MEMORANDUM AND ORDER

F. GOSSETT, Magistrate Judge

This is an action under the Employee Retirement Income Security Act, 29 U.S.C. § 1001, et seq. ("ERISA") and is now before the magistrate judge on the plaintiff's Motion for Leave to Conduct Discovery Outside the Administrative Record (Filing No. 25). Having considered the defendant's ("Coventry's") response (Filing No. 27), together with Coventry's "administrative record" (Filing No. 23), I find that the motion should be granted.

I. BACKGROUND

The court has reviewed the "administrative record" filed by Coventry in order to determine whether additional discovery is warranted.

In August 2005, the plaintiff, Steven Zalkin, was diagnosed with the multicentric form of Castleman Disease. Multicentric Castleman Disease is a rare aggressive multisymptom disorder that can result in death due to infection and also transformation into non-Hodgkin's lymphoma. (Filing No. 23-4 at p. 83/112, COVZ000357). The disease is not officially designated as a cancer; however, it resembles lymphoma. (Filing No. 23-3 at p. 104/107, COVZ000271; Filing No. 23-4 at p. 75/112, COVZ000349). There is no standard treatment for Castleman Disease. (Filing No. 23-4 at p. 83/112, COVZ000357).

By April 1, 2008, plaintiff had received two courses of prednisone, two courses of rituximab, and one course of high dose dexamethasone. (Filing No. 23 at p. 63/99, COVZ000067). He had suffered a severe reaction in his lower legs, possible due to the Castleman Disease, but which was resolved with hydroxyzine. Plaintiff was developing fatigue, drenching night sweats, and persistent itching in his legs; he had not yet suffered from infections or bleeding, and had no new adenopathy. ( Id.) Each course of treatment caused a partial remission, but each time the plaintiff's symptoms recurred. On April 22, 2008, plaintiff's doctors recommended that he pursue a course of CVP chemotherapy plus rituximab. (Filing No. 23 at p. 64/99, COVZ000068).

Coventry, the plaintiff's health insurance carrier, states that it issued and administered a fully-funded HMO point of service open access plan, sponsored by the plaintiff's business, Alamar Corporation. (Filing No. 20, Answer at ¶ 2).

By letter dated May 12, 2008, Coventry approved the plaintiff's request for coverage of an Autologous Stem Cell transplant evaluation at the UNMC. (Filing No. 23 at p. 52/99, COVZ000056). The letter noted that approval of coverage for the evaluation did not ensure approval of coverage of the transplant itself.

Request for Preauthorization

On May 19, 2008, plaintiff sought preauthorization from Coventry to proceed with "high intensity chemotherapy, followed by an autologous peripheral stem cell transplant." (Filing No. 23 at p. 58/99, COVZ000062). Coventry denied the request for preauthorization on May 23, 2008, advising the plaintiff that the proposed treatment was "experimental and investigational" and coverage was excluded under sections 7.1.1, 7.1.2, and 7.2.59 of the insurance policy. (Filing No. 23 at p. 97-98/99, COVZ000101-102; Policy, Filing No. 23-6 at pp. 49 54/90; COVZ000546 551)

Coventry's initial decision was based on a case report apparently procured through a peer review analysis organization or business called "mcmc" and prepared by Howard J. Fingert, MD. Dr. Fingert's brief report, printed on the letterhead of "mcmc" notes that the University of Nebraska Medical Center ("UNMC") was "one of the most experienced transplant centers caring for patients with lymphoproliferative disorders, and some promising experience has been published using auto transplant for Castleman's disease from this same center," and "[a]necdotal experience has been published using transplant in this diagnosis[.]" Citing two references, the unsigned report concludes that the experience was "not sufficient to conclude that this approach can provide reliable and durable clinical utility." (Filing No. 23 at pp. 93-94/99, COVZ000097-98).

First Level Appeal

On May 28, 2008, plaintiff's attending physician, Dr. James Armitage, appealed the transplant denial and provided Coventry with peer review literature from the UNMC and the Mayo Clinic demonstrating that the proposed transplant procedure had value in the treatment of Castleman Disease. (Filing No. 23-2 at pp. 5-30/64, COVZ000107-133). Dr. Armitage advised that an auto stem cell transplant is, at the present time, a treatment most likely to cause a durable remission of Castleman Disease, and there is no standard of care treatment for the disease.

Plaintiff's first-level appeal was denied on May 30, 2008, based on additional medical opinions, again procured through "mcmc." None of the "mcmc" consultants conferred with the plaintiff's attending physician. Coventry presented the following questions to the consulting physicians:

1. Is the requested rare disease autologous stem cell transplant experimental/investigational per plan language?

2. If not considered experimental/investigational, is it medically necessary?

James Wortman, M.D. — Citing 18 references, Dr. Wortman opined that a stem cell transplant for Castleman Disease would be considered experimental/investigational, as there were only a small number of anecdotal reports of success in the literature due to the rarity of the disease. The successes discussed in the literature were "mainly in the neurologic disturbances associated with this disease." The plaintiff's symptoms, however, were generalized. While SCT was a reasonable choice for this patient, there were "a number of other experimental/investigational alternatives available for this disease." (Filing No. 23-4 at p. 75/112, COVZ000349). Dr. Wortman went on to discuss five experimental or investigational alternatives. His discussion concludes:

[Author] Dispenziera reviews 16 patients with autologous stem cell transplants and out of those, 14 had improvement or stabilizations of their neurologic abnormalities. This current patient has no neurologic complaints and manifests his disease in a much different way. Although a stem cell transplant is a reasonable option in a patient such as this, with a number of anecdotal successes, there are other alternative therapeutic treatments programs along with observation alone that have met with some success. Bone marrow transplantation is one of these, but due to a limited number of studies would still be considered experimental/investigational and therefore not medically necessary based on policy language.

(Filing No. 23-4 at p. 76/112, COVZ000350).

Kenneth Pienta, M.D. — Citing three resources, Dr. Pienta advised that "optimal standard therapies" have not been established for Castleman Disease, a rare heterogeneous group of diseases involving proliferation of lymphoid cells. "Currently, most patients receive treatments derived from past experience with non-Hodgkin lymphoma." The role of bone marrow transplant and more recently developed courses of treatment remain undefined for this disease. Thus, he concluded, the requested autologous stem cell transplant was experimental/investigational per plan language.

Second Level Appeal

Dr. Armitage filed a second level appeal on June 17, 2008, on plaintiff's behalf. Dr. Armitage requested the opportunity to speak before the appeal committee and asked that an expert in lymphoma and transplantation attend the hearing. (Filing No. 23-3 at p. 5/107, COVZ000172). Plaintiff was promptly advised that the Corporate Grievance Committee would review the second level appeal on June 23, 2008 at 10:00 AM.

Apparently, the appeal was held by conference call with Dr. Armitage, Coventry's Corporate Grievance Committee, and the plaintiff. The appeal information was reportedly reviewed by Lee P. Hartner, MD, Board Certified Medicine, Hematology and Oncology; Debra Esser, MD, VP Medical Affairs, Board Certified Family Practice; Pat Schaefer, RN, Director Health Services. (Filing No. 23-3 at p. 26/107, COVBZ000193). There is no indication that any expert in lymphoma or transplantation attended the hearing or was consulted. No transcript of the hearing was provided in Coventry's "administrative record."

Documents generated by Coventry indicate that its denial of plaintiff's second level appeal was based mainly on Dr. Hartner's opinion, which Coventry procured through the "mcmc" organization. (Filing No. 23-3 at pp. 20-21/107, COVZ000187-88). Dr. Hartner's signed opinion (Filing No. 23-3 at pp. 17-19/107; COVZ000184-186) is similar in content and format to the other opinions procured by Coventry through "mcmc." He did provide additional information that multicentric Castleman disease is thought to be related to "Polyneuropathy, organomegaly, endocrinopathy, monoclonal gammopathy, and skin changes (POEMS) syndrome." While emerging data supported the use of autologous transplantation in the treatment of POEMS syndrome, the use of this treatment had not been "well studied" in Castleman disease, and the literature consisted only of case reports. There was no standard treatment for Castleman disease, and no controlled trials existed to support the use of any recommended therapies. Dr. Hartner opined that it was "certainly reasonable to consider the use of autologous stem cell transplantation in this patient with multicentric Castleman's disease who has failed prior therapy." The treatment, however, was considered experimental/investigational, based on the language of the Coventry insurance policy, which made "no exception for the rarity of certain disorders, which precludes the determination of any consensus regarding the safety and efficacy of particular treatments." (Filing No. 23-3 at p. 18/107, COVZ000185).

Complaint Filed with the Nebraska Department of Insurance

Noting that his second level appeal was denied within 24 hours of the conference call, plaintiff filed a complaint against Coventry with the State of Nebraska Department of Insurance ("State"). Plaintiff advised that he was told Coventry uses an independent firm that would send his information to experts in the field to determine his appeal; however, the individuals who reviewed his appeal were not experts in the treatment of Castleman disease and did not have any scientific data in relationship to the treatments and successes of treating Castleman disease. (Filing No. 23-3 at p. 26/107, COVZ000193).

In response to the State's inquiry, Coventry provided numerous documents, together with the explanation that

It is Coventry's policy with respect to any request for any transplant, if the participating specialist determines the member is a candidate for transplantation is to authorize an evaluation at an approved facility. The evaluation is to determine if the individual is an appropriate candidate for the specific transplant in question and if the transplant is deemed to be a covered benefit in accordance with the plan contract (e.g. not deemed to be experimental or investigational, ect. [sic]). Additionally, the evaluation will be used by the participating specialist to determine if another course of treatment may be more beneficial for that individual patient under a given set of circumstances. The evaluation does not approve coverage for the transplant.

* * * *

Castleman's Disease is a rare condition. However, it is not the case that any treatment for a rare condition will de facto be considered investigational or experimental. Any request Coventry receives for treatment for a rare condition or any condition from that matter is, reviewed by a medical director and compared to medical necessity criteria. If medical necessity criteria is unavailable, then the treatment request would be reviewed using evidence based medicine, peer reviewed literature and/or same specialty external expert review to make a determination.

(Filing No. 23-3 at pp. 29-30/107, COVZ000196 197).

In response, the State investigator noted that all of Coventry's consulting physicians had agreed there was no standard of care for the plaintiff's condition, and at least two of the consultants acknowledged the stem cell transplant as a reasonable option. One physician referenced "other experimental/investigational alternatives available," suggesting that any reasonable treatment options would be similarly denied. Another physician specifically attributed the lack of supporting literature to the rarity of the disorder, and the insurance policy made no exception for the rarity of certain disorders. The State investigator concluded that the case file suggested a disparate impact of the policy language resulting in the treatment of rare conditions being held de facto experimental by Coventry.

Coventry's August 1, 2008 response repeated its original statement. Coventry again stated that the Coventry policy does provide access to a reasonable level of covered treatment for insureds with rare conditions, citing an instance where it decided to allow coverage for the treatment of Amyloidosis. (Filing No. 23-4 at pp. 3-6/112, COVZ000277-280).

The administrative complaint is not discussed in the plaintiff's brief, and the records supplied by Coventry appear to be silent as to the State's resolution of the matter.

LEGAL ANALYSIS

The issue presented in this motion is whether the plaintiff may conduct discovery outside the administrative record produced by Coventry (Filing No. 23). Following the Supreme Court's decision in Metropolitan Life Ins. Co. v. Glenn (which did not specifically address the scope of discovery appropriate in cases arising under ERISA), the Eighth Circuit has observed:

There is no doubt that Glenn changed ERISA review in some ways. First, the Supreme Court determined specifically that when the entity that administers the plan "both determines whether an employee is eligible for benefits and pays benefits out of its own pocket" a conflict of interest exists. Glenn, 128 S.Ct. at 2346 . Prior to Glenn, this Court held the opposite. See, e.g., Chronister I, 442 F.3d at 655 ("[I]t is wrong to assume a financial conflict of interest from the fact that a plan administrator is also the insurer.") (quoting McGarrah v. Hartford Life Ins. Co., 234 F.3d 1026, 1030 (8th Cir. 2000)).

Similarly, under this Court's pre- Glenn precedent, a financial conflict of interest would not trigger less-deferential review unless the claimant could show that the conflict was causally connected to the specific decision at issue. See Woo v. Deluxe Corp., 144 F.3d 1157, 1160 (8th Cir. 1998); McGarrah, 234 F.3d at 1030 . Glenn makes clear that, while a causal connection might be important in determining the appropriate level of scrutiny for a plan administrator's decisionmaking, such a connection is not required. Glenn, 128 S.Ct. at 2351 ("The conflict of interest . . . should prove more important . . . where circumstances suggest a higher likelihood that it affected the benefits decision. . . ."). Under Glenn, courts must analyze the facts of the case at issue, taking into consideration not only the conflict of interest, but also other factors that might bear on whether the administrator abused its discretion. Id.

"Although the Glenn Court ruled that a Plan Administrator's conflict of interest is a factor that can be explored in an ERISA case, it did not discuss how the Plaintiff went about establishing such a conflict." Almeida v. Hartford Life Accident Ins. Co., 2010 WL 743520 at *4, Case No. 09-cv-1556 (D. Colo., March 2, 2010). In Chronister, the Eighth Circuit noted that it was not faced with determining whether Glenn changes the discovery limitations in ERISA cases. Chronister, 563 F.3d at 775 n. 2. "Simply put, Glenn's effect on the discovery rules for ERISA cases remains unclear." Winterbauer, 2008 WL 4643942 at *5.

It is clear that, after Glenn, a reviewing court may no longer disregard an ERISA plan's structural conflicts of interest without further analysis. See Denmark v. Liberty Life Assurance Co. of Boston, 566 F.3d 1, 9 (1st Cir. 2009). A structural conflict of interest occurs when, as here, the plan administrator both determines eligibility and pays the claims. The conflict of interest must be considered as a factor when conducting a deferential review of the plan administrator's decision to deny benefits. "Evidence of a conflict of interest may appear on the face of the Plan, by evidence of improper incentives, or through proof of a pattern or practice of unreasonably denying meritorious claims." Almeida v. Hartford Life Accident Ins. Co., ___ F. Supp. 2d ___, 2010 WL 743520 at *2, Case No. 09-cv-1556 (D. Colo., March 2, 2010) (citing Glenn, 128 S. Ct. at 2354 (Roberts, C.J., concurring)). The significance of the factor will depend on the circumstances of the particular case. Glenn, 128 S. Ct. at 2346 .

The Glenn decision was not intended to "bring about near universal review by judges de novo-i.e., without deference-of the lion's share of ERISA plan claims denials." Glenn, 128 S. Ct. at 2350 . However, even in a case which is subject to deferential review, the court must analyze the facts of the case at issue, considering all the factors that might bear on whether the Plan administrator abused its discretion. See Chronister, 563 F.3d at 775 .

In this instance, the plaintiff seeks leave to conduct discovery regarding:

1. The conflict of Coventry (Plan Administrator) because of direct financial interest to Coventry when claims are denied;

2. The background, training, and credentials of the reviewing parties at Coventry making the initial denial of benefits;

3. The conflict of interest of the reviewing parties at Coventry making the initial denial of benefits;

4. The conflict of those so-called experts employed by Coventry in the appeals process, including how they were determined to be most knowledgeable about the issues in question, how many previous times they have been retained by Coventry, what direct financial interest they potentially have because of remuneration paid and because of the prospect of additional retention;

5. Whether Coventry had the most up to date scientific studies and treatises in their possession or if they stopped investigating when they found those they believed supportive of Coventry's position;

6. How many other similar cases have been determined and to what result;

7. Why autologous stem cell treatment has been accepted as appropriate treatment repeatedly by Coventry but it made a distinction for treatment in this case, and whether such distinction was rational[.]

Filing No. 26, Plaintiff's brief at pp. 9-10.

In this case, the plaintiff sought insurance coverage for treatment of a very rare condition. Coventry's "administrative record" demonstrates that it solicited medical opinions through an organization ("mcmc"), the independence and qualifications of which are nowhere disclosed. The qualifications of the doctors from whom Coventry and/or "mcmc" procured opinions are nowhere disclosed. The medical opinions upon which Coventry apparently relies appear on the letterhead of "mcmc," and all contain a boilerplate disclaimer to the effect that the physician had no conflicts of interest. Only one of the opinions bears a signature. The terms of Coventry's arrangement with "mcmc" for the procurement of medical opinions remain unknown. It is unknown how or whether Coventry or the "mcmc" organization screens physicians for relevant levels of expertise or conflicts of interest, financial or otherwise.

"[T]he extent of any alleged conflict of interest could be shown by how Defendant instructs third party consultants, doctors and reviewers and/or whether Defendant provides incentives to them." Almeida, 2010 WL 743520 at *3 (citing Glenn, 128 S. Ct. at 2354 (Roberts, C.J., concurring), in turn, citing Armstrong v. Aetna Life Ins. Co., 128 F.3d 1263, 1265 (8th Cir. 1997)). In Almeida, the court allowed the plaintiff to conduct discovery on "any financial incentives given to employees involved in denying her claim." Id. Citing Glenn, 128 S. Ct. at 2351 , the court also ordered the insurance company to disclose its claims manuals, training and claims guidelines, and other statistical data relevant to the issue of whether it had a history of biased claims administration.

In McGahey v. Harvard Univ. Flexible Benefits Plan, 260F.R.D. 10, 12 (D. Mass. 2009), the court granted the plaintiff's motion pursuant to Fed.R.Civ.P. 56(f) to conduct discovery directed to the issue of whether the insurance company's experts "were in fact truly independent":

The court is persuaded that McGahey is entitled to some additional discovery related to the opinions of three experts who performed independent medical examinations (IME) for Harvard. . . . These three experts stand apart from the other fourteen experts and treating physicians who opined on McGahey's disability because of their diagnosis of McGahey with "symptom magnification," or, less politely, faking. At least on the surface, this opinion is difficult to reconcile with the diagnoses of McGahey's treating physicians who are sufficiently convinced of her complaints of pain to prescribe powerful (and potentially addictive) painkillers like OxyContin and Percocet.

Relying solely on the language of the Group Policy (Filing No. 23-4), Coventry maintains that it has discretionary authority in determining eligibility for benefits because "[t]he Group Policy expressly grants Coventry the `sole and absolute discretion' to determine whether procedures and treatments are experimental or investigational. . . . Therefore, the abuse of discretion standard applies, and matters outside the Administrative Record are not subject to discovery."

As discussed in footnote 3, supra , Coventry has not identified all of the applicable ERISA plan documents, yet demands a deferential standard of review. The "administrative record" upon which it relies is neither organized nor authenticated. Coventry's redactions of its "administrative record" have not been authorized or explained.

As required by the Eighth Circuit's decision in Chronister, 563 F.3d at 775 , this court has considered the facts of the case at issue, the nature of the claim, and Coventry's structural conflict of interest. Coventry has insisted that the course of treatment advocated by the plaintiff's treating physicians is "experimental" simply because the plaintiff suffers from a rare condition for which there is no standard of care. In arguing that the proposed treatment was not "necessary," Coventry's consulting experts suggested alternative treatments which would also be "experimental" under whatever guidelines they used to interpret Coventry's policy language. Coventry's own consulting experts appear to agree that it would be reasonable to pursue the course of treatment proposed by the plaintiff's treating physician.

Coventry acknowledges, as it must, that it has a structural conflict of interest. Under Glenn, "not all conflicts are created equal." Hogan-Cross v. Metropolitan Life Ins. Co., 568 F. Supp. 2d 410 (S.D.N.Y. 2008).

Their significance in any given case depends upon all of the circumstances, including those suggesting a higher or lower likelihood that the conflict affected the decision. Information bearing on the manner in which a conflicted plan administrator compensates outside consultants could be highly pertinent. Maintenance of compensation arrangements that create economic incentives for consultants to recommend denial or termination of benefits would have a material bearing on the likelihood that the administrator's conflict affects its benefit determinations.

ORDER

IT IS ORDERED that plaintiff's Motion for Leave to Conduct Discovery Outside the Administrative Record (Filing No. 25) is granted, as follows:

1. The plaintiff may serve discovery requests to obtain information concerning the topics proposed in the plaintiff's motion, consistent with the numerical limits imposed by the applicable Federal Rules of Civil Procedure.

2. If any document is withheld from production or disclosure on the grounds of privilege or work product, the producing party shall disclose the following information about each such document withheld: a description of the document withheld with as much specificity as is practicable without disclosing its contents, including (a) the general nature of the document; (b) the identity and position of its author; (c) the date it was written; (d) the identity and position of its addressee; (e) the identities and positions of all persons who were given or have received copies of it and the dates copies were received by them; (f) the document's present location and the identity and position of its custodian; and (g) the specific reason or reasons why it has been withheld from production or disclosure.

A party may object to this order by filing an "Objection to Magistrate Judge's Order" within 14 days after being served with the order. The objecting party must comply with all requirements of NECivR 72.2.