing a treatment plan. See Trial Ex. 1-0006 (2011 Guidelines) ("[t]he treatment plan . . . considers . . . [i]nterventions needed to address co-occurring behavioral health or medical conditions"); Trial Ex. 2-0008 (2012 Guidelines) (same); Trial Ex. 3-0008 to -0009 (2013 Guidelines) (same); Trial Ex. 4-0007 to -0008 (2014 Guidelines) (instructing providers to "collect[] information from the member and other sources, and complete[] an initial; evaluation of . . . risk stemming from co-occurring behavioral health or medical conditions"); Trial Ex. 5-0010 (2015 Guidelines) (instructing providers to "collect information from the member and other sources, and complete[] an initial evaluation of . . . co-occurring and behavioral health and physical conditions"); Trial Ex. 6-0011 (2016 Guidelines) (same); Trial Ex. 7-0011 (2016 Guidelines (June)) (same); Trial Ex. 8-0008 (2017 Guidelines) (same). The criteria in the Guidelines that actually govern coverage determinations with respect to the treatment of co-occurring conditions, however, are not consistent with generally accepted standards of care. Instead, in all relevant years the Guidelines instruct that determination of the appropriate level of care for the purposes of making coverage decisions should be based only on whether treatment of the current condition is likely to be effective at that level of care whereas treatment of co-occurring conditions need only be sufficient to "safely manage" them or to ensure that their treatment does not undermine treatment of the current condition.¹³ Conversely, the Guidelines omit any evaluation of whether a member's co-occurring conditions can be effectively treated in the requested level of care, or whether those conditions complicate or aggravate the member's situation such that an effective treatment plan requires a more intensive level of care than might otherwise be appropriate.

13 The specific provisions of the Guidelines that reflect an approach to co-occurring conditions that is inconsistent with generally accepted standards of care are identified by Plaintiffs in the Claims Chart with the short form "Co-occurring" in the "Why Flawed" column of the chart. Plaintiffs challenge over thirty specific Guidelines on this basis. Although the Court does not cite each one of these provisions here, it has reviewed all of them, as well as the parties' arguments and supporting testimony as they relate to the challenged provisions, and finds that each of the provisions listed on the Claims Chart that is challenged on this basis is inconsistent with generally accepted standards of care requiring effective treatment of both the patient's current condition (i.e., the patient's primary condition for which treatment is sought) and any co-occurring medical or behavioral health conditions.

108. UBH witnesses testified that the Guidelines are consistent with generally accepted standards of care with respect to treatment of co-occurring conditions because they instruct that a member should be placed at a level of care where the member's "current condition" can be treated both safely and effectively, and the term "current condition" encompasses co-occurring conditions. See Trial Tr. at 977:8-16 (Martorana); Trial Tr. at 1178:4-11 (Simpatico); Trial Tr. at 1387:4-14 (Allchin). That testimony was not credible because the plain language of the Guidelines supports a contrary conclusion; instead, the Court finds that these witnesses were simply offering post hoc rationalizations for Guidelines that transparently fail to provide for the effective treatment of co-occurring conditions. This is particularly obvious in the 2015 through 2017 versions of the Guidelines, which contain a list of requirements for admission in the Common Criteria that uses different words to describe the treatment of the member's "current condition" and the member's "co-occurring conditions." In particular, while the list requires that a member's "current condition can be safely, efficiently, and effectively assessed and/or treated in the proposed level of care," the very next requirement is that "[c]o-occurring behavioral health and medical conditions can be safely managed." See Trial Ex. 5-0008 (2015 Guidelines) ¶¶ 1.5, 1.6; Trial Ex. 6-0009 (2016 Guidelines) ¶¶ 1.5, 1.6; Trial Ex. 7-0009 (2016 Guidelines (June)) ¶¶ 1.5, 1.6; Trial Ex. 8-0007 (2017 Guidelines) fifth and sixth black bullet points under heading "Common Admission Criteria for All Levels of Care." In other words, UBH distinguished between treatment of the current condition (which must be both safe and effective) and treatment of co-occurring conditions (which need only be safe). When questioned by the Court, UBH's witnesses were unable to offer a convincing explanation for their interpretation and essentially conceded that the actual words that UBH used in the Guidelines did not support their testimony. Dr. Martorana, for example, could not explain why different words were used in the Common Criteria to describe the treatment of co-occurring conditions as compared to treatment of the current condition, responding that he did not "pick these words" and that when UBH approved them it did "not think it through in the way" the Court was thinking about the question. Trial Tr. 976:24-977:5 (Martorana). Likewise, Dr. Simpatico testified that he "[didn't] know" why different words were used and that he would "approve [an] edit" to the Common Criteria stating that co-occurring conditions must be safely and effectively treated. Trial Tr. 1179:23-1180:1 (Simpatico). Finally, Dr. Allchin testified that the separate reference to the safe treatment of co-occurring conditions was mere surplusage designed to emphasize that treatment of co-occurring conditions must be safe, even though the previous provision (under his interpretation of the Guidelines) already required that treatment of co-occurring conditions - which he testified are included in the term "current condition" - must be "safely, efficiently, and effectively assessed" at the proposed level of care. Dr. Allchin, like Dr. Martorana, noted in response to the Court's questions that he "didn't write the Guidelines" and conceded that it would be reasonable to interpret them as establishing separate standards for the treatment of the current condition and treatment of co-occurring conditions. Trial Tr. 1389:6-1390:14 (Allchin).

109. The Court's interpretation of the Guidelines with respect to the treatment of co-occurring conditions finds further support in a document drafted by Mr. Niewenhous, dated December 9, 2015, entitled "Guideline Touchbase Call." Trial Ex. 512. Under the general heading "Development of the [Utilization Management] Model" and the subheading "Current Model" there is a bullet point that states: "Is not organized to manage the needs of members with concurrent medical and behavioral health conditions." Trial Ex. 512-0007. Mr. Niewenhous testified that this statement reflected the fact that "in [UBH's] commercial business the services focus on the reasons why somebody came into treatment at that point." Trial Tr. 304:19-305:3 (Niewenhous).

c. Whether UBH Guidelines deviate from generally accepted standards of care by failing to err on the side of caution in favor of higher levels of care when there is ambiguity and pushing patients to lower levels of care where such a transition is safe even if the lower level of care is likely to be less effective

110. As discussed above, it is a generally accepted standard of care that patients should be placed at the least restrictive level of care that is both safe and effective and that practitioners should err on the side of caution when there is uncertainty by placing patients at the higher level of care. Further, the fact that a lower level of care may be less restrictive does not justify moving the patient to that level of care if it is also likely to be less effective in treating the patient's overall condition - including the underlying condition and any co-occurring conditions - even if movement to the lower level of care may be safe. UBH's Guidelines do not adhere to these principles. Instead, they actively seek to move patients to the least restrictive level of care at which they can be safely treated, even if a lower level of care may be less effective for that patient.¹⁴

14 With the exception of Common Criteria ¶ 5 in the 2011 and 2012 Guidelines and Common Criteria ¶ 6 in the 2013 Guidelines, the Court finds that all of the provisions in the Claims Chart that are identified as "Drive Towards Lower Levels of Care" are inconsistent with generally accepted standards of care due to this flaw.

111. In the 2011 through 2013 Guidelines, one of the requirements for admission to a given level of care was that treatment could not be safely and effectively provided in a less intensive level of care. See Trial Ex. 1-0005 (2011 Guidelines) Common Criteria ¶ 5 ("The member's current condition cannot be effectively and safely treated in a lower level of care even when the treatment plan is modified, attempts to enhance the member's motivation have been made, or referrals to community resources or peer supports have been made"); Trial Ex. 2-0006 (2012 Guidelines) ¶ 5 (same); Trial Ex. 3-0008 (2013 Guidelines) Common Criteria ¶ 6 (same). On its own, this requirement is not inconsistent with generally accepted standards of care (though other provisions of the Guidelines in these years improperly pushed members to lower levels of care, as discussed below). See Trial Tr. at 232:12-18 (Fishman) (testifying that he doesn't "particularly object" to the language in paragraph 6 of the Common Criteria of the 2013 Guidelines). Beginning in 2014, however, UBH added limiting language to this provision, allowing for continued coverage only when "[t]he member's current condition cannot be safely, efficiently and effectively assessed and/or treated in a less intensive setting due to acute changes in the member's signs and symptoms and/or psychosocial and environmental factors (i.e., the 'why now' factors leading to admission)." Trial Ex. 4-0007 (2014 Guidelines) Common Criteria and Best Practices for All Levels of Care: Admission, second black bullet point (emphasis added); see also Trial Ex. 5-0008 (2015 Guidelines) Common Criteria and Best Practices for All Levels of Care ¶ 1.4 (same); Trial Ex. 6-0009 (2016 Guidelines) ¶ 1.4 (same); Trial Ex. 7-0009 (2016 Guidelines (June)) Common Criteria and Best Practices for All Levels of Care ¶ 1.4 (same); Trial Ex. 8-0007 (2017 Guidelines) Common Criteria and Clinical Best Practices for All Levels of Care: Common Admission Criteria for All Levels of Care, first black bullet point ("The member's current condition cannot be safely, efficiently, and effectively assessed and/or treated in a less intensive level of care"). In doing so, the Guidelines drove members to lower levels of care even when treatment of the member's overall and/or co-occurring conditions would have been more effective at the higher level of care.

112. This focus on moving members to lower levels of care once their acute symptoms have been addressed can be seen in the Best Practices provisions of the Guidelines for all relevant years. See Trial Ex. 1-0006 (2011 Guidelines) Common Criteria ¶ 7 ("The goal of treatment is to improve the member's presenting symptoms to the point that treatment in the current level of care is no longer required."); Trial Ex. 2-0006 (2012 Guidelines) Common Criteria ¶ 7 (same); Trial Ex. 3-0008 (2013 Guidelines) Common Criteria ¶ 8 (same); Trial Ex. 4-0011 (2014 Guidelines) second bullet under "Clinical Best Practices: Evaluation & Treatment Planning" ("Treatment focuses on addressing the 'why now' factors to the point that the member's condition can be safely, efficiently, and effectively treated in a less intensive level of care . . ."); Trial Ex. 5-0011 (2015 Guidelines) Clinical Best Practices ¶ 4.1.7 (same); Trial Ex. 6-0013 (2016 Guidelines) ¶ 4.1.7 (same); Trial Ex. 7-0013 (2016 Guidelines (June)) ¶ 4.1.7 (same); Trial Ex. 8-0008 (2017 Guidelines) sixth black bullet (same, except "'why now' factors" is replaced with "factors precipitating admission").

113. Further, in each version of the Guidelines, there are other provisions that add requirements for continued service at a particular level of care that push patients to lower levels of care even though services at the lower level of care may not be as effective in treating the patient's condition. In particular, in all years there are provisions in the Guidelines that state that coverage at a particular level of care will be discontinued unless moving to a lower level of care is unsafe. See, e.g., Trial Ex. 1-0078 to -0079 (2011 Guidelines) Continued Service Criteria ¶¶ 2, 8 (for continued coverage, member must demonstrate, inter alia, "a significant likelihood of deterioration in functioning/relapse if transitioned to a less intensive level of care" and either measurable progress or "clear and compelling evidence that continued treatment at this level of care is required to prevent acute deterioration or exacerbation that would then require a higher level of care"); Trial Ex. 2-0082 (2012 Guidelines) Continued Service Criteria ¶ 6 (requiring "evidence that relapse or a significant deterioration in functioning would be imminent if the member was transitioned to a lower level of care . . ." ); Trial Ex. 3-0089 (2013 Guidelines) Continued Service Criteria ¶ 6 (same); Trial Ex. Ex. 4-0007 (2014 Guidelines) first sub-bullet under "Discharge" (providing that coverage ends when "[t]he 'why now' factors which led to admission have been addressed to the extent that the member can be safely transitioned to a less intensive level of care or no longer requires care"); Trial Ex. 5-0009 (2015 Guidelines) Discharge Criteria ¶ 3.1.1 (same); Trial Ex. 6-0010 (2016 Guidelines) Discharge Criteria ¶ 3.1.1 (same); Trial Ex. 7-0010 (2016 Guidelines (June)) ¶ 3.1.1 (same); Trial Ex. 8-0007 (2017 Guidelines) Common Discharge Criteria for All Levels of Care, first sub-bullet (same). These provisions fall short because they require discontinuation of coverage once it is safe to move to a lower level of care without regard to whether treatment at a lower level of care will be effective. As discussed above, Dr. Martorana testified that a patient would not be discharged under the Guidelines unless treatment at the lower level was both safe and effective. Trial Tr. 1064:3-1065:7 (Martorana). That testimony was not credible, however, because Dr. Martorana was unable to point to specific provisions in the Guidelines establishing the existence of such a requirement.

114. Not only do the Guidelines in all relevant years contain provisions that improperly instruct clinicians to consider only safety and not effectiveness in deciding whether to move a patient to a lower level of care; they also deviate from generally accepted standards of care by using language that strongly conveys to clinicians that they should err on the side of moving members to lower levels of care even where there is uncertainty about whether such a move is safe. For example, in the 2011 Guidelines, one of the requirements for receiving continued services at a given level of care was "[m]easurable and realistic progress has occurred or there is clear and compelling evidence that continued treatment at this level of care is required to prevent acute deterioration or exacerbation that would then require a higher level of care." Trial Ex. 1-0078 (2011 Guidelines) Continued Service Criteria ¶ 8 (emphasis added); see also Trial Ex. 1-0019 (2011 Guidelines) Intensive Outpatient Program: Mental Health Conditions ¶ 7 (continued coverage at this level requires that "[t]he provider and, whenever possible, the member collaborate to update the treatment plan every 3 to 5 treatment days in response to changes in the member's condition or provide compelling evidence that continued treatment in the current level of care is required to prevent acute deterioration or exacerbation of the member's current condition"); Trial Ex. 2-0020 (2012 Guidelines) Intensive Outpatient Program: Mental Health Conditions ¶ 7 (same); Trial Ex. 2-0049 (2012 Guidelines) Intensive Outpatient Program: Substance Use Disorders ¶ 8 (same); Trial Ex. 2-0063 (2012 Guidelines) Residential Rehabilitation: Substance Use Disorders ¶ 5 (same, except "provider" is replaced with "treating psychiatrist/addictionologist"). The parties' witnesses were in agreement that the "clear and compelling evidence" language used by UBH is not a medical term at all. Trial Tr. 137:1-9 (Martorana); Trial Tr. 1239:2-3 (Simpatico); Trial Tr. 1584:1-6 (Alam). Nor can there be any doubt that these words, based on their plain meaning, set a high threshold for continued services at a given level of care and precluded coverage if the clinician was merely uncertain as to whether treatment at a lower level of care would be safe (much less effective). Indeed, as noted above, Dr. Simpatico conceded, when pressed, that the "clear and compelling" standard used by UBH set an "impossible metric." Trial Tr. 1238:9-1240:24, 1242:8-9 (Simpatico).

115. Even when UBH did not use the words "compelling evidence" and "clear and compelling," the Guidelines for all years emphasized that when considering whether a lower level of care would be safe, clinicians should focus on "acute" symptoms and/or deterioration that was "significant," "severe" or "imminent," again deviating from generally accepted standards of care by discouraging them from taking into account the effective treatment of the patient's overall condition. See, e.g., Trial Ex. 2-00062 (2012 Guidelines) Residential Rehabilitation: Substance Use Disorders (providing coverage where "[t]here is a high risk of developing severe withdrawal symptoms which cannot be safely treated in a lower level of care"); Trial Ex. 3-0089 (2013 Guidelines) Continued Service Criteria ¶ 6 ("The member's current symptoms and/or history provide evidence that relapse or a significant deterioration in functioning would be imminent if the member were transitioned to a lower level of care . . . ."); Trial Ex. 5-0008 (2015 Guidelines) Common Criteria ¶ 1.5 ("Assessment and/or treatment of acute changes in the member's signs and symptoms and/or psychosocial and environmental factors (ie., the 'why now' factors leading to admission) require the intensity of services provided in the proposed level of care."); Trial Ex. 5-0081 (2015 Guidelines) Rehabilitation, Residential: Substance-Related Disorders, Admissions Criteria ¶ 1.3.2 (coverage where the "member is in immediate or imminent danger of relapse, and the history of treatment suggests that the structure and support provided in this level of care is needed to control the recurrence").

116. Starting in 2014, the drive to lower levels of care, even if they were likely to be less effective in treating a patient's overall condition, was also reflected in the way UBH defined the purpose of treatment, namely, as addressing the "why now" factors that precipitated admission. See, e.g., Trial Ex. 4-0027 (2014 Guidelines) Intensive Outpatient Program, Preamble ("The course of treatment in an Intensive Outpatient Program is focused on addressing the 'why now' factors that precipitated admission . . . ."); Trial Ex. 5-0030 (2015 Guidelines) Intensive Outpatient Program, Preamble (same); Trial Ex. 5-0033 (2015 Guidelines), Outpatient, Preamble ("The course of treatment in Outpatient is focused on addressing the 'why now' factors that precipitated admission . . . .").

d. Whether UBH Guidelines deviate from generally accepted standards of care by precluding coverage for treatment to maintain level of function

117. As discussed above, it is well-established that effective treatment of mental health and substance use disorders includes treatment aimed at preventing relapse or deterioration of the patient's condition and maintaining the patient's level of functioning. UBH Guidelines deviate from that standard by requiring a finding that services are expected to cause a patient to "improve" within a "reasonable time," and further restricting the concept of "improvement" to "reduction or control of the acute symptoms that necessitated treatment in a level of care." See Trial Ex. 1-0005 (2011 Guidelines) Common Criteria ¶ 6; Trial Ex. 2-0007 (2012 Guidelines) Common Criteria ¶ 6; Trial Ex. 3-0008 (2013 Guidelines) ¶ 7; Trial Ex. 4-0009 (2014 Guidelines) Common Criteria, Admission column, first black bullet; Trial Ex. 5-0008 to -0009 (2015 Guidelines) ¶ 1.8; Trial Ex. 6-0010 (2016 Guidelines) Common Criteria and Clinical Best Practices for All Levels of Care, Admission Criteria ¶ 1.8; Trial Ex. 7-0010 (2016 Guidelines (June)) Common Criteria and Clinical Best Practices for All Levels of Care, Admission Criteria ¶ 1.8; Trial Ex. 8-0007 (2017 Guidelines) Common Criteria and Clinical Best Practices for All Levels of Care, Common Admission Criteria for All Levels of Care, fifth black bullet point.¹⁵

15 The Court refers to these provisions collectively as the "Improvement Criteria."

118. Mr. Niewenhous testified that UBH's Improvement Criteria were borrowed from Chapter 6 of the CMS Manual. Trial Tr. 317:2-330:25 (Niewenhous). In all challenged versions, however, UBH modified the language used in the CMS Manual to provide for more limited coverage of services aimed at maintaining level of function. As discussed above, the CMS Manual provides for coverage of "[s]ervices [that are] . . . reasonably . . . expected to improve the patient's condition." Trial Ex. 656-00026. To meet this requirement, services must be "designed to reduce or control the patient's psychiatric symptoms so as to prevent relapse or hospitalization, and improve or maintain the patient's level of functioning." Id. (emphasis in original). The CMS Manual goes on to explain how this requirement can be satisfied, stating:

It is not necessary that a course of therapy have as its goal restoration of the patient to the level of functioning exhibited prior to the onset of the illness, although this may be appropriate for some patients. For many other psychiatric patients, particularly those with long-term, chronic conditions, control of symptoms and maintenance of a functional level to avoid further deterioration or hospitalization is an acceptable expectation of improvement. "Improvement" in this context is measured by comparing the effect of continuing treatment versus discontinuing it. Where there is a reasonable expectation that if treatment services were withdrawn the patient's condition would deteriorate, relapse further, or require hospitalization, this criterion is met.

Some patients may undergo a course of treatment that increases their level of functioning, but then reach a point where further significant increase is not expected. Such claims are not automatically considered noncovered because conditions have stabilized, or because treatment is now primarily for the purpose of maintaining present level of functioning. Rather, coverage depends on whether the criteria discussed above are met. Services are noncovered only where the evidence clearly establishes that the criteria are not met; for example, that stability can be maintained without further treatment or with less intensive treatment.

Trial Ex. 656-0026 to -0027 (CMS Manual). While borrowing bits and pieces of the standard set forth above, UBH made important modifications in its Guidelines that focused on acuity and precluded coverage of treatment services aimed at maintenance.

119. First, in contrast to the CMS Manual, which requires that there must be a reasonable expectation of improvement in the patient's "condition," the Improvement Criteria in the UBH Guidelines require that there must be a reasonable expectation of improvement in "the member's presenting problems" and UBH also added the modifying phrase "within a reasonable period of time." As discussed above, the term "presenting problems" refers to acute symptoms rather than the member's underlying - and often chronic - condition and the "reasonable period of time" requirement further reinforces the idea that improvement, under the UBH Guidelines, is about crisis stabilization rather than maintenance of function.

120. Second, in all relevant years UBH omitted the second sentence in the block quote above, which makes clear that under the CMS standard (and generally accepted standards of care), improvement is not limited to crisis stabilization but rather, includes services to maintain function.

121. The acute focus of UBH's Improvement Criteria was made even more explicit in 2012, when UBH added a sentence (to be designated as Paragraph 1.8.1 of the Common Criteria starting in 2015) explaining that "[i]mprovement of the member's condition is indicated by the reduction or control of the acute symptoms that necessitated treatment in a level of care." See, e.g., Trial Ex. 5-0009 (2015 Guidelines) ¶ 1.8.1; see also Trial Ex. 307-0002 (July 2010 Minutes of Coverage Determination Committee meeting, chaired by Mr. Niewenhous) (including the following "conclusion" with respect to discussion about "least intensive LOC" in the context of custodial care and inpatient services: "Add clarification that reasonable expectation of improvement in the patient's condition is improvement in the patient's acute condition"). That limitation, which does not correspond to any similar limitation in the CMS Manual, remained in the Guidelines for all subsequent years of the class period.

122. Further, while the Improvement Criteria provision for all relevant years contains a sentence that roughly corresponds to the sentence in the block quote above that begins "Improvement in this context" (to be designated 1.8.2 of the Common Criteria starting in 2015), calling for a weighing of "effectiveness of treatment" against likelihood of deterioration, beginning in 2014 UBH replaced the phrase "evidence that the member's condition will deteriorate" used in the CMS Manual with "evidence that the member's signs and symptoms will deteriorate," further emphasizing that the focus of the inquiry was to be on control of acute symptoms. See, e.g., Trial Ex. 5-0009 (2015 Guidelines) ¶ 1.8.2 ("Improvement in this context is measured by weighing the effectiveness of treatment against evidence that the member's signs and symptoms will deteriorate if treatment in the current level ends. Improvement must also be understood within the broader framework of the member's recovery, resiliency and wellbeing.").

123. The Court does not find credible the testimony offered by Dr. Martorana that the Improvement Criteria set forth two separate definitions of improvement in the sections that were eventually numbered ¶ 1.8.1 and ¶ 1.8.2. See Trial Tr. 987:7-20 (Martorana). Under this interpretation, only the first definition (found in ¶ 1.8.1) measures improvement with reference to acute symptoms whereas the second (alternative) definition (found in ¶ 1.8.2) defines improvement with reference to the likelihood of deterioration in the member's overall condition. In support of this interpretation, UBH points to the last sentence of ¶ 1.8.2, which instructs that "[i]mprovement must also be understood within the broader framework of the member's recovery and/or resiliency goals." Yet this interpretation is not consistent with the modifier "in this context" in ¶ 1.8.2. The most reasonable interpretation of this language is that it refers to the preceding sentence, found in ¶ 1.8.1 in the later versions of the Guidelines, which states that improvement is "indicated by the reduction or control of the acute symptoms that necessitated treatment in a level of care." (In contrast, the sentence that precedes the sentence beginning "[i]n this context" in the CMS Manual is the one that UBH chose to omit from its own provision, making clear that in the case of chronic conditions, improvement can mean "control of symptoms and maintenance of a functional level to avoid further deterioration or hospitalization.") That reading is also consistent with the Guidelines as whole, which repeatedly emphasize that treatment must be aimed at reduction or control of acute signs and symptoms, as discussed above. The last sentence of ¶ 1.8.2, instructing clinicians that improvement must "also be understood within the broader framework" of the member's recovery, resiliency and wellbeing merely pays lip service to generally accepted standards of care without offering any concrete guideline for incorporating them into the Improvement Criteria. The use of the word "also" in that sentence further makes clear that the sentence is merely an add-on that is not intended modify the requirements that precede it in the Improvement Criteria provision. In sum, the Court concludes that the Improvement Criteria provision in the UBH Guidelines for all versions of the Guidelines that are at issue in this case set forth a unified standard that is inconsistent with generally accepted standards of care.

124. Finally, the Court finds that specific additional criteria for residential treatment of mental health conditions in 2011 and intensive outpatient treatment in 2011 and 2012, as well as language in the preamble in the intensive outpatient treatment Guidelines for 2014-2107, cited by UBH in its post-trial brief, do not cure the deficiency discussed above with respect to those particular levels of care. See UBH Post-Trial Brief at 80-81. While the criteria and language cited by UBH use various formulations to refer to treatment to prevent deterioration, they do not override the excessively narrow requirements for continued coverage contained in the Improvement Criteria, which are applicable to all levels of care as part of the Common Criteria.

e. Whether UBH Guidelines deviate from generally accepted standards of care by precluding coverage based on lack of motivation

125. Plaintiffs contend UBH' Guidelines deviate from generally accepted standards of care by requiring discharge as soon as a patient becomes unwilling or unable to participate in treatment. The Court finds that the Guidelines for 2011 through 2013 are consistent with generally accepted standards of care with respect to consideration of a patient's motivation in determining the appropriate level of care but that the Guidelines for 2014 through 2017 deviate from those standards.

126. The parties' experts appear to be in agreement that for all levels of care that are at issue in this case, it is not appropriate under generally accepted standards of care to expect patients to be motivated to participate when they initially seek treatment; instead, there should be attempts to motivate a patient to participate in treatment before treatment at that level of care is discontinued. See Trial Tr. 116:1-3 (Fishman) ("To ask people to be motivated at the door is to ask people to be well before they get into treatment."); Trial Tr. 996:4-12 (Martorana) ("if the member is displaying an inability . . . to participate in treatment or [is] unwilling to participate in treatment, then we would expect the treatment plan to change" by "bring[ing] into play any number of interventions," including "motivational interventions").

127. While the CMS Manual suggests that a lack of motivation to participate may be a reason to preclude coverage of treatment at the Partial Hospitalization level, see Trial Ex. 656-0029 to -0034 (CMS Manual) § 70.3 (Partial Hospitalization Services), it also makes clear that the ability to participate in treatment is particularly critical in partial hospitalization programs because such programs are designed to provide short-term, acute care. That level of care is not at issue in this case and the evidence in the record does not support the conclusion that the standards that apply to motivation at that level also apply to the levels of care that are at issue here. Conversely, the evidence in the record does not support the conclusion that it is never appropriate to discontinue treatment at a given level of care based on a patient's lack of motivation to participate. Rather, to the extent that a patient should be placed at a level of care that is effective, generally accepted standards of care do not preclude discontinuation of treatment at a particular level of care if attempts to motivate a patient have failed and it is unlikely that treatment will be effective at that level due to lack of participation. Of course, it may be that effective treatment will require the patient to move to a higher level of care in the face of such a lack of motivation. See Trial Tr. 115:17-22 (Fishman) ("[S]ometimes it's lack of motivation or reluctance or even frank opposition to treatment that requires a certain intensity of treatment to get to persuade them to get with the program and to do better and to become cooperative and to become motivated.").

128. Beginning in 2014, UBH's common Discharge Criteria clearly violated the standards set forth above by providing that the "continued stay criteria are no longer met" when the "member is unwilling or unable to participate in treatment and involuntary treatment or guardianship is not being pursued." See Trial Ex. 4-0008 (2014 Guidelines) Common Criteria and Best Practices for All Levels of Care, second bullet under "Discharge"; Trial Ex. 5-0010 (2015 Guidelines) Common Criteria and Best Practices for All Levels of Care, Discharge Criteria, ¶ 3.1.5; Trial Ex. 6-0011 (2016 Guidelines) Common Criteria and Best Practices for All Levels of Care, Discharge Criteria, ¶ 3.1.5; Trial Ex. 7-0011 (2016 Guidelines (June)) Common Criteria and Best Practices for All Levels of Care, Discharge Criteria, ¶ 3.1.5; Trial Ex. 8-0007 (2017 Guidelines) Common Criteria and Best Practices for All Levels of Care, Common Discharge Criteria for All Levels of Care, fifth bullet point under only black bullet. Under these provisions, lack of motivation is a basis for discharge and discontinuation of coverage regardless of whether attempts to motivate the patient may eventually be effective or whether it is likely that treatment at this level of care is likely to be effective despite the patient's low motivation. Moreover, UBH's assertion that lack of motivation is a "single non-dispositive factor" in the Discharge Criteria for these years, see UBH post-trial brief at 83, is flatly contradicted by the plain language of these provisions.

129. The Guidelines for 2011, in contrast to the Guidelines discussed above, do not make lack of motivation an automatic reason for discontinuation of coverage at a given level of care. Instead, while making "active" participation a requirement for continued service, they leave room for coverage at a given level of care, even where the patient is not actively participating in treatment, for an "initial period of stabilization and/or motivational support." Trial Ex. 1-00078 (2011 Guidelines) Continued Service Criteria ¶ 4. Similarly, the Continued Service Criteria in the 2012 and 2013 Guidelines allow coverage to continue at a given level of care even where there is a "[l]ack of progress" if it is being addressed by "an intervention to engage the member in treatment." See Trial Ex. 2-0082 (2012 Guidelines) Continued Service Criteria ¶ 5; Trial Ex. 3-0089 (2013 Guidelines) Continued Service Criteria ¶ 5. The Court finds that these requirements are not inconsistent with the generally accepted standards of care discussed above.

f. Whether UBH Guidelines deviate from generally accepted standards of care by failing to address the unique needs of children and adolescents

130. One of the most troubling aspects of UBH's Guidelines is their failure to address in any meaningful way the different standards that apply to children and adolescents with respect to the treatment of mental health and substance use disorders. Throughout the Class Period, UBH failed to adopt separate level-of-care criteria tailored to the unique needs of children and adolescents. Nor do the Guidelines instruct decision-makers to apply the criteria contained in the Guidelines differently when the member is a child or adolescent.

131. While the clinical Best Practices provisions of the Guidelines contain "specific things that are very pertinent to children and adolescents," Trial Tr. 1376:19-22 (Allchin), these provisions are aimed at treatment providers rather than UBH staff who make coverage determinations. The criteria in the Guidelines that must be satisfied to obtain coverage, that is, the actual rules that govern coverage determinations, make no distinctions based on the unique needs of children and adolescents. In fact, as Dr. Triana testified, "UBH has never adopted any special set of rules for children and adolescents." Trial Tr. 1737:25-1738:2 (Triana); see also Trial Tr. 1673:11-14 (Alam) (conceding that UBH Guidelines "do not contain separate criteria for children and adolescents").

132. Generally accepted standards of care do not require that UBH create an entirely separate set of guidelines to address the needs of children and adolescents. They do, however, require that UBH's Guidelines instruct decision-makers to apply different standards when making coverage decisions involving children and adolescents, where applicable, including relaxing the criteria for admission and continued stay to take into account their stage of development and the slower pace at which children and adolescents generally respond to treatment. UBH has failed to meet this requirement for all relevant years.

g. Whether UBH deviates from generally accepted standards of care by using an overly broad definition of "custodial care" in its Guidelines, coupled with an overly narrow definition of "active" treatment and "improvement"

133. Under generally accepted standards of care, "custodial care" has a specific, narrow definition, which appears in the CMS Manual:

Custodial care serves to assist an individual in the activities of daily living, such as assistance in walking, getting in and out of bed, bathing, dressing, feeding, and using the toilet, preparation of special diets, and supervision of medication that usually can be self-administered. Custodial care essentially is personal care that does not require the continuing attention of trained medical or paramedical personnel. In determining whether a person is receiving custodial care, the intermediary or carrier considers the level of care and medical supervision required and furnished. It does not base the decision on diagnosis, type of condition, degree of functional limitation, or rehabilitation potential.

Trial Ex. 654-0029 (CMS Manual) Section 110, Custodial Care. This definition is found in Chapter 16 of the CMS Manual, listing General Exclusions from Coverage, and applies to services relating to mental health and substance use disorders, as well as medical services.

134. In all challenged versions of the Guidelines, UBH has broadened the concept of custodial care beyond the generally accepted definition of that term in several important ways, as set forth in more detail below.¹⁶ First, while generally accepted standards of care limit custodial services to those that "do[] not require the continuing attention of trained medical or paramedical personnel," the UBH Guidelines include a definition of "custodial care" under which even "skilled services" may be excluded from coverage on the basis that they are custodial. Second, UBH borrows the concept of "active care" - which is a separate requirement for Medicare coverage of inpatient hospitalization and partial hospitalization in the CMS Manual - and treats it as the flip side of custodial care, not only for coverage of inpatient services but also residential treatment. In doing so, it expands the concept beyond the definition used in the CMS Manual by including additional requirements that are focused on pushing patients to lower levels of care and terminating coverage as soon as the patient's acute symptoms have been addressed, regardless of whether treatment at a lower level of care is likely to be effective. Finally, UBH adds provisions related to improvement and maintenance of function that import into the concept of custodial care the shortcomings discussed above relating to these concepts.

16 The Court has reviewed all of the Guidelines that Plaintiffs challenge under the "Custodial" category in the Claims Chart and finds that each of them is deficient because of the overly narrow approach to custodial care adopted by UBH, as set forth below.

135. For all relevant years, UBH had a CDG addressing the exclusion of coverage for custodial care provided in both acute inpatient units and residential treatment centers. See Trial Ex. 10-0003 (Custodial Care and Inpatient Services CDG, effective August 1, 2010-December 1, 2011 ("2011 Custodial Care CDG")) (UBH "maintains that treatment of a behavioral health condition in an acute inpatient unit or [residential treatment center] is not for the purposes of providing custodial care, but for the active treatment of a behavioral health condition."); Trial Ex. 47-0003 (Custodial Care and Inpatient Services CDG, effective December 1, 2011-January 1, 2013 ("2012 Custodial Care CDG")) (same); Trial Ex. 84-0003 (Custodial Care and Inpatient & Residential Services, effective January 1, 2013-February 1, 2014 ("2013 Custodial Care CDG")) (addressing custodial care "in a psychiatric inpatient or residential setting"); Trial Ex. 108-0002 (Custodial Care and Inpatient & Residential Services, effective February 1, 2014-March 1, 2015) ("2014 Custodial Care CDG")) (same); Trial Ex. 148-0003 (Custodial Care and Inpatient & Residential Services, effective March 1, 2015-April 1, 2016 ("2015 Custodial Care CDG")) (addressing custodial care in "psychiatric inpatient and residential treatment settings"); Trial Ex. 195-0003 (Custodial Care and Inpatient & Residential Services, effective April 1, 2016-March 1, 2017 ("2016 Custodial Care CDG")) (same); Trial Ex. 221-0003 (Custodial Care (Inpatient & Residential Services), effective May 1, 2017 ("2017 Custodial Care CDG")) (same). These CDGs (hereinafter, the "Custodial Care CDGs") explain that an acute inpatient setting "provides 24-hour nursing care and monitoring, assessment and diagnostic services, treatment and specialty medical consultation services" whereas a residential treatment center "provides overnight mental health services to members who do not require 24-hour nursing care and monitoring offered in an acute inpatient setting but who do require 24-hour structure." See, e.g., Trial Ex. 47-0004 (2012 Custodial Care CDG).

136. While there are minor differences between the Custodial Care CDGs, all of them focus on three interrelated concepts: "custodial care," "active treatment" and "improvement." See, e.g., Trial Ex. 84-0003 (2013 Custodial Care CDG) (containing key points in these three categories, with each of the three terms placed in bold). In all but the earliest version of the Custodial Care CDG, UBH defines "custodial care" as including clinical services under some circumstances.¹⁷ The 2012, 2013 and 2014 Custodial Care CDG's do this using the following language, contained in the "key points" section of these CDGs:

17 The earliest versions of the CDGs and LOCGs that are at issue in this case exclude coverage of "custodial care" but do not provide an express definition of that term. See Trial Ex. 10-0003 (2011 Custodial Care CDG); Trial Ex. 1-0057 (2011 Guidelines). The first time the definition appeared in UBH's Guidelines was in the custodial care CDG that came into effect on December 1, 2011. See Trial Ex. 47-0003 (Custodial Care and Inpatient Services CDG, effective December 1, 2011- January 1, 2013). As these early Guidelines do not define "custodial care" they are not flawed in this particular respect, though they deviate from generally accepted standards of care related to custodial care for all of the other reasons discussed in this section.

Custodial Care in a psychiatric inpatient or residential setting is the implementation of clinical or non-clinical services that do not seek to cure, or which are provided during periods when the member's behavioral health condition is not changing, or does not require trained clinical personnel to safely deliver services . . . .

Trial Ex. 47-0003 (2012 Custodial Care CDG) second black bullet point (emphasis added); Trial Ex. 84-0003 (2013 Custodial Care CDG) first black bullet point (same); Trial Ex. 108-0003 (2014 Custodial Care CDG) first black bullet point (same). Similarly, the 2015, 2016 and 2017 Custodial Care Guidelines, while revising the definition of custodial care, continued to deem "custodial" any services "for the primary purpose of . . . maintaining a level of function (even if the specific services are considered to be skilled services)." Trial Ex. 148-0003 (2015 Custodial Care CDG) second sub-bullet under "custodial care" back bullet in key points; Trial Ex. 195-0003 (2016 Custodial Care CDG) (same); Trial Ex. 221-0003 (2017 Custodial Care CDG) (same). These definitions are inconsistent with generally accepted standards of care, as reflected in the definition of custodial care in the CMS Manual (quoted above), which limits custodial care to unskilled services. See Trial Ex. 654-0029 (CMS Manual, Chapter 16) Section 110, Custodial Care; see also Trial Tr. 120:12-121:13 (Fishman) (explaining that defining custodial as any "services that do not require continued administration by trained medical personnel" is not consistent with generally accepted standards because lower levels of residential treatment do not require medical personnel).

137. The definitions of custodial care quoted above also deviate from generally accepted standards of care because they deem services - even skilled clinical services - to be "custodial" whenever the patient's condition is stable, that is, "during periods when the member's behavioral health condition is not changing," such as when the patient's "presenting signs and symptoms . . . have been stabilized, resolved, or a baseline level of functioning has been achieved" or when the patient "is not responding to treatment or otherwise not improving." Trial Ex. 84-0003 (2013 Custodial Care CDG); see also Trial Ex. 148-0003 (2016 Custodial Care CDG) (skilled services deemed custodial if they are for the purpose of "maintaining a level of function"). This is inconsistent with the generally accepted standard, discussed above, that calls for treatment to be provided when needed to maintain a patient's level of function or to prevent deterioration. See, e.g., Trial Tr. 558:3-7 (Plakun) ("determining whether a service is custodial" should not "depend on the degree of functional limitation or rehabilitation potential"); Trial Ex. 654-0029 (CMS Manual) Chapter 16, Section 110, Custodial Care (providing that the determination of whether services are custodial should not be based on "rehabilitation potential"). Likewise, the fact that a patient is not "responding" to treatment is not a generally accepted ground for withholding services, at least where a patient still has the potential to respond to treatment. Trial Tr. 114:15-22, 117:6-17 (Fishman). UBH's interpretation of the term "custodial" is unreasonable in light of what is generally accepted.

138. The shortcomings of the definition of "custodial care" in the Custodial Care CDGs are compounded and reinforced by the provisions of the Custodial Care CDGs that address "active care," which is described as the opposite of "custodial care." See Trial Ex. 10-0003 (2011 Custodial Care CDG) Key Points (UBH "maintains that treatment of a behavioral health condition in an acute inpatient unit or [residential treatment center] is not for the purpose of providing custodial care, but is for the active treatment of a behavioral health condition."); Trial Ex. 47-0003 (2012 Custodial Care CDG) Key Points (same); Trial Ex. 84-0003 (2013 Custodial Care CDG) Key Points (care is custodial when "[t]he intensity of active treatment . . . is no longer required"); Trial Ex. 108-0003 (2014 Custodial Care CDG) Key Points (same); Trial Ex. 148-0003 (2015 Custodial Care CDG) Key Points ("services provided in psychiatric and residential treatment settings that are not active and are solely for the purposes of Custodial Care as defined below are excluded"); Trial Ex. 195-0003 (2016 Custodial Care CDG) Key Points (same); Trial Ex. 221-0003 (2017 Custodial Care Guideline) Key Points (same).

139. The concept of "active treatment" is addressed in Chapter 2 of the CMS Manual, governing coverage of Inpatient Psychiatric Hospital Services. See Trial Ex. 655-0007 (CMS Manual) Chapter Two, Section 30.2.2.1 (entitled "Principles for Evaluating a Period of Active Treatment"). Section 30.2.2.1 provides that services meet the "active treatment" requirement if they are: 1) "Provided under an individualized treatment or diagnostic plan;" 2) "Reasonably expected to improve the patient's condition or for the purpose of diagnosis; and" 3) "Supervised and evaluated by a physician." The parties are in agreement that this definition of "active treatment" reflects generally accepted standards. See Plaintiff's Post-Trial Brief at 56; UBH Proposed Findings of Fact and Conclusions of Law at 92, ¶¶ 545-546; see also Trial Ex. 10-0008 (2010 Custodial Care CDG) (citing CMS Chapters 2 and 16).

140. In the 2011 through 2015 Custodial Care CDGs, UBH included the three requirements of Section 30.2.2.1 quoted above in its definition of "active treatment" but also added the following two requirements:

• Unable to be provided in a less restrictive setting; and

• Focused on interventions that are based on generally accepted standard medical practice and are known to address the critical presenting problem(s), psychosocial issues and stabilize the patient's condition to the extent that they can be safely treated in a lower level of care.

See Trial Ex. 10-0003 (2011 Custodial Care CDG) Key Points; Trial Ex. 47-0003 (2012 Custodial Care CDG) Key Points (same); Trial Ex. 84-0003 (2013 Custodial Care CDG) Key Points (same); Trial Ex. 108-0003 (2014 Custodial Care CDG) Key Points (same); Trial Ex. 148-0003 (2015 Custodial Care CDG) Key Points (same). (Hereinafter, the Court refers to the first of these requirements as the "less restrictive setting" requirement and the second as the "critical presenting problems" requirement.)

141. In 2016, UBH revised the Custodial Care CDG to include the "strict definition" of active treatment (that is, only the three requirements contained in the CMS definition of "active treatment") but did not eliminate the "less restrictive setting" requirement; rather, it moved this additional requirement to two different bullet points in the Custodial Care CDG. See Trial Ex. 195-0003 (2016 Custodial Care CDG) third bullet ("Active Treatment in an inpatient or residential treatment setting is a clinical process involving the 24-hour care of members that . . . cannot be managed in a less restrictive setting.") and fifth bullet ("Optum maintains that inpatient or residential treatment . . . cannot be provided in a less restrictive setting."); see also Trial Ex. 221-0003 (2017 Custodial Care CDG) (same language as 2016 Custodial Care CDG); Trial Ex. 537 (March 2016 email exchange between Martorana, Niewenhous and Urban regarding adoption of "strict definition" of "active treatment" from CMS Manual, in which Urban told Martorana that "[a]lthough (unable to be managed in a lower level of care) is not included in CMS' definition of 'active treatment,' . . . [w]e can still cite this in the custodial care CDG and I can make sure it remains").¹⁸

18 Although the 2016 and 2017 Custodial Care CDGs no longer include the "critical presenting problems requirement," the focus on treatment of acute symptoms reflected in that requirement is preserved in the definition of "improvement" in these CDGs, as discussed further below.

142. UBH's Custodial Care CDGs, therefore, provide that treatment is not "active" (and is thus custodial) whenever it is "[]able to be provided in a less restrictive setting." But the mere fact that it is possible to provide services in a less restrictive setting does not mean that such a setting is the appropriate one for a particular patient. Rather, as discussed above, generally accepted standards call for a multi-dimensional assessment of the patient to determine where treatment will be both safe and most effective, erring on the side of caution. It is unreasonable to conclude that services are not "active" just because they could, in theory, be provided somewhere else.

143. Similarly, it is also not consistent with generally accepted standards to limit "active treatment" to interventions that "address the critical presenting problem(s), psychosocial issues" and "stabilize the patient's condition to the extent that they can be safely treated in a lower level of care," - another requirement added to the definition of "active treatment" by UBH in its Custodial Care CDGs for 2011 through 2015. This is just another way of pushing patients to lower levels of care where it is safe to do so even though treatment at the lower level may not be as effective, an approach that is inconsistent with generally accepted standards of care for the reasons discussed above.

144. The Court does not find persuasive UBH's reliance on the CMS Manual in support of the "less restrictive setting" requirement. See UBH Post-Trial Brief at 88 (citing Trial Ex. 656-0025 to -0026). UBH points to the last sentence of the "Reasonable Expectation of Improvement" section of the coverage criteria in Section 70.1 of Chapter 6 of the CMS Manual, which addresses outpatient hospital psychiatric services. That sentence states that "[s]ervices are noncovered only where the evidence clearly establishes that the criteria are not met; for example, that stability can be maintained without further treatment or with less intensive treatment." Trial Ex. 656-0026 to -0027 (CMS Manual) (emphasis added). As the Court has already found, UBH has borrowed words from Section 70.1 but has not preserved the broader meaning of that section, which makes clear that coverage of outpatient hospitalization services should be continued at that level even if a patient's condition has stabilized so long as there is a "reasonable expectation that if treatment services were withdrawn the patient's condition would deteriorate, relapse further, or require hospitalization."

145. UBH further narrowed its definition of what constitutes custodial care by incorporating its overly-restrictive definition of "improvement" in the Custodial Care CDGs as a counterpart to the "active care" requirement. In particular, UBH defines "improvement" in the Custodial Care CDGs (as in the Guidelines generally) as "reduction or control of the acute symptoms that necessitated hospitalization or residential treatment." Trial Ex. 10-0003 (2011 Custodial Care CDG) Key Points, sixth black bullet point; Trial Ex. 47-0003 (2012 Custodial Care CDG) Key Points, seventh black bullet point; Trial Ex. 84-0003 (2013 Custodial Care CDG) Key Points, seventh black bullet point; Trial Ex. 108-0003 (2014 Custodial Care CDG) Key Points, seventh black bullet point; Trial Ex. 148-0003 (2015 Custodial Care CDG) Key Points, fourth black bullet point; Trial Ex. 195-0003 (2016 Custodial Care CDG) Key Points, fourth black bullet point; Trial Ex. 221 (2017 Custodial Care CDG) Coverage Rationale. Thus, for UBH, only those services that are expected to reduce or control acute symptoms count as "active treatment" sufficient to avoid a finding that the services are custodial (and consequently excluded from coverage). The application of this narrow definition of "improvement" results in an overemphasis on acuity in the Custodial Care CDGs and precludes coverage of services needed to maintain function or prevent deterioration. In sum, UBH's concepts of custodial care, active treatment, and improvement are intertwined in the Custodial Care CDGs to preclude coverage of services that would not be considered custodial under generally accepted standards of care.

146. Similar flaws related to the custodial care exclusion are also found in UBH's LOCGs governing coverage of residential treatment. In the 2011 Guidelines, for example, one of the requirements for coverage of residential treatment is that treatment must not be for "the purpose of providing custodial care, but is for the active treatment of a mental health condition." Trial Ex. 1-00027 (2011 Guidelines) Residential Treatment Center: Mental Health Conditions ¶ 5(a); Trial Ex. 1-0057 (2011 Guidelines) Residential Rehabilitation: Substance Use Disorders ¶ 5(a). These LOCGs go on to set forth the flawed definition of active treatment discussed above, modifying the CMS definition by adding the "less restrictive setting" and "critical presenting problems" requirements. Id. The same is true for the LOCGs for residential treatment in the 2012 and 2013 Guidelines, which prohibit coverage of services that are "custodial" rather than "active" and use the same five-part definition of "active treatment." See Trial Ex. 2-0059 to -0060 (2012 Guidelines) Residential Detoxification: Substance Use Disorders ¶ 6(b)(iv)-(v); Trial Ex. 2-0064 (2012 Guidelines) Residential Rehabilitation: Substance Use Disorders ¶ 5(b)(iv)-(v); Trial Ex. 3-0034 to -0035 (2013 Guidelines) Mental Health Conditions: Residential Treatment Center ¶ 6(d)-(e); Trial Ex. 3-0069 (2013 Guidelines) Substance Use Disorders: Residential Rehabilitation ¶ 6(d)-(e).

147. Beginning in 2012, the residential treatment LOCGs also added definitions of "custodial care" that mirrored the definitions used in the Custodial Care CDGs for the same years. See, e.g., Trial Ex. 2-0029 (2012 Guidelines) Residential Treatment Center: Mental Health Conditions ¶ 5(a); Trial Ex. 2-0059 (2012 Guidelines) Residential Detoxification: Substance Use Disorders ¶ 6(a); Trial Ex. 2-0063 (2012 Guidelines) Residential Rehabilitation: Substance Use Disorders ¶ 5(a); Trial Ex. 3-0034 (2013 Guidelines) Mental Health Conditions: Residential Treatment Center ¶ 5; Trial Ex. 3-0068 to -0069 (2013 Guidelines) Substance Use Disorders: Residential Rehabilitation ¶ 5; Trial Ex. 4-0043 (2014 Guidelines) Residential Treatment Center: Mental Health Conditions, "Continued Service" and "Discharge" columns; Trial Ex. 4-0077 (2014 Guidelines) Residential Rehabilitation: Substance Use Disorders, "Continued Service" and "Discharge" columns; Trial Ex. 5-0038 to -0039 (2015 Guidelines) Residential Treatment: Mental Health Conditions ¶ 2.2; Trial Ex. 5-0082 (2015 Guidelines) Rehabilitation, Residential: Substance-Related Disorders ¶ 2.2.2; Trial Ex. 6-0043 to -0044 (2016 Guidelines) Residential Treatment Center: Mental Health Conditions ¶ 2.2; Trial Ex. 6-0091 (2016 Guidelines) Rehabilitation, Residential: Substance-Related Disorders ¶ 2.2; Trial Ex. 7-0044 (2016 Guidelines adopted June 2016) Residential Treatment Center: Mental Health Conditions ¶ 2.2; Trial Ex. 7-0092 (2016 Guidelines adopted June 2016) Rehabilitation, Residential: Substance-Related Disorders ¶ 2.2; Trial Ex. 8-0018 (2017 Guidelines) second sub-bullet under second black bullet under "Residential Treatment Center Continued Service Criteria"; Trial Ex. 8-0036 (2017 Guidelines) second black bullet under "Rehabilitation, Residential Continued Service Criteria." These definitions are flawed for the reasons discussed above.

148. At trial, a UBH witness testified that the definition of custodial care used in the Guidelines was based on custodial care exclusions in class members' plans. See Trial Tr. 899:10-20 (Dehlin) (testifying that the definition of "custodial care" in the Custodial Care CDG was "verbatim, but if not verbatim, incredibly close to the language from the definition of the most common definition of custodial care from Exhibit 1654 [UBH's summary exhibit regarding custodial care definitions of Plans of Claim Sample]"). A review of the custodial care definitions in the plans of the Claim Sample reflects that 25 members of the Claim Sample had benefit plans that used the three-part definition of "custodial care" that UBH began using in 2015 in its Custodial Care CDGs and in its residential treatment LOCGs. The court has reviewed the Plans of these Claim Sample members, however, and does not find that any of them include the overly restrictive definitions of "active treatment" and "improvement" that significantly expand the concept of custodial care in UBH's CDGs and LOCGs. See Trial Ex. 1654-0001 to -0005. Further, even if the inclusion of this language in some class members' Plans might limit coverage for those class members to exclude even some services that are consistent with generally accepted standards of care - a question the Court does not address here- it does not justify the application of standards that do not reflect generally accepted standards of care to class members whose plans do not contain this language.

h. Whether UBH Guidelines deviate from generally accepted standards of care by imposing mandatory prerequisites rather than a multidimensional approach

149. As discussed above, decisions about the level of care at which a patient should receive treatment must be multi-dimensional, taking into account a wide variety of information about the patient and allowing clinicians to weigh the dimensions against one another. Plaintiffs contend the very structure of UBH's Guidelines, containing a list of Common Criteria that are mandatory, is inconsistent with the holistic approach that is required under generally accepted standards of care. While a list of required criteria does not necessarily deviate from generally accepted standards of care, UBH's Guidelines are nonetheless flawed to the extent that they instruct clinicians to collect a wide array of information under their Best Practices provisions but do not allow for adequate consideration of this information in the rules and requirements that govern coverage determinations. This flaw results in many of the deviations from generally accepted standards of care that are discussed above.

5. Whether UBH Guidelines are Consistent With ASAM

150. As discussed above, ASAM is a recognized source of generally accepted standards of care and reflects, inter alia, the following generally accepted standards of care: 1) treatment should not be limited to crisis stabilization and the treatment of acute presenting symptoms but rather, should be aimed at providing effective treatment of the patient's overall condition, including chronic and co-occurring medical and behavioral health conditions; 2) patients should treated at the least restrictive level of care that is both safe and effective and should be moved to a lower level of care only where the lower level is likely to be safe and just as effective as treatment at the higher level of care in addressing a patient's overall and co-occurring conditions; 3) clinicians should err on the side of caution by placing the patient in a higher level of care when there is ambiguity or uncertainty as to the appropriate level of care; 4) treatment services should be provided to maintain functioning or prevent deterioration; 5) determination of the appropriate level of care must take into account the unique needs of children and adolescents; and 6) placement determinations should be based on a holistic, multidimensional approach that allows a wide variety of factors to be taken into account and weighed against one another. UBH's Guidelines deviate from these standards in a multitude of ways, as set forth above. This has been the case throughout the Class Period, including before and after the 2013 publication of the ASAM third edition. Indeed, in an internal UBH email exchange in 2012 with the subject line "Use of ASAM criteria poll," one of UBH's regional medical directors opined that the ASAM Criteria "usually will result in more authorization as they are more subjective and broader than our LOCG/CDGs." See Trial Ex. 348-0001 to -0002 (email dated July 18, 2012 from Dr. Michael Haberman to Dr. Lorenzo Triana).

151. Many of the deviations from ASAM Criteria in the UBH Guidelines are reflected in the edits proposed by Mr. Jerry Shulman, a co-editor of ASAM (along with Dr. Fishman), who was hired by UBH in 2013 to compare UBH's Guidelines with ASAM Criteria and propose revisions to bring them into line with ASAM. See Trial Ex. 402-006 (describing services to be performed by Mr. Shulman); Trial Tr. 1626:10-20 (Alam). Mr. Shulman principally critiqued the March 2013 CDG for Treatment of Substance Use Disorders, Trial Ex. 412-0015 to -0045, and the substance use disorder sections of the 2012 Level of Care Guidelines, Trial Ex. 412-0046 to -0098. Among other things, he found that UBH's continued service criteria were more restrictive than ASAM Criteria. See Trial Ex. 412-0058 (proposing two additional alternative grounds for coverage in the continued service criteria of the 2013 Guidelines, which would have significantly expanded coverage under the Guidelines: "5. The member has not yet resolved the problems that justified admission but is working on them and making progress. OR 6. The member has resolved the problems that justified admission but new problems have surfaced which can only be dealt with safely at the current level of service."); see also Trial Ex. Ex. 412-0036 (proposing same additional grounds for coverage to the Treatment of Substance Use Disorder CDG). He also identified ways UBH's Guidelines failed to appropriately consider co-occurring conditions, Trial Ex. 412-0093, and explained that coverage criteria that are limited to "stabilization" create a "likelihood of the member experiencing further problems," facing "additional risk," and needing "additional treatment." Trial Ex. 412-0053.

152. The most glaring inconsistency between UBH Guidelines and the ASAM Criteria relates to coverage of residential treatment at levels 3.1, 3.3 and 3.5. UBH Guidelines simply do not provide criteria for coverage of services at these levels. Thus, when Mr. Shulman began his comparison of the UBH Guidelines with ASAM Criteria, he could not find criteria that applied to levels of residential treatment below level 3.7 and called Dr. Alam, at UBH, to ask where they were. Trial Tr. 1639:16-19 (Alam). Dr. Alam (incorrectly) told Mr. Shulman that UBH does not cover those levels of care. Id.; see also Trial Ex. 412-13 ("Optum/ASAM Crosswalk"¹⁹ created by Mr. Shulman as part of his report reflecting his understanding that levels 3.1, 3.3 and 3.5 are "not an Optum member benefit"). Consistent with this understanding, one of Mr. Shulman's proposed edits of UBH's Guidelines was to make clear in the title of two of the residential treatment guidelines for substance use disorders that they related specifically to services at ASAM level 3.7. See Trial Ex. 412-0089 and -0093.

19 At trial, Dr. Alam explained that Mr. Shulman's "crosswalk" was "essentially a lineup of [UBH's] criteria next to the ASAM [C]riteria to allow sort of a back and forth, some matching . . . ." Trial Tr. 1627:13-16 (Alam); see also Trial Tr. 1638:24-1639:2 (Alam) (one of the purposes of the crosswalk was to "make it easier to see the differences and similarities between ASAM and [UBH's] guidelines").

153. In contrast to what UBH told Mr. Shulman, it has represented to Connecticut regulators that ASAM levels 3.1, 3.3 and 3.5 are, in fact, covered by its Guidelines, namely, in the admission criteria for Residential Rehabilitation. See Trial Ex. 402-0005 (2013 Crosswalk); Trial Ex. 506-0005 (2015 Crosswalk). Yet Dr. Fishman offered extensive testimony at trial that the Residential Rehabilitation criteria in the UBH Guidelines are not consistent with ASAM when applied to levels of residential treatment that are lower than level 3.7, and that testimony is largely uncontroverted. See Trial Tr. 124:6-126:16 (Fishman) (testifying that sections 1.3 and 1.4 of Rehabilitation, Residential LOCG in 2015 Guidelines overemphasize acuity and imminent danger for lower levels of residential treatment even if these criteria are appropriate at the 3.7 level); Trial Tr. 143:11-144:23 (Fishman) (testifying that requirement in Residential Rehabilitation: Substance Use Disorders LOCG in the 2011 Guidelines that treating psychiatrist or addictionologist update the treatment plan every five days is appropriate at level 3.7, where care is medically monitored, but is not appropriate at lower levels of residential care); Trial Tr. 223:12-16 (Fishman) (testifying as to paragraph 3 of the same guideline, requiring that "psychiatric evaluations and consultations are available 24 hours a day," that this requirement would be appropriate for the medically monitored level 3.7, but "would not be appropriate for 3.5, 3.3, and 3.1."). To the extent that UBH witness Dr. Robinson-Beale testified generally that it was her understanding that UBH's Guidelines covered these levels of care even though they are not specifically "called out," that testimony was not credible. See Trial Ex. 1657 (Robinson-Beale Depo. excerpt) at 189:3-190:2. Similarly, Dr. Alam's testimony that a member would not be denied coverage of residential treatment at the 3.5 level "merely because there's not a separate and distinct level in the UBH Guidelines" and that he was unaware of any denials of coverage at that level based on the fact that "there was not a specific and distinct 3.5 level of care guideline" is not sufficient to establish that UBH Guidelines do, in fact, provide coverage criteria appropriate for that level of care.

154. In its post-trial brief, UBH essentially conceded that its Guidelines do not provide for coverage of residential treatment at ASAM levels 3.1, 3.3 or 3.5. See UBH Post-Trial Brief at 91-93. Instead, UBH offers a hodge-podge of excuses for this omission, none of which is convincing. First, as to level 3.1, UBH argues it is not required to have criteria for this level of residential treatment because services at this level are evaluated under separate guidelines that have not been challenged in this case for determining coverage of sober living arrangements. See UBH Post-Trial Brief at 92 (citing Trial Tr. 406:12-25 (Niewenhous); 1024:9-12, 1137:25-11338:3 (Martorana)). Yet ASAM expressly states that level 3.1 "is not intended to describe or include sober houses, boarding houses, or group homes where treatment services are not provided." Trial Ex. 662-245 (ASAM Criteria). Residential treatment at level 3.1, in contrast, must provide at least five hours a week of treatment. Trial Ex. 662-244. As the UBH Guidelines Plaintiffs challenge in this case purport to provide coverage criteria for residential treatment of substance use disorders, the testimony that UBH applies different guidelines for determining coverage of sober living arrangements is beside the point. Moreover, UBH has not demonstrated that the Guidelines it says it applies to sober living homes are appropriate with respect to level 3.1 residential treatment, that is, programs that include both a residence component and a clinical component.

155. Nor is Dr. Martorana's vague testimony that only some benefit plans cover sober living homes, see UBH Post-Trial Brief at 92 (citing Trial Tr. 1024:4-8), sufficient to establish that any class member's Plan excludes treatment at level 3.1. As is apparent from witness testimony and ASAM itself, terms such as "sober living home" and "halfway house" are used colloquially to refer both to sober living programs that include a clinical component (making them residential treatment), and those that do not. UBH has not offered evidence that Dr. Martorana's testimony even relates to sober living programs that meet the definition of residential treatment; nor has it pointed to any plans that exclude coverage at that level.

156. UBH's responses as to levels 3.3 and 3.5 are similarly unconvincing. As to level 3.3, UBH states (in a footnote) that "Plaintiffs offered no evidence that any class members sought coverage for treatment at a Level 3.3 facility, and Plaintiffs' experts did not opine as to this level of care." UBH Post-Trial Brief at 92-93 n. 66; see also Trial Ex. Ex. 651-0002 ("Historically, we haven't covered the lower levels of residential. However, if we move to using ASAM, I don't see how we are able to deny the lower levels if the member has a residential benefit."); Tr. 1809:14-25 (Niewenhous). As to level 3.5, UBH suggests that some of the class members' Plans do not cover treatment at this level of care because they require that treatment be provided by medical professionals, effectively limiting coverage of residential treatment to services provided at the 3.7 level. While these arguments might be relevant to remedies they do not change the Court's findings with respect to liability.

I. Whether UBH Guidelines Complied With State Laws

For the reasons set forth below, the Court finds that during the class period UBH violated the laws of Illinois, Connecticut, Rhode Island, and Texas by failing to apply criteria that were in compliance with the laws of those states for making coverage determinations relating to substance use disorders treatment.

1. Illinois

157. Effective August 18, 2011, Illinois law mandated that all "[m]edical necessity determinations for substance use disorders shall be made in accordance with appropriate patient placement criteria established by the American Society of Addiction Medicine." 215 Ill. Comp. Stat. § 5/370c(b)(3) (effective Aug. 18, 2011). In 2015, Illinois amended the provision that contained this requirement by adding the following sentence: "No additional criteria may be used to make medical necessity determinations for substance use disorders." 215 Ill. Comp. Stat. 5/370c(b)(3) (effective September 9, 2015). The Court finds that the plain language of the original provision required that UBH use the ASAM Criteria rather than its own Guidelines to make coverage determinations for treatment of substance abuse disorders. See F.D.I.C. v. Meyer, 510 U.S. 471, 476 (1994) (statutory terms given their "ordinary or natural meaning").

158. To the extent that the Illinois statute as originally enacted was in any way unclear, the circumstances surrounding the amendment of the provision in 2015 support the conclusion that the amendment was meant to clarify rather than modify the original provision. See Block v. Office of Ill. Sec'y of State, 988 N.E.2d 718, 721-722 (App. Ct. Ill. 2013) ("While a material change in a statute made by an amendatory act is presumed to change the original statute, that presumption is rebutted where the circumstances surrounding the enactment of the amendment indicate that the legislature intended to interpret, rather than change, the original act."). It is particularly significant that the 2015 amendment merely adds a sentence; it does not change the language used with respect to the actual requirement that ASAM Criteria must be used. Further, the only mention of the provision in the transcript of the Illinois Senate session in which the bill was addressed was a passing reference describing it as a provision that "specifies" that ASAM Criteria are to be used for making medical necessity determinations. See Ill. Senate Tr., 2015 Reg. Sess. No. 52 at p. 11. This reference was made in the context of a discussion of an unrelated issue and there was no suggestion that the ASAM requirement was new. Nor has the Court found anything in the legislative history suggesting that the amendment was intended to modify the original ASAM requirement.

159. UBH's carefully phrased argument that the 2015 amendment was "[c]onsistent with the . . . recommendation" of a working group assembled by the Illinois Department of Insurance, misleadingly implying that the amendment was in response to a recommendation by that working group, is not persuasive. See UBH Post-Trial Brief at 106. The working group's report, dated January 2017, states that "[s]ometimes providers use ASAM guidelines while payers use other guidelines" and goes on to state, "[i]t will be beneficial to continue to work to find a consistent set of criteria so needed services can be provided." See Ill. Dept. of Ins. Working Group re Treatment and Coverage of Substance Abuse Disorders and Mental Illness Annual Report, January 2017 at 2. To the extent that UBH characterizes this statement as a "recommendation" it certainly cannot be a recommendation that the language cited above ("No additional criteria may be used to make medical necessity determinations for substance use disorders.") be added to the 2011 Illinois law requiring use of ASAM Criteria as that provision had already been amended at the time the working group was formed and the amended provision had been in effect for more than a year when the report was published. Further, the statement UBH quotes does not address what the law actually required, either in 2011 or in 2015, contrary to UBH's representation in its brief. See UBH Post-Trial Brief at 106 (pointing to the above statement in support of the assertion that "[t]he group noted that the original version of the law did not require the use of ASAM" when the report makes no mention of the original version, or indeed any version, of the law). Moreover, to the extent the statement in this 2017 report might be read to imply that payers are permitted to use their own guidelines (or at least, were permitted to do so as of January 2017), UBH's argument proves too much as it is undisputed that at least as of 2015, insurers were only permitted to use ASAM Criteria and not their own guidelines to make medical necessity determinations.

160. Likewise, the Governor's initial veto of the 2015 statute that amended the ASAM requirement (among many other things) does not support the conclusion that the 2015 amendment stating that "[n]o additional criteria may be used to make medical necessity determinations for substance use disorders" imposed a new requirement on health benefit plans under Illinois law.²⁰ To the contrary, the Governor's letter supports the opposite conclusion. The letter addresses the bill in which the amendment was contained, the Heroin Crisis Act, and begins with a description of "important changes" in the law relating to the opioid crisis, including a requirement that private insurers cover "at least one opioid antagonist, as well as acute treatment and clinical stabilization services." Notably absent is any discussion of changes governing the standards that must be used to make coverage determinations with respect to such benefits. The Governor does express concern "about a very costly mandate on the State's Medicaid providers" in the bill, namely, a requirement that he said "mandates that fee-for-service and medical assistance Medicaid programs cover all forms of medication assisted treatment of alcohol or opioid dependence, and . . . removes utilization controls and prior authorization requirements." Again, however, the Governor does not address the use of ASAM Criteria or whether the original language of the 2011 law, requiring that coverage determinations be "in accordance with" ASAM Criteria, precluded the use of other criteria by private insurers. Finally, the Governor ends the letter by stating that he would support the bill if certain specific changes were made. None of those changes relates to the amendment at issue here.

20 The Court found Governor Rauner's letter at http://www.ilga.gov/legislation/fulltext.asp?DocName=09900HB0001gms&GA=99&LegID=83490&SessionId=88&SpecSess=0&DocTypeId=HB&DocNum=0001&GAID=13&Session=. The URL provided by UBH in its brief resulted in an error message.

161. UBH did not start using the ASAM Criteria for Illinois substance use disorder claims until January 2016. Trial Tr. 951:16-20 (Martorana); see also Trial Ex. 273-0002 (September 2015 Guideline Applicability Tool); Trial Ex. 274-0002 (January 2016 Guideline Applicability Tool). Because it was required to use ASAM Criteria to make medical necessity determinations for claims governed by Illinois law as of August 18, 2011, its use of its own Guidelines as to those claims violated Illinois law. Further, even if the original 2011 version of the law permitted UBH to use its own Guidelines so long as they were consistent with the ASAM Criteria, at least until the law was amended in 2015, UBH's Guidelines did not comply with Illinois law because they were not consistent with ASAM Criteria, as discussed above.

2. Connecticut

162. Connecticut has required insurers to use the ASAM Criteria, or a set of criteria that UBH "demonstrates to the Insurance Department is consistent with" the ASAM Criteria, since October 1, 2013. Conn. Gen. Stat. § 38a-591c(a)(3) (2017); 2013 Conn. Legis. Serv. 13-3. UBH concedes that it has never used the ASAM Criteria in Connecticut. To establish compliance with Connecticut law, UBH points to the "crosswalks" it submitted to Connecticut regulators in 2013 and 2015. See Trial Exs. 402 & 506. The Court finds that UBH has failed to comply with Connecticut law throughout the class period because its Guidelines are not "consistent with" the ASAM Criteria for the reasons discussed above. Moreover, in the "crosswalks" UBH submitted to Connecticut regulators in 2013 and 2015, it materially mischaracterized the UBH Guidelines by stating that "the criteria from all 3 ASAM levels [3.1, 3.3 and 3.5] are included in the admission criteria for Reside[n]tial Rehabilitation." Trial Exs. 402-0005 & 506-0005. At the time these statements were made to Connecticut regulators, UBH knew them to be false, as reflected in the Shulman Report, discussed above.

3. Rhode Island

163. Since July 10, 2015, Rhode Island has required that payors such as UBH "rely upon the criteria of the American Society of Addiction Medicine when developing coverage for levels of care for substance-use disorder treatment." 27 R.I. Gen. Laws § 27-38.2-1(g) (2015); 2015 R.I. Pub. Laws 15-236 (15-H 5837A). While the law does not preclude UBH from developing its own guidelines to make coverage determinations, it requires that those guidelines must be consistent with ASAM Criteria; merely listing ASAM as a reference or borrowing a definition is not sufficient to meet this requirement. For the reasons discussed above, UBH's Guidelines are not consistent with ASAM Criteria and therefore UBH has failed to comply with Rhode Island law.

164. The requirement under Rhode Island law that coverage guidelines must "rely on" ASAM Criteria is not limited to in-network providers. UBH's reliance on 27 R.I. Gen Laws § 27-38.2-4 in support of that proposition is misplaced.²¹ That subsection of Rhode Island's Parity Act provides:

21 In its post-trial brief and proposed findings of fact and conclusions of law, UBH cites 27 R.I. Gen. Laws § 27-38.4, which does not exist. Based on context, the Court concludes that UBH intended to cite 27 R.I. Gen. Laws § 27-38.2-4. UBH did not provide the specific language of the provision, which it mischaracterizes in its brief by ignoring the language after the semicolon.

The health care benefits outlined in this chapter apply only to services delivered within the health insurer's provider network; provided, that all health insurers shall be required to provide coverage for those benefits mandated by this chapter outside of the health insurer's provider network where it can be established that the required services are not available from a provider in the health insurer's network.

27 R.I. Gen. Laws § 27-38.2-4. The health care benefits referenced in this subsection are mental health and substance-use disorder coverage. This section merely provides that a health benefit plan is not required to cover mental health or substance use disorder treatment by an outside provider unless that treatment is not available through its own providers. It does not limit the requirement that health benefit plans "rely on" ASAM Criteria in making coverage determinations related to substance use disorder treatment.

4. Texas

165. For the entire class period, insurance companies were required to apply criteria issued by the Texas Department of Insurance ("TDI Criteria" or "TCADA Guidelines") in making medical necessity determinations with respect to claims for substance use disorder treatment when an individual's plan was governed by Texas law and treatment was sought from a provider or facility in Texas. 28 Tex. Admin. Code § 3.8011 (1991).

166. Throughout the class period, UBH's Guideline Applicability Tool, used by UBH's Care Advocates and Peer Reviewers to determine which guidelines to apply to a member's benefit request, consistently shows that Texas guidelines were to be applied to coverage requests for substance use disorder treatment in Texas under plans governed by Texas law. Trial Tr. 389:5-20 and 394:2-7 (Niewenhous); Trial Tr. 430:12-431:3 (Niewenhous); Trial Ex. 450 (May 2014 Guideline Applicability Tool); Trial Ex. 268 (October 2014 Guideline Applicability Tool); Trial Ex. 270 (January 2015 Guideline Applicability Tool); Trial Ex. 271 (March 2015 Guideline Applicability Tool); Trial Ex. 272 (May 2015 Guideline Applicability Tool); Trial Ex. 273 (September 2015 Guideline Applicability Tool); Trial Ex. 274 (January 2016 Guideline Applicability Tool); Trial Ex. 275 (May 2016 Guideline Applicability Tool); Trial Ex. 276 (July 2016); Trial Ex. 277 (August 2016 Guideline Applicability Tool); Trial Ex. 278 (January 2017 Guideline Applicability). In addition, three UBH witnesses testified that throughout the class period UBH has used TDI Criteria to make coverage determinations with respect to claims governed by Texas law. See Trial Tr. 951:21-952:2 (Martorana); Trial Tr. 1377:21-1378:1 (Allchin); Trial Tr. 430:5-431:3 (Niewenhous).

167. On the other hand, Plaintiffs introduced into evidence an email from Mr. Niewenhous dated May 26, 2015 in which he stated, in part, as follows:

Question from Houston about whether the TCADA guidelines apply or the CDGs Former required by State reg, latter thought to apply because of Parity. Houston has been using the CDGs.

Trial Ex. 493 (emphasis added). Mr. Niewenhouse testified that "Houston" referred to UBH's Care Advocacy Center in Houston. Trial Tr. 395:4-6 (Niewenhous). This evidence supports the conclusion that UBH violated Texas law at some point during the class period by applying its own CDG's rather than the TDI Criteria, though the email does not establish how long the violation lasted or which CDGs UBH applied. The class list provided by UBH also reflects that UBH applied its own Guidelines, rather than TDI Criteria, to claims for residential treatment of substance use disorder in Texas during the class period. See Trial Ex. 255 (class list). In particular, it shows that the claims of numerous class members for such services were denied on the basis of UBH's Guidelines rather than the TDI Criteria. Id.; see also Trial Ex. 896 (Stipulation re Trial Ex. 255) ¶¶ 2-3. In the face of this evidence, the Court finds that the testimony offered Mr. Niewenhous and Drs. Martorana and Allchin that UBH applied the TDI Criteria is not credible, at least to the extent they implied that UBH consistently applied TDI Criteria to claims for benefits that were governed by Texas law during the class period. Plaintiffs have demonstrated by a preponderance of the evidence that during the class period UBH violated Texas law by applying its own Guidelines to claims for benefits that should have been decided under TDI Criteria.

J. UBH's Guideline Development Process

168. Throughout the class period, UBH reviewed its Guidelines annually, revising them in response to input from clinicians and professional organizations. Trial Tr. 1688:5-15 (Triana); Trial Ex. 1658 (Beaty Depo.) at 84:07-18.

169. With respect to the LOCGs, the revision process was conducted in several stages. The first stage of the process typically started in June of the preceding year, when UBH distributed its Guidelines to internal and external behavioral health professionals and professional societies to solicit suggestions for revisions. Trial Tr. 937:18-938:2 (Martorana). Individuals with degrees in social work - Jerry Niewenhous and Loretta Urban in 2011-2016, and Erik Rockswold in 2017 - reviewed the feedback and created working drafts of the revised LOCGs. Trial Tr. 1696:8-14 (Triana); Trial Ex. 1661 (Urban Depo.) at 15:19-24, 42:5-43:2; Trial Ex. 904 (Rockswold Depo.) at 18:20-24. Throughout the year, these individuals also tracked various sources on UBH's "hierarchy of evidence," including government sources, guidelines and consensus statements issued by professional associations, graded reviews of the literature, and peer-reviewed research. Trial Ex. 1661-0002 to -0003 (Urban Depo.) at 38:15-41:19. The working drafts incorporated revisions based on this research as well. Id. The working drafts of the revised LOCGs, along with a grid summarizing the feedback discussed above, were then submitted to the LOCG Work Group for consideration. Trial Tr. 1696:3-1697:5 (Triana). The LOCG Work Group over the relevant years included UBH's Chief Medical Officers, Dr. Robinson-Beale, Dr. Bonfield, and Dr. Bruce Bobbitt; senior clinicians, Dr. Lorenzo Triana, Dr. Pete Brock, and Dr. Andrew Martorana; as well as Jerry Niewenhous. Id. Finally, after the LOCG Work Group had considered the working drafts and the feedback that had been received, and had made any changes that it found appropriate, it submitted the proposed revised LOCGs to the BPAC (from 2011-2016) or the Utilization Management Committee (2017) for review and approval. Trial Tr. 1707:21-1708:10 (Triana).

170. A similar process was followed with respect to the CDGs. First, Loretta Urban developed working drafts of the CDGs, which were then circulated to clinicians, both within UBH and outside it, for feedback. Trial Ex. 1661-001 (Urban Depo.) at 12:08-14. The revised CDG drafts and feedback were then passed on to the Coverage Determination Committee ("CDC") for consideration and further revision. Trial Tr. 337:14-23, 414:9-12 (Niewenhous). Finally, the CDC submitted the proposed revised CDGs to BPAC (2011-2016) or the Utilization Management Committee (2017) for review and approval. Trial Tr. 337:21-23, 338:5-6 (Niewenhous); Trial Tr. 697:25-698:16 (Triana); Trial Tr. 1821:3-011 (Niewenhous); Trial Ex. 1657 (Robinson-Beale Depo.) at 201:11-202:6.

171. The internal UBH clinicians who provided feedback on the working drafts of the revised Guidelines were typically medical doctors or health care professionals with at least a masters-level education. Trial Tr. 1689:22-1690:1 (Triana). The external clinicians who provided feedback included clinicians who were selected from UBH's provider network. Trial Tr. 1691:10-16, 1692:8-11 (Triana). They also included clinicians who were members of UBH's Behavioral Specialty Advisory Committee ("BSAC"), an internal committee that includes representatives of various specialty associations, including the American Psychiatric Association, the American Psychological Association, the National Association of Social Workers, the National Association of Psychiatric Health Systems and ASAM. Trial Tr. 1692:2-5 (Triana). Clinicians were asked questions such as whether UBH's Guidelines were "easy to use" or if there were "criteria which should be added or deleted." Trial Ex. 1114 (January 20, 2012 letter requesting feedback from UBH provider regarding LOCGs). They were not specifically asked if the Guidelines were consistent with generally accepted standards of care. Id. They were paid $150 for submitting written comments on the Guidelines. Id.

172. The National Committee for Quality Assurance ("NCQA") and the Utilization Review Accreditation Commission ("URAC") are the two leading organizations that accredit utilization management processes for major health plans and for freestanding health utilization management organizations. Trial Tr. 1766:6-8 (Goddard). To earn accreditation, both URAC and NCQA require that a health insurer's guideline development process includes consultation with actively practicing providers with relevant medical knowledge, consideration of evidence-based treatment, an annual review process (and update of guidelines if appropriate) and approval by a clinical director. Trial Tr. 1768:19-1769:4, 1770:6-1771:7 (Goddard); see also Trial Exs. 1012-0154 (URAC Health Utilization Management, Version 7.0, HUM 1 Review Criteria) & 1011-0007 (NCQA UM 2 Clinical Criteria for UM Decisions). These accreditations are based on the process that an organization uses in developing its guidelines, not the substantive content of those guidelines. Trial Tr. 1784:13-21 (Goddard).

173. UBH employee John Beaty was responsible for UBH's accreditation with NCQA and URAC during the class period. Trial Ex. 1658 (Beaty Depo.) at 12:04-08. He confirmed that UBH received accreditation for the LOCGs from both NCQA and URAC during the entire class period. Trial Ex. 1658 (Beaty Depo.) at 83:22-85:07, 87:3-88.

174. While the process UBH uses to develop its Guidelines satisfies all of the requirements for accreditation, the Court concludes that it is also fundamentally flawed because it is tainted by UBH's financial interests.

175. UBH earns money by charging fees for its services as the behavioral health administrator for various health plans. As discussed above, UBH administers two types of plans: fully insured and self-funded plans. Based on the stipulated list of coverage denials from UBH's records that meet class definitions, see Trial Ex. 255 (Class List), and the parties' stipulation regarding per-member-per-month rates for class members, see Trial Ex. 711, the Court finds that more class members' Plans were self-funded than were fully insured (39,257 as compared to 27,734) but that the [ REDACTED] Further, on a per-member-per-month basis, UBH made between [REDACTED] See Trial Ex. 711-0014.

176. For fully insured plans, UBH bears the risk that the benefit expense for the services it approves will be more than it projected when it fixed its premium, which reduces UBH's profit. Trial Tr. 840:6-14 (Dehlin). Likewise, although UBH does not bear the same risk with respect to self-funded plans, it has an incentive to keep benefit costs down for customers who purchase such plans. Trial Tr. 803:12-21 (Triana) ("[Y]ou have to also approach the health plans and the customers that you have plans with, and you have to address and let them know that you may be changing a guideline. And one of the things that they may be asking is what are, potentially, the cost implications to that. So it's important to be able to answer those kinds of questions, because they are the customers.").

177. Because of the financial incentives to keep benefit expense down, UBH regularly prepares detailed financial forecasts that include projections of expected benefit expense and benefit expense targets it wants to achieve. Trial Ex. 1660 (Brock Dep.) at 216:1-219:9. UBH also tracks its performance in relation to those benefit expense forecasts and targets, noting monthly trends and taking action to address benefit expenses that exceed its projections. See, e.g., Trial Ex. 745; Trial Ex. 783-0009.

178. One area in which UBH maintains detailed "utilization" data relates to average length of stay ("ALOS") for which UBH approves coverage. As ALOS increases, the cost of associated benefits increases, either for UBH or, in the case of self-funded plans, its customers. Trial Tr. 761:12-21 (Triana). Therefore, UBH carefully monitors "utilization" data with regard to ALOS for particular levels of care. See, e.g., Trial Ex. 783-0031 to -0038; Trial Ex. 745. UBH also sets ALOS targets for each level of care, and tracks them every month. See e.g., Trial Tr. 759:15-760:17 (Triana); Trial Ex. 720-0015.

179. UBH's Guidelines have a direct impact on benefit expense and therefore are closely tied to the financial incentives discussed above. While the incentives related to fully insured and self-funded plans are not identical, with respect to both types of plan UBH has a financial interest in keeping benefit expense down. Further, even if the financial incentives may be stronger as to one or the other category of plan, any resulting shortcomings in its Guideline development process taints its decision-making as to both categories of plan because UBH maintains a uniform set of Guidelines for fully insured and self-funded plans.

180. The Court finds that the financial incentives discussed above have, in fact, infected the Guideline development process. In particular, instead of insulating its Guideline developers from these financial pressures, UBH has placed representatives of its Finance and Affordability Departments in key roles in the Guidelines development process throughout the class period. For example, Peter Brock, the head of UBH's Affordability Department, and Fred Motz, from UBH's Finance Department, were both members of the BPAC, the committee responsible for approving the LOCGs and CDGs. Trial Tr. 703:3-16 (Triana); Trial Ex. 482-0002 (BPAC minutes dated January 20, 2015 showing members). Another Affordability representative, Michael Powell, was also on the BPAC through at least 2015. See, e.g., Trial Ex. 482-0002. Brock's successor as head of the Affordability Department, Nisha Patterson, became a member of the Utilization Management Committee ("UMC"), which replaced the BPAC in 2016. Trial Ex. 552-0002.

181. In addition to including representatives of Finance and Affordability on the committees with ultimate authority to approve the Guidelines, UBH provided detailed relevant financial briefings to other members of those committees who were not members of Finance or Affordability. For example, Dr. Triana, Chair of the BPAC and then the UMC, and committee member Dr. Martorana, were both briefed in detail on a monthly basis on UBH's financial metrics and its performance related to benefit expense targets. See, e.g., Trial Ex. 783 (example of monthly business review sent to Drs. Triana and Martorana); Trial Ex. 720 (ALOS report sent to Dr. Triana); Trial Ex. 745 (email discussion of "June close" sent to Dr. Triana); Trial Tr. 755:5-17 (Triana); Tr. 1122:20-1123:9 (Martorana). These reports were also sent to committee members from Finance and Affordability. See, e.g., Trial Ex. 783 (December 2014 email also sent to, inter alia, BPAC members Margaret Brennecke, Peter Brock, James Davis, and Nisha Patterson); Trial Ex. 482 (January 2015 minutes showing BPAC members); Trial Ex. 745 (July 2013 email also sent to, inter alia, BPAC members Michael Powell, Peter Brock, Brett Hart, James Davis, and future BPAC members Patterson and Motz); Trial Ex. 368 (March 2013 minutes showing BPAC members).

182. UBH witnesses testified that financial considerations were rarely discussed at BPAC meetings and that the Finance Department Representative Fred Motz rarely attended or spoke, see Trial Tr. 786:3-788:9 (Triana). That evidence does not show that financial considerations did not play a role in the development of UBH's Guidelines, however, given that the committee members were intimately familiarity with the financial implications of their decisions in creating and revising the Guidelines. In any event, the record is replete with evidence that UBH's Guidelines were viewed as an important tool for meeting utilization management targets, "mitigating" the impact of the 2008 Parity Act, and keeping "benex" down. See, e.g., Trial Ex.768-0009 (2014 presentation describing "[c]ontinued use of concurrent review to ensure appropriate utilization" as the "Mitigation Strateg[y]" for Parity's "[r]emoval of day and visit limits on IP, Intermediate and OP"); Tr. 307:4-24 (Niewenhous).

183. First, the very fact that the Guidelines were riddled with requirements that provided for narrower coverage than is consistent with generally accepted standards of care gives rise to a strong inference that UBH's financial interests interfered with the Guideline development process. The Court finds, for example, that the "why now" factors introduced by Dr. Bonfield were aimed more at keeping "benex" down than they were at ensuring that members received coverage of services that was consistent with generally accepted standards of care. And it is consistent with that goal that the LOCG Working Group did not change the "why now" provisions of the Guidelines in response to criticism from a BSAC member who represented AACAP. See Trial Ex. 516-0007 (feedback from BSAC representative Dr. Alan Axelson, stating that "[w]hile I understand the focus on 'why now' interventions, I am very concerned that the overemphasis of this type of treatment has contributed to an ineffective and inefficient overall treatment system"); Trial Tr. 743:24-747:6 (Dr. Triana conceding that although the Level of Care Working Group discussed Dr. Axelson's comments, UBH "did not make a change in the 'why now' language for 2016"). Similarly, the overemphasis on moving members to a "less restrictive setting" in the Guidelines, discussed above, was influenced, at least in part, by cost considerations. See Trial Ex. 437-0001 ("I think that taking out the restrictive setting language [from the medical necessity definition] is okay because it is likely that the least costly service would also be offered in a less restrictive environment.").

184. Other decisions by UBH during the class period further support the conclusion that its financial self-interest was a critical consideration in deciding what criteria would be used to make coverage decisions and when Guidelines would be revised.

185. One example that illustrates the heavy emphasis that UBH places on financial considerations when deciding whether Guidelines should be changed is UBH's decision in late 2016 not to amend its Guidelines with respect to Applied Behavioral Analysis ("ABA"), a treatment for autism spectrum disorder. Although the Utilization Management Committee had approved a Guideline broadening coverage of that treatment, UBH's CEO, Martha Temple, overruled the recommendation, cautioning UBH staff, "[w]e need to be more mindful of the business implications of guideline change recommendations." Trial Tr. 904-0004 to -0005, -0008 (Rockswold) (testimony that UMC approved the Guideline change but that the CEO vetoed the change); Trial Ex. 812-0001 (12/16/16 email from Martha Temple to UBH staff, including UMC members Nisha Patterson and Adam Easterday).

186. Another example is UBH's decision making with respect to coverage of Transcranial Magnetic Stimulation ("TMS"), a treatment for major depressive disorder. For many years, UBH denied coverage of this treatment on the basis that it was experimental, but by around 2013 or 2014, the FDA had approved TMS and outside reviewers were sometimes overruling UBH's denials of coverage. Trial Tr. 766:9-767:11 (Triana). Because UBH was "getting pressure" to cover TMS, see Trial Ex. 758, it commissioned an internal study of the "financial impact" of covering TMS claims where medically necessary. Trial Tr. 767: 4-11. Fred Motz, of UBH's Finance Department, conducted the analysis and UBH "estimated [a] cost per patient" in the range of $9,000 to $14,000. Trial Ex. 749-0004. The Clinical Policy Committee, with the benefit of this analysis, then considered a number of factors, including the impact to benefit expense and the "return on investment" ("ROI") if it revised the Guidelines to cover TMS treatment in accordance with national standards. Id. The Committee recommended that UBH approve TMS claims only for members of self-funded plans, that is, plans where UBH was not responsible for paying the benefits, and not for members of the fully insured plans. Trial Ex. 749-0005. However, UBH's in-house counsel, Adam Easterday, advised Carolyn Regan, UBH's then-Vice President for Clinical Policy, that UBH could not make such a distinction. Trial Ex. 758-0003 ("Bottom line is that from legal perspective we cannot deny some commercial requests and approve others based on our financial arrangements. Since we have found TMS to be proven under some circumstances we need to cover it for all commercial plans when it meets the criteria."). In the face of this advice, Regan told Mr. Niewenhous, "[w]e will need to manage [the TMS benefit] very tightly." Id. The discussions about how to avoid or mitigate the financial impact of covering TMS included BPAC members Lorenzo Triana, Bill Bonfield, Fred Motz, Peter Brock, Michael Powell, Gerry Niewenhous, and Rhonda Robinson-Beale. See Trial Ex. 423.

187. Perhaps the most telling example of the emphasis UBH placed on financial considerations in its decision making with respect to the Guidelines relates to UBH's decision not to adopt the ASAM Criteria for making substance use disorder coverage determinations.

188. On numerous occasions throughout the Class Period - in 2012, 2013, 2014, and 2016 - UBH considered adopting the ASAM Criteria as its standard clinical coverage criteria for substance use disorders in lieu of the LOCGs and CDGs. Trial Tr. 802:4-16 (Triana) (2012); Trial Ex. 382-0003 (2013); Trial Tr. 1631:6-9 (Alam) (2013); Trial Ex. 430-0002 to -0006 (2014); Trial Ex. 524-0002 to -0004 (2016). Each time the issue came up, the UBH clinicians who specialized in addiction medicine (the "SUDs Team") recommended adopting the ASAM Criteria. Trial Tr. 1653:22-25 (Alam); Trial Ex 420; Trial Ex. 430; Trial Ex. 548-0033, -0041. Dr. Alam, a Senior Medical Director at UBH and a substance use disorder specialist, testified that there was consensus among all of UBH's addiction psychiatrists that the company should adopt the ASAM Criteria. Trial Tr. 1654:6-16 (Alam). Dr. Martorana, who supported adopting the ASAM Criteria and participated in the discussions at UBH about whether to adopt them, testified that he never heard anyone raise a clinical objection to adopting the ASAM Criteria. Trial Tr. 1122:8-19 (Martorana). Even Martha Temple - UBH's effective CEO and not a clinician - recognized that UBH should adopt the ASAM Criteria "to get in line with evidence based guidelines for our policies around Substance Use." Trial Ex. 524-0004. Ms. Temple's first request, though, was for someone to let her know the "impact" of the potential change. Trial Ex. 524-0004. The Court finds that this statement was a reference to the financial impact of adopting the ASAM Criteria.

189. Despite the clear consensus among UBH's addiction specialists that the ASAM Criteria were preferable to UBH's own Guidelines from a clinical standpoint, UBH consistently refused to replace its standard Guidelines with ASAM Criteria without first obtaining approval from the Finance Department. See, e.g., Trial Ex. 524-0002 (moving forward would require "'green light' from finance"); Trial Ex. 548-0034 ("BPAC requested that there be a financial review of possible impact of adoption of ASAM [C]riteria prior to moving forward"). But Finance would not approve the change because "a meaningful and valid BenEx modeling of the impact of a move to ASAM [C]riteria . . . [was] not possible due to the paucity of robust and relevant data." Trial Ex. 548-0034 (original emphasis). See also Trial Ex. 524-0002 ("As part of one of the SUD's work streams, we looked at adopting the ASAM guidelines but NEVER received a 'green light' from finance because they could not estimate the financial impact on BenEx in changing from using the UBH guidelines to ASAM. I recently had Martin push finance again . . . and the response was the same."). In other words, UBH rejected the recommendation of its clinicians with respect to the use of ASAM Criteria because it could not be sure that use of the ASAM Criteria would not increase BenEx. See, e.g., Trial Ex. 452-0008; Trial Tr. 781:7-782:3 (Triana); Trial Tr. 1122:8-19 (Martorana) (no clinical objections to ASAM Criteria); Trial Ex. 524-0002 (reason finance would not sign off was that "they could not estimate the financial impact on BenEx in changing from using the UBH guidelines to ASAM"); Trial Tr. 1669:2-5 (Alam) (testimony that proposed "rollout" of ASAM pilot would be terminated if it led to an increase in utilization); Trial Ex. 548-0042 (noting "[p]ossible impact on benex cost" as a "limitation" of ASAM); Trial Ex. 348-0001 to -0002 (UBH medical director warning that the ASAM Criteria "usually will result in more authorization as they are more subjective and broader than our LOCG/CDGs").

K. Exhaustion of Administrative Remedies by Class Members

190. For the purposes of this case, UBH does not dispute that all named Plaintiffs exhausted their administrative remedies. See Wit Dkt. No. 296 (Joint Proposed Pretrial Order) at 3 (stating that "UBH does not assert [the defense of failure to exhaust administrative remedies] with respect to the named Plaintiffs"). Therefore, the Court finds that each of the named Plaintiffs has exhausted administrative remedies or is deemed to have done so. Further, because the classes bring purely facial challenges to the Guidelines, the claims of named Plaintiffs put UBH on notice of the absent class members' claims, thus fulfilling the purposes of UBH's internal grievance procedure. Therefore, the Court finds that any exhaustion requirements contained in class members' plans that apply to any claims asserted in this action are excused. See Des Roches v. California Physicians' Serv., 320 F.R.D. 486, 499 (N.D. Cal. 2017) (citing Leon v. Standard Ins. Co., 2016 WL 768908, at *4 (C.D. Cal. Jan. 28, 2016); In re Household Int'l Tax Reduction Plan, 441 F.3d 500, 502 (7th Cir. 2006) ("[R]equiring exhaustion by the individual class members would merely produce an avalanche of duplicative proceedings and accidental forfeitures, and so is not required."); Barnes v. AT & T Pension Benefit Plan-Nonbargained Program, 270 F.R.D. 488, 494 (N.D. Cal. 2010) (same)).

191. Similarly, the Court finds that requiring class members to exhaust administrative remedies would be futile because their claims are based on UBH's application of faulty Guidelines in making benefits determinations and the evidence shows that the same Guidelines UBH used to make initial coverage determinations were also used to decide appeals. See Trial Ex. 257-0015 (2012 UMPD) (providing that UBH appeal reviewer must base decision on Guidelines); Trial Ex. 258-0015 (2013 UMPD) (same); Trial Ex. 259-0016 (2014 UMPD) (same); see also Trial Ex. 258-0024 (2013 UMPD) (notification of appeal decision required to cite Guidelines upon which decision was based); Trial Ex. 259-0024 (2014 UMPD) (same); Trial Ex. 260-0015 (2015 UMPD) (same); Trial Ex. 1186-0015 (2016 UMPD) (same); Trial Ex. 262-0018 (2017 UMPD) (same); Trial Ex. 257-0028 (2012 UMPD) (same).

192. UBH's witnesses testified that members' Plans vary with the respect to the administrative appeals that are available to them. See, e.g., Trial Tr. 839:6-8 (Dehlin) (Plans vary with respect to appeal rights); Trial Tr. 948: 22-949:6 (Martorana) (some Plans provide for an independent external appeal). Many class members from the Claim Sample pursued administrative appeals of UBH's denial of benefits. See Trial Ex. 1655 (summary exhibit for Claim Sample). However, UBH offered evidence that some class members who did not exhaust available administrative remedies were required under their Plans to exhaust those remedies before they could bring a legal action against UBH. See, e.g., Trial Ex. 1535-0057 (plan for class member 659) (providing that "[y]ou cannot bring any legal action against us to recover reimbursement until you have completed all the steps [described in the plan]"); Trial Ex. 1557-0084 (plan for class member 6600) (requiring exhaustion of administrative remedies both as to claims for reimbursement and as to claims "for any other reason"); Trial Ex. 1583-0085 (plan for class member 12605) (same); Trial Ex. 1633-0090 (plan for class member 7292) (same); Trial Ex. 1655 (summary exhibit showing that these class members did not file administrative appeals). Because the Court finds that any exhaustion required of class members is excused, and further finds that exhaustion would have been futile, it need not reach the question of whether the terms of any specific class member's Plan required exhaustion of administrative remedies as to the claims asserted in this action; nor does it decide whether UBH preserved any exhaustion defense it may have had as to these members by providing them adequate notice of internal appeal requirements and of their right to bring a civil action. See Bechtol v. Marsh & McLennan Cos., Inc., No. C07-1246 MJP, 2008 WL 238588, at *4 (W.D. Wash. Jan. 28, 2008) (deeming ERISA claims exhausted based on employer's failure to provide proper notice to employee of internal grievance procedure and right to bring civil action).

I. CONCLUSIONS OF LAW

A. Breach of Fiduciary Duty Claim

193. Plaintiffs bring their Breach of Fiduciary Duty Claim under 29 U.S.C. §§ 1132(a)(1)(B), which allows "participants" of ERISA plans to bring a civil action to "enforce [their] rights under the terms of the plan, or to clarify [their] rights to future benefits under the terms of the plan" and under 29 U.S.C. § 1132(a)(3)(A), which allows ERISA participants "to enjoin any act or practice which violates any provision of this subchapter or the terms of the plan." As stated above, the parties have stipulated that all of the named Plaintiffs were participants in plans governed by ERISA at the time of their noncoverage determination. Therefore, Plaintiffs are "participants" within the meaning of these sections.

194. The specific ERISA provision upon which Plaintiffs base their Breach of Fiduciary Duty Claim is 29 U.S.C. § 1104(a)(1), which sets forth the duties of ERISA plan fiduciaries. Section 1104 (a)(1) provides, in relevant part, that a "a fiduciary shall discharge his duties with respect to a plan solely in the interest of the participants and beneficiaries and

(A) for the exclusive purpose of:

(i) providing benefits to participants and their beneficiaries; and

(ii) defraying reasonable expenses of administering the plan;

(B) with the care, skill, prudence, and diligence under the circumstances then prevailing that a prudent man acting in a like capacity and familiar with such matters would use in the conduct of an enterprise of a like character and with like aims;

. . . and

(D) in accordance with the documents and instruments governing the plan insofar as such documents and instruments are consistent with the provisions of this subchapter and subchapter III.

29 U.S.C. § 1104(a)(1)(A), (B) & (D). Plaintiffs assert that UBH has breached its fiduciary duty by violating its duty of loyalty (29 U.S.C. § 1104(a)(1)(A)), its duty of due care (29 U.S.C. § 1104(a)(1)(B)), and its duty to comply with plan terms (29 U.S.C. § 1104(a)(1)(D)).

195. "[F]ederal courts have the authority to enforce the [administrative] exhaustion requirement in suits under ERISA, and . . . as a matter of sound policy they should usually do so." Amato v. Bernard, 618 F.2d 559, 568 (9th Cir. 1980). Plaintiffs contend the exhaustion requirement does not apply to their Breach of Fiduciary Duty Claim to the extent it is based on alleged breaches of the duties of loyalty and due care, while UBH argues that it applies to claims based on all three of the duties that Plaintiffs assert have been violated. The Court assumes without deciding that the exhaustion requirement applies to claims for breach of fiduciary duty based on breach of the duty of loyalty, breach of the duty of due care and breach of the duty to comply with plan terms. It finds for the reasons discussed above, however, that the requirement is satisfied as to the named Plaintiffs and excused as to the class members, and in any event, that exhaustion is not required because it would have been futile.

196. The elements Plaintiffs must prove to prevail on their Breach of Fiduciary Duty Claim are: (1) UBH was a Plan fiduciary; (2) UBH breached its fiduciary duty; and (3) the breach caused harm to Plaintiffs. LYMS, Inc. v. Millimaki, No. 08-CV-1210-GPC-NLS, 2013 WL 1147534, at *9 (S.D. Cal. Mar. 19, 2013), supplemented, No. 08-CV-1210-GPC-NLS, 2013 WL 3353838 (S.D. Cal. July 2, 2013) (citing Brosted v. Unum Life Ins. Co., 421 F.3d 459, 465 (7th Cir. 2005)).

197. UBH was a plan fiduciary with respect to Plaintiffs' Plans by virtue of its designation as administrator of mental health and substance use benefits under their Plans. See 29 U.S.C. § 1002(14)(A) ("fiduciary" includes "any administrator"). Further, Plaintiffs' Plans delegated discretionary authority to UBH to interpret and apply plan terms, and UBH exercises that authority when it makes coverage determinations and more broadly, when it adopts Guidelines to standardize its coverage determinations and to ensure that those determinations are consistent with generally accepted standards of care. Thus, when it adopts and applies its Guidelines to coverage determinations, UBH is required to act in a manner that is consistent with the fiduciary duties set forth above, that is, the duty of loyalty, the duty of due care and the duty to comply with plan terms. See Lockheed Corp. v. Spink, 517 U.S. 882, 887 (1996) ("[O]nly when fulfilling certain defined functions, including the exercise of discretionary authority or control over plan management or administration' does a person become a fiduciary under ERISA") (internal quotation marks and citation omitted).

198. The Supreme Court has explained that "[n]othing in ERISA requires employers to establish employee benefits plans . . . [or] . . . mandate[s] what kind of benefits employers must provide if they choose to have such a plan." Id. at 887; see also Curtiss-Wright Corp. v. Schoonejongen, 514 U.S. 73, 78 (1995) ("Employers or other plan sponsors are generally free under ERISA, for any reason at any time, to adopt, modify, or terminate welfare plans."). Consequently, if UBH were the plan sponsor with respect to Plaintiffs' Plans, it would be acting in a capacity that is analogous to the settlor of a trust, rather than as a fiduciary, and would not owe Plaintiffs the fiduciary duties discussed above. See Lockheed Corp., 517 U.S. at 890. The evidence at trial established, however, that UBH was not a plan sponsor with respect to Plaintiffs' Plans and did not have the authority to modify the terms of Plaintiffs' Plans. Rather, as plan administrator, it only had authority to interpret and apply the terms of Plaintiffs' Plans. Consequently, UBH was not functioning as a plan settlor when it adopted the Guidelines or when it applied them to Plaintiffs' claims.

199. The parties agree, as a general proposition, that the question of whether UBH breached its fiduciary duty to comply with plan terms is governed by an abuse of discretion standard of review. They disagree, however, on the standard of review that applies to the questions of whether UBH breached its fiduciary duties of loyalty and due care. UBH contends these claims are so closely tied to the question of whether its Guidelines are proper that it is entitled to the same deference with respect to interpretation of plan terms as it afforded with respect to the claim for failure to comply with plan terms. Plaintiffs disagree, arguing that these claims are subject to de novo review. The Court does not need to decide this question because it concludes, for the reasons stated below, that UBH has breached its fiduciary duty under the abuse of discretion standard.

200. Under the abuse of discretion standard, a "plan administrator's decision 'will not be disturbed if reasonable.'" Stephan v. Unum Life Ins. Co. of Am., 697 F.3d 917, 929 (9th Cir. 2012) (quoting Conkright v. Frommert, 559 U.S. 506, 521 (2010) (internal quotation marks omitted)). Under this standard, an administrator's decision is entitled to deference unless it is "'(1) illogical, (2) implausible, or (3) without support in inferences that may be drawn from the facts in the record.'" Id. (quoting Salomaa v. Honda Long Term Disability Plan, 642 F.3d 666, 676 (9th Cir. 2011)). "This abuse of discretion standard, however, is not the end of the story." Id. Rather, "the degree of skepticism with which [courts] regard a plan administrator's decision when determining whether the administrator abused its discretion varies based upon the extent to which the decision appears to have been affected by a conflict of interest." Id. For example, in Stephan, the court found that there was a structural conflict of interest that had to be considered in determining whether or not there had been an abuse of discretion because the defendant, Unum Life Insurance Company, played a "dual role as plan administrator, authorized to determine the amount of benefits owed, and insurer, responsible for paying such benefits." Id. (citing Metro. Life Ins. Co. v. Glenn, 554 U.S. 105, 114 (2008)). In Glenn, the Supreme Court held that an insurance company, like an employer, may have a conflict of interest even though the insurance company charges the employer "a fee that attempts to account for the cost of claims payouts." 554 U.S. at 114. The Court found that under these circumstances the claim payout may not come from the insurance company's own pocket "to the same extent" it does when an employer is a plan administrator, but there is, nonetheless, a conflict of interest. 554 U.S. at 114. One reason for this conflict of interest, the Court explained, is that "the employer's own conflict may extend to its selection of an insurance company" because "[a]n employer choosing an administrator in effect buys insurance for others and consequently (when compared to the marketplace customer who buys for himself) may be more interested in an insurance company with low rates than in one with accurate claims processing." Id.

201. The degree of skepticism that is appropriate when a plan administrator has a conflict of interest depends upon the circumstances. As the Court explained in Glenn, "where circumstances suggest a higher likelihood that [the conflict] affected the benefits decision, including, but not limited to, cases where an insurance company administrator has a history of biased claims administration," more skepticism is warranted. 554 U.S. at 117. On the other hand, the conflict "should prove less important (perhaps to the vanishing point) where the administrator has taken active steps to reduce potential bias and to promote accuracy, for example, by walling off claims administrators from those interested in firm finances, or by imposing management checks that penalize inaccurate decision making irrespective of whom the inaccuracy benefits." Id.

202. The evidence introduced at trial supports the conclusion that significant skepticism is warranted in determining whether UBH abused its discretion when it adopted the Guidelines that are challenged in this case. First, the evidence shows that UBH had a structural conflict of interest throughout the class period because a large portion of its revenues came from fully insured plans. Moreover, the evidence shows that even as to the self-funded plans, UBH felt pressure to keep benefit expenses down so that it could offer competitive rates to employers. Second, regardless of whether the financial incentive to keep benefit expenses down was stronger with respect to the fully insured plans or the self-funded plans, the conflict of interest affected all members equally, regardless of which type of plan they were insured under, because UBH used a single set of Guidelines to make coverage determinations. Third, UBH did not ensure that the internal process it set up for adopting and revising the Guidelines insulated the individuals who developed the Guidelines from financial considerations. To the contrary, UBH included administrators from its Finance and Affordability Departments on the committees that ultimately had to approve the Guidelines. Further, as to those individuals who were involved in the Guideline development process who were not in those Departments, such as Mr. Niewenhous, UBH made sure that on a regular basis they received detailed financial information about "utilization," including whether targets set by UBH in particular categories of services were being met. Finally, the evidence at trial established that the emphasis on cost-cutting that was embedded in UBH's Guideline development process actually tainted the process, causing UBH to make decisions about Guidelines based as much or more on its own bottom line as on the interests of the plan members, to whom it owes a fiduciary duty. This was apparent from UBH's handling of TMS and ABA benefits, discussed above. Most striking, however, was the obvious impact of financial considerations on UBH's decision making as to the adoption of the ASAM Criteria. UBH's refusal to adopt the ASAM Criteria was not based on any clinical justification. Indeed, all of its clinicians recommended that the ASAM Criteria be adopted. The only reason UBH declined to adopt the ASAM Criteria was that its Finance Department wouldn't sign off on the change. In other words, UBH's Finance Department had veto power with respect to the Guidelines and used it to prohibit even a change in the Guidelines that all of its clinicians had recommended. This evidence establishes that UBH has a conflict of interest that has had a significant impact on decision-making as to the development of the Guidelines. Therefore, in applying the abuse of discretion standard to Plaintiffs' Breach of Fiduciary Duty Claim, the Court views UBH's decision making with significant skepticism.

203. Applying the standard of review discussed above, and based on the Findings of Fact related to the challenged Guidelines and UBH's Guideline development process, the Court finds, by a preponderance of the evidence, that UBH has breached its fiduciary duty by violating its duty of loyalty, its duty of due care, and its duty to comply with plan terms by adopting Guidelines that are unreasonable and do not reflect generally accepted standards of care.

204. As discussed above, the final element of Plaintiffs' Breach of Fiduciary Duty Claim is that the breach must have caused harm to Plaintiffs. The Court finds that this requirement is met. As the Court found on summary judgment, the harm that Plaintiffs allege resulted from UBH's breach of fiduciary duty is the denial of their right to fair adjudication of their claims for coverage based on Guidelines that were developed solely for their benefit. See Wit, Dkt. No. 286 at 24-25. The Court declines to revisit that conclusion.

205. UBH argues that to the extent that the Denial of Benefits Claim is asserted under both 29 U.S.C. § 1132(a)(1)(B) and § 1132(a)(3)(A), the Court should dismiss the latter claim on the basis that the former claim provides adequate relief. UBH relies on the rule that equitable relief under § 1132(a)(3) is not available if § 1132(a)(1)(B) provides an adequate remedy. See Varity Corp. v. Howe, 516 U.S. 489, 512 (1996). It is well-established, however, that under Varity, claims asserted under § 1132(a)(1)(B) and § 1132(a)(3) "may proceed simultaneously so long as there is no double recovery." Moyle v. Liberty Mut. Ret. Ben. Plan, 823 F.3d 948, 961 (9th Cir. 2016), as amended on denial of reh'g and reh'g en banc (Aug. 18, 2016). As the Court has not yet addressed the question of remedies, UBH's request that the Court dismiss the Breach of Fiduciary Duty Claim asserted under § 1132(a)(3)(A) is premature.

206. For these reasons, the Court finds that UBH is liable with respect to the Breach of Fiduciary Duty Claim.

B. Denial of Benefits Claim

207. Plaintiffs assert the Denial of Benefits Claim under 29 U.S.C. § 1132(a)(1)(B) and 29 U.S.C. § 1132(a)(3)(B). As stated above, 29 U.S.C.§ 1132(a)(1)(B) allows ERISA plan participants to bring a civil action to "enforce [their] rights under the terms of the plan, or to clarify [their] rights to future benefits under the terms of the plan." Section 1132(a)(3)(B) allows ERISA plan participants to bring a civil action to "to obtain other appropriate equitable relief (i) to redress such violations or (ii) to enforce any provisions of this subchapter or the terms of the plan."

208. To the extent Plaintiffs are required to exhaust their administrative remedies with respect to the Denial of Benefit Claim, the Court concludes that the requirement is met as to the named Plaintiffs and excused as to the remaining class members. It further finds that as to any class members who did not exhaust their administrative remedies that exhaustion would have been futile.

209. To prevail on their Denial of Benefits Claim, Plaintiffs must establish by the preponderance of the evidence that: 1) one condition of coverage under the class members' Plans was that the requested treatment was consistent with generally accepted standards of care and/or the standards mandated by state law; 2) when determining whether a request for coverage satisfied its Guidelines, UBH was interpreting and applying those plan terms; 3) UBH's Guidelines were not consistent with generally accepted standards or the standards mandated by state law; and 4) UBH denied Plaintiffs' requests for coverage for outpatient, intensive outpatient, or residential treatment based in whole or in part on UBH's Guidelines.

210. Plaintiffs' claim for Denial of Benefits is reviewed under an abuse of discretion standard. The Court applies that standard with significant skepticism for the reasons discussed above.

211. One condition of coverage under each class member's Plan was that the services for which coverage was requested are consistent with generally accepted standards of care and/or the standards mandated by state law. In applying its Guidelines to class members' requests for coverage, UBH was interpreting the terms of their Plans.

212. Applying the standard of review discussed above, and based on the Findings of Fact related to the challenged Guidelines and UBH's Guideline development process, the Court finds, by a preponderance of the evidence, that UBH's Guidelines were unreasonable and an abuse of discretion because they were more restrictive than generally accepted standards of care.

213. In addition to plan terms requiring UBH to use generally accepted standards of care, UBH was specifically required, pursuant to the laws of Illinois, Connecticut, Rhode Island, and Texas, to administer requests for benefits pursuant to Plans governed by those states' laws in accordance with those laws. For the reasons stated above, the Court finds that UBH did not adhere to these state law requirements.

214. UBH denied Plaintiffs' requests for coverage for outpatient, intensive outpatient, or residential treatment based in whole or in part on UBH's Guidelines.

215. UBH argues that to the extent that the Denial of Benefits Claim is asserted under both 29 U.S.C. § 1132(a)(1)(B) and § 1132(a)(3)(B), the Court should dismiss the latter claim on the basis that the former claim provides adequate relief, again relying on Varity. For the reasons discussed above, the Court finds that UBH's request is premature.

216. For these reasons, the Court finds that UBH is liable with respect to the Denial of Benefits Claim.

IT IS SO ORDERED. Dated: February 28, 2019

/s/_________

JOSEPH C. SPERO

Chief Magistrate Judge