(i Arbitrary and capricious procedure
BC/BS, as the fiduciary under ERISA of Ms. Kekis's health insurance plan, was bound to act only in accordance with the express terms of the policy. See ERISA § 404, 29 U.S.C. § 1104(a)(1)(D) (West Supp. 1992) ("Fiduciary duties"). As a fiduciary, BC/BS cannot disregard any policy requirement, nor can it implicitly incorporate into the policy any term that does not expressly appear. See, e.g., Clarke v. Bank of New York, 687 F. Supp. 863, 867 (S.D.N.Y. 1988) (citing, e.g., Miles v. New York State Teamsters Conf., 698 F.2d 593, 599 (2d Cir.), cert. denied, 464 U.S. 829, 78 L. Ed. 2d 108, 104 S. Ct. 105 (1983)); see also Morgan v. Mullins, 643 F.2d 1320, 1321 (8th Cir. 1981). The fiduciary's duties are dictated by the policy, such that a fiduciary who ignores the terms of the policy and inserts its own terms acts arbitrarily and capriciously. Cf., e.g., Kulakowski v. Rochester Hospital Serv. Corp., 779 F. Supp. 710, 716-17 (W.D.N.Y. 1991) (arbitrary and capricious for fiduciary "to ignore the language of [the Contract] and to insert a new exclusion . . . that no coverage would be provided for therapy which is 'investigational'" (quotation omitted) (citation omitted)); Dozsa v. Crum & Forster Ins. Co., 716 F. Supp. 131, 139 (D.N.J. 1989) (abuse of discretion for fiduciary to "apply a different exclusion having different criteria from the Plan's exclusion").
When it reviewed Ms. Kekis's request for coverage, BC/BS applied the experimental/investigative services exclusion in a manner that was not provided for in the policy itself. While the policy gives BC/BS discretion to exclude from coverage those services and procedures which are "experimental/investigative services," the policy provides a strict, specific definition of what constitutes such services. The policy defines experimental/investigative services as "any service or procedure we do not recognize as accepted medical practice." See Def. exh. "A" P 18. The policy then defines that term ("any service or procedure we do not recognize as accepted medical practice") as a service or procedure that BC/BS determines "has no proven medical value." See id. In short, the policy provides insureds such as Ms. Kekis a working definition of experimental/investigative services: a procedure is experimental/investigative under the policy only if, in the view of BC/BS, the service or procedure has no proven medical value. If BC/BS determines that a service or procedure does have proven medical value, then the experimental/investigative exclusion clause does not provide a legitimate basis for denying coverage.
The court finds that BC/BS did not apply this definition of experimental/investigative services when it reviewed Ms. Kekis's request for coverage. The evidence revealed that BC/BS substituted its own definition of experimental/investigative services that differs from the definition set forth in the plan. In fact, BC/BS continued to apply its own, incorrect definition of experimental/investigative throughout these proceedings. Specifically, BC/BS erroneously applied the common, "dictionary" definition of experimental/investigative when it reviewed and denied Ms. Kekis's request for coverage, rather than the definition of experimental/investigative services set forth in its own policy. While the dictionary might provide the most widely accepted, reasonable definition of the words "experimental" and "investigative," the insurance policy provides the binding definition of those terms. The fact that the policy, which BC/BS drafted, carries an arguably uncustomary definition of experimental/investigative services is immaterial. National Grange Mutual Ins. Co. v. Continental Casualty Ins. Co., 650 F. Supp. 1404, 1408 (S.D.N.Y. 1986) ("exclusions from coverage 'must be specific and clear in order to be enforced. They are not to be extended by interpretation or implication, but are to be accorded a strict and narrow interpretation'" (citation omitted)). Since BC/BS, as fiduciary of Ms. Kekis's policy, applied a definition of experimental/investigative services that is not the definition mandated by the policy, BC/BS acted arbitrarily and capriciously when it relied upon this definition in denying her request for coverage. Cf. Kulakowski, 779 F. Supp. at 716-17; Dozsa, 716 F. Supp. at 139.
There are numerous indications in the record that BC/BS relied upon an impermissible standard when it refused Ms. Kekis's request. The most vivid evidence of this misapplication was presented in the testimony of Dr. Davies, BC/BS's medical director. Both Dr. Davies and Henry F. Becker, Vice President of BC/BS, testified that Dr. Davies enjoys broad discretion in making decisions concerning the legitimacy of proposed medical treatments. To the extent that Dr. Davies's recommendations involve medical judgments, they are generally left undisturbed by BC/BS management. Dr. Davies testified that he recommended rejection of Ms. Kekis's request because it was obvious that she would be participating in a "study" conducted under a research protocol. As an initial matter, Dr. Davies found the use of a protocol to be particularly significant because protocols, in his view, are generally used to describe an experiment using human subjects to be performed in furtherance of a larger study to investigate the effects of the treatment. See Def. Mem. at 40; Def. exh. "S" (excerpt from National Cancer Institute Investigator's Handbook, defining protocol as "the detailed written plan of a clinical experiment"). Contra, e.g., Adams, 757 F. Supp. at 675 (use of protocol does not mean that treatment is experimental); Dozsa, 716 F. Supp. at 138.
Indeed, the information that Dr. Davies reviewed was replete with references, most conspicuously from Ms. Kekis's own oncologists, that she would be participating as part of a research program testing the efficacy of HDC-ABMT vis-a-vis standard chemotherapy, the results of which were not yet known. See Def. exit. "K" (package sent by Strong Memorial, including letters and protocol, described supra pp. 9-11). Moreover, Dr. Davies had read several published articles prepared by distinguished scientists, each stating that the value of HDC-ABMT compared to standard chemotherapy is unknown and still under investigation. See, e.g., Def. Mem. at 31-37 (citing articles). Given that Ms. Kekis clearly proposed to be part of an investigation, and many of her own doctors-- including Dr. DiPersio, who would supervise the procedure-- confirmed the investigative nature of her plan, Dr. Davies concluded that her treatment constituted an experimental/investigative service that should be excluded from coverage.
No one can seriously question Dr. Davies's good faith in reaching his conclusion. His credentials are impressive, he testified sincerely and his conclusion regarding the experimental and investigative nature of Ms. Kekis's treatment-- under the ordinary definition of the terms "experimental" and "investigative"-- is substantiated by the affidavits of two other experts in the field. See Eddy Aff. (2/24/93) P 15 ("it is my unequivocal opinion there was absolutely no medical evidence that the HDC/ABMT procedure as proposed for Michelene D. Kekis is anything but experimental"); Herberman Aff. (2/26/93) P 23 ("it is my unequivocal opinion, based upon my years and training as a cancer researcher, that HDC/ABMT is experimental, and more importantly the particular procedure advanced by Dr. DiPersio and the University of Rochester is experimental"). His experience and credentials notwithstanding, Dr. Davies, as BC/BS's medical director, was bound to apply the policy's definition of experimental/investigative services, not his own definition or one supplied by experts in the field.
Dr. Davies clearly did not follow the standard mandated by the BC/BS policy when he reviewed Ms. Kekis's request for coverage. Instead, he considered numerous irrelevant factors, including evidence that she would receive treatment under a protocol and as part of an investigative study. All of these facts are immaterial to consideration of the only permissible inquiry: does HDC-ABMT have any proven medical value? If BC/BS found the answer to be yes, then it could not properly rely upon the experimental/investigative services clause to deny Ms. Kekis's request for coverage. While Dr. Davies asked many other questions concerning the experimental nature of HDC-ABMT, he did not consider this fundamental question. Therefore, he did not review Ms. Kekis's request for coverage in accordance with the express terms of the policy, and his recommendation which was purportedly based upon the policy terms was arbitrary and capricious.
See Dozsa, 716 F. Supp. at 139.
The court's finding that Dr. Davies, and ultimately BC/BS, acted arbitrarily and capriciously in interpreting the policy follows the rulings in Kulakowski v. Rochester Hosp. Serv. Corp., supra, 779 F. Supp. 710, and Dozsa v. Crum & Forster Ins. Co., supra, 716 F. Supp. 131. In both of those cases, the defendant insurer relied upon an exclusion clause to deny the plaintiff's request for coverage for HDC-ABMT. In Kulakowski, the insurer relied upon its finding that the proposed treatment was "investigational" and thus excluded from coverage under the policy. Kulakowski, at 716.
The insurer's denial of coverage was peculiar, however, because the governing policy allowed an exclusion for "experimental" treatments but made no mention of excluding "investigational" treatments. Id. at 716-17. Since the insurer relied upon criteria-- the treatment was investigational-- that were not a legitimate basis for excluding coverage under the policy, the court found that the exclusion was arbitrary and capricious. Id.
The court in Dozsa was presented with an astonishingly similar situation. See generally Dozsa, 716 F. Supp. 131. After reviewing the insurer's denial of coverage in light of the policy provision upon which the insurer relied, the court concluded:
What [the insurers] did in effect was to ignore the language of . . . [the] Medical Plan and to insert a new exclusion. The new exclusion was that no coverage would be provided for therapy which is "investigational" and therapy which is not backed by a consensus in peer reviewed medical literature is "investigational". That may very well be the terms of an exclusion in [the insurer's] own medical policies and in other medical plans which it administers. It is clearly not a basis for exclusion in the . . . Medical Plan.