Schwartz v. Oxford Health Plans, Inc., 175 F. Supp. 2d 581 (S.D.N.Y. 2001)

U.S. District Court for the Southern District of New York - 175 F. Supp. 2d 581 (S.D.N.Y. 2001)
June 11, 2001

175 F. Supp. 2d 581 (2001)

Elizabeth J. SCHWARTZ, Plaintiff, v. OXFORD HEALTH PLANS, INC. and Oxford Health Plans (N.Y.), Inc., Defendants.

No. 99 CIV 3369 DC.

United States District Court, S.D. New York.

June 11, 2001.

Kravet & Vogel, LLP, New York City by Joseph A. Vogel, for Plaintiff.

*582 Morrison Cohen Singer & Weinstein, LLP, New York City, by Arthur Ciampi, Kieran X. Bastible, for Defendants.


CHIN, District Judge.

Plaintiff Elizabeth Schwartz, a cancer patient since 1991, is insured by defendants Oxford Health Plans, Inc. and Oxford Health Plans (N.Y.), Inc. (together, "Oxford"). She commenced this action pursuant to the Employee Retirement Income Security Act of 1974, as amended ("ERISA"), to recover medical insurance benefits for treatment she received at Memorial Sloan-Kettering Cancer Center ("Sloan-Kettering"), as well as a declaratory judgment clarifying the extent to which she may obtain reimbursement for her ongoing treatment expenses. Having settled plaintiff's claims for treatment expenses incurred on or before January 31, 2000, the parties have stipulated to a summary bench trial on plaintiff's "going forward" claimsthat is, the claims for treatment expenses incurred after January 31, 2000.

The principal issue before the Court is the standard Oxford will use to determine the amount of money it will reimburse Schwartz for her treatment expenses at Sloan-Kettering. Plaintiff contends that Oxford should be required to reimburse her for the full amount of Sloan-Kettering's charges (subject to the deductible and coinsurance provisions of her health plan), unless Oxford demonstrates that such charges are significantly greater than the amounts regularly charged by other New York cancer centers for the same procedures and medications. In contrast, Oxford maintains that it need only process Schwartz's claims in accordance with its present procedures.

For the reasons set forth below, judgment will be entered in favor of Schwartz. The following constitute my findings of fact and conclusions of law, pursuant to Fed.R.Civ.P. 52(a).

  FINDINGS OF FACT  A. Plaintiff's Illness and Medical Treatment

Schwartz, an attorney, has worked for the law firm Paul, Weiss, Rifkind, Wharton & Garrison ("Paul Weiss") since 1989. (Affidavit of Elizabeth J. Schwartz ("Schwartz Aff."), dated May 12, 2000, ¶ 5). In 1991, she was diagnosed with cancer and, since then, has regularly received medical treatment at Sloan-Kettering, a comprehensive cancer center of national renown.[1] (Id. ¶¶ 2, 5). With the exception of a four-month medical leave of absence in 1994, Schwartz has continued to work, on a "substantially full time basis," at Paul Weiss. (Id. ¶ 5).

The treatments needed to stabilize Schwartz's cancer can only be delivered effectively at a comprehensive cancer center such as Sloan-Kettering. (Id. ¶ 2). There are four comprehensive cancer centers in New York: Columbia Presbyterian, New York University Hospital, Albert Einstein College Hospital, and Sloan-Kettering. (Id.; Affidavit of Joseph A. Vogel ("Vogel Aff."), dated May 15, 2000, Ex. 18). Unlike Sloan-Kettering, each of the three other comprehensive cancer centers in New York is a contracted facility within Oxford's network of providers and thus accepts discounted fees from Oxford for the services it performs. (Supplemental *583 Affidavit of James Kilcoyne ("Kilcoyne Reply Aff."), dated May 26, 2000, ¶ 4).

  B. Plaintiff's Insurance Coverage

Throughout the period of plaintiff's employment, Paul Weiss has provided a group health insurance plan for its employees. When Schwartz was first diagnosed with cancer in 1991, the Paul Weiss group health insurance plan was underwritten and administered by Connecticut General Life Insurance Company ("Connecticut General"). (Schwartz Aff. ¶ 6). Thereafter, in 1995, Paul Weiss switched to a plan offered and administered by Oxford. (Id.).

The Oxford insurance plan, also known as the "Freedom Plan," offered employees a choice between managed care through in-network HMO providers and indemnity insurance through out-of-network providers.[2] Because Sloan-Kettering was an out-of-network provider, Schwartz decided to enroll in the Freedom Plan's "out-of-network" option so that she could continue her treatment at Sloan-Kettering with the oncologists who had been monitoring and treating her cancer since the initial diagnosis. In doing so, she "understood that ... [she] would not be receiving medical care at the lower costs offered through Oxford's `in network' HMO doctors." (Id.). Rather, she concluded, based upon her reading of the Freedom Plan, that the Freedom Plan coverage for Sloan-Kettering's cancer treatment charges would be equivalent to the indemnity coverage she had previously received from Connecticut General. (Id.).

  C. The Freedom Plan

The terms of the Freedom Plan are set forth in the Oxford Member Handbook and Supplemental Certificate of Coverage ("Handbook"). (Vogel Aff., Ex. 20). Under these terms, Schwartz's non-network treatment at Sloan-Kettering was covered as follows: subject to specified deductibles and coinsurance, Oxford would fully reimburse the charges to the extent they did not exceed the "usual, customary and reasonable" ("UCR") rates for such services and procedures. (Handbook at 8-9; Affidavit of James Kilcoyne ("Kilcoyne Aff."), dated May 11, 2000, ¶ 4). Once Schwartz satisfied a cumulative annual deductible of $300, Oxford would reimburse 80% of any further charges until an out-of-pocket maximum of $2300 per year in combined deductibles and coinsurance was met. At that point, 100% of the UCR would be reimbursed. (Schwartz Aff. ¶ 7).

The term "UCR" is addressed three times in the Handbook. First, the Handbook defines UCR as follows:

 A UCR schedule is a compilation of maximum allowable charges for various medical services. They vary according to the type of provider and geographic location. Fee schedules are calculated using data compiled by the Health Insurance Association of America (HIAA) and other recognized sources. What We Cover/reimburse is based on the UCR. For example: Our benefit is 80% of the cost of Covered Services. We will reimburse you 80% of the UCR for that service. If the charges exceed the UCR, you must pay the difference plus the 20% Coinsurance.

(Handbook at 9). Second, in a section entitled "Exclusions and Limitations," the *584 Handbook provides that "charges that are in excess of the UCR charges as determined by Us for Covered Services are excluded from coverage and are the Member's responsibility." (Id. at 17). Third, in the "Definitions" portion of the Handbook, UCR is defined as "[t]he amount charged or the amount We determine to be the reasonable charge, whichever is less, for a particular Covered Service in the geographical area it is performed." (Id. at 26).

Under the indemnity insurance provisions of the Handbook, it is the participant not the health care provider who shares a contractual relationship with Oxford. (Responsive Affidavit of Elizabeth J. Schwartz ("Schwartz Reply Aff."), dated May 25, 2000, ¶ 2). Accordingly, Schwartz as a participant enrolled in the indemnity insurance portion of the Freedom Plan is personally liable for the fees charged by Sloan-Kettering if Oxford refuses to pay them. (S Reply Aff. ¶ 1; Handbook at 20). This scenario differs greatly from that of a participant who elects HMO coverage, for that individual bears no financial responsibility for the cost of services received from network providers. (Schwartz Reply Aff. ¶ 16).

The Freedom Plan indemnity coverage also differs from an HMO because Oxford shares no contractual relationships with doctors. In the Oxford HMO, the right of a network provider to payment is governed by his or her individual contract with Oxford, which generally requires the provider to accept a discounted fee as full payment. (Schwartz Reply Aff. ¶ 17; Oxford HMO Certificate of Coverage ("HMO Handbook"), attached as Ex. 35 to Vogel Aff., at 10). Furthermore, "[r]eimbursement to Network Facilities [i.e. hospitals] is based on a discounted fee-for-service basis in accordance with individually negotiated agreements." (HMO Handbook Addendum at 1).

  D. Valuation of UCR for Medical Services

Oxford determines its UCR rates based on data published by the Health Insurance Association of America in combination with the Health Care Financing Administration's Resource Based Relative Value Scale (the "RBRVS").[3] (Vogel Aff., Ex. 23). HIAA is an industry trade group for indemnity insurance carriers that has collected physician charge data throughout the United States and resold it to various insurance providers, including Oxford. (Kilcoyne Aff. ¶ 7). With respect to the medical and surgical procedures, Oxford bases its out-of-network fee schedules on information obtained from HIAA's Prevailing Healthcare Charges System (the "PHCS").[4] (Id.). The PHCS data is used despite the following disclaimer, which appears in the medical edition of the PHCS:

 The PHCS data, whether charge data or conversion factor data, are provided to subscribers for informational purposes only. Ingenix, Inc. disclaims any endorsement, approval, or recommendation of particular uses of the data. There is neither a stated nor an implied "reasonable and customary" charge (either actual or derived) .... Any interpretation and/or use of the data by the subscriber is sole[ly] and exclusively at the discretion of the subscriber.

(Vogel Aff., Ex. 7). The PHCS is mentioned nowhere in the Handbook.

*585 HIAA assembles its PHCS data from information contributed by approximately 130 indemnity insurance companies, who in turn obtain information from physicians regarding the fees they charge for various medical and surgical procedures. HIAA then collects and organizes the physician fee data by geographical region, current procedural terminology ("CPT") code, and various charge percentiles.[5] (Kilcoyne Aff. ¶ 8). Oxford calculates its UCRs based on the level of coverage purchased by a member's employer. Thus, in this case, the UCR for Schwartz's treatments at Sloan-Kettering was calculated at the 90th percentile the level of coverage purchased by Paul Weiss. In other words, under the Freedom Plan, Schwartz was to receive reimbursement "at a level at which only 10% of providers would bill at or above that amount for the procedure involved." (Id. ¶ 10).

HIAA publishes at least six editions of its PHCS data, including hospital, surgical, and medical editions. (Vogel Aff., Ex. 9 at I.2). The hospital edition includes information for both inpatient and outpatient charges, which are categorized by diagnostic related group within major diagnostic category and geographic area.[6] (Id. at I.1). Despite the availability of a PHCS hospital edition, Oxford used the medical and surgical editions which only list the charges of independent physicians who perform the services in question at their offices to calculate its UCR for Schwartz's treatments at Sloan-Kettering. (Deposition of Douglas C. Anderson ("Anderson Dep."), at 12-15). Although the medical and surgical editions do not provide either hospital outpatient or pharmaceutical charge data, Oxford continued to use these editions to disallow certain Sloan-Kettering charges for outpatient treatments and chemotherapy drugs and supplies. (Anderson Dep. at 14-16; Exs. 14, 16).

  E. Valuation of UCR for Pharmaceuticals

Oxford determines the UCS for pharmaceuticals obtained from out-of-network providers in several ways. For pharmaceuticals assigned a "J code" by the federal government, Oxford reimburses out-of-network providers 110% of the wholesale price of the pharmaceutical, as determined by the "Medispan Database."[7] (Kilcoyne Aff. ¶ 13). For pharmaceuticals that have been assigned a National Drug Classification number, Oxford reimburses out-of-network providers "at 110% of the lowest observed average wholesale price" of the pharmaceutical. (Id. ¶ 14). When Oxford is unable to identify an average wholesale price, Oxford reimburses out-of-network providers at 110% of the "Medicare Fee," a measure derived from the Health Care Financing Information or a local Medicare carrier such as Empire Blue Cross of New York. (Id. ¶ 15).

None of these measures is specified in the Handbook. HIAA publishes, in addition to medical and hospital editions, a *586 PHCS pharmaceutical edition, but Oxford did not use this edition to calculate its UCR for Schwartz's medication charges. (Anderson Dep. at 16). Schwartz, as a retail consumer of chemotherapy drugs, has no access to the wholesale or "gray" market for pharmaceuticals. (Schwartz Reply Aff. ¶ 3).

  F. Oxford's Reimbursement of Sloan-Kettering Procedures

From the time Schwartz began cancer treatment in 1991, Sloan-Kettering has prepared the paperwork necessary to present reimbursement claims to Oxford. Each month, Schwartz receives a bill from Sloan-Kettering for her outpatient charges, as well as separate bills from Sloan-Kettering's physician practice groups for their professional charges. Detailed billings are always available to Oxford on request. (Schwartz Reply Aff. ¶ 6).

Throughout 1995, 1996, and the first half of 1997, Oxford routinely reimbursed Sloan-Kettering for the cancer treatment Schwartz received there. During this period, Oxford never complained that Sloan-Kettering's charges were excessive. (Schwartz Aff. ¶ 8). In the second half of 1997, however, the relatively cooperative relationship between Schwartz and Oxford took a turn for the worse. Oxford "cut back drastically" on Schwartz's Freedom Plan benefits, and simultaneously attempted to convince Schwartz to seek cancer treatment from an in-network provider. (Id.). Such a provider, she was told by Oxford, would accept Oxford's HMO fees as full payment for all treatment charges and would thus "relieve [Schwartz] of any concerns about out-of-pocket expenses." (Id.).[8] Oxford's decision to reduce its reimbursement did not correlate to any increase in Sloan-Kettering's charges. In fact, Schwartz observed no substantial increase in Sloan-Kettering's rates since 1997. (Id. ¶ 4).

An illustration of Oxford's reimbursement of Schwartz's treatment at Sloan-Kettering can be found in her records for March and May 2000. Out of a total of $8,186.26 in charges,[9] Oxford reimbursed Schwartz less than half only $4,080.96. (Vogel Aff. Ex. 33). In addition to using 110% of wholesale to calculate the amount it would reimburse for Schwartz's pharmaceuticals, Oxford unilaterally determined the dosage for which it would reimburse. For example, instead of reimbursing Schwartz for 90 milligrams of pamidronate (the dosage Schwartz received on March 10, 2000), Oxford only reimbursed Schwartz for 30 milligrams $236.27 out of a $1,079.10 charge. (Schwartz Reply Aff. ¶ 11; Ex. 33). Oxford also refused to reimburse Schwartz for any part of the oxycodone and capecitabine she received on March 17, 2000, even though both drugs along with dosage and, in the case of capecitabine, CPT code were identified in the billing statement. (Id.).

  G. Grievance Proceedings

Once Oxford reduced its level of reimbursement, Schwartz complained to the *587 benefits manager at Paul Weiss, but was unable to obtain relief. (Schwartz Aff., ¶ 9). Thereafter, she sent Oxford a letter protesting its exclusion of certain Sloan-Kettering charges from coverage and seeking complete reimbursement for the excluded amounts. In the letter, dated March 25, 1998, Schwartz itemized the various charges excluded by Oxford for the period of July to December 1997. (Vogel Aff., Ex. 22).

Oxford rejected the request by letter dated April 13, 1998, stating that the Physician Reimbursement Department had determined that Sloan-Kettering and its physicians were "correctly reimbursed" and that "no additional coverage is warranted." (Id. Ex. 23). Schwartz was advised, for the first time, that (1) Oxford reimbursed out-of-network providers 110% of the base "Average Wholesale Price" for pharmaceuticals; (2) reimbursement for chemotherapy drugs could not be made accurately because Sloan-Kettering's electronic billing system did not indicate the quantity of such drugs; (3) Oxford determined its UCRs based on data published by the HIAA in combination with the RBRVS; and (4) due to the foregoing, it would reimburse only 60% of Schwartz's cost for Sloan-Kettering's diagnostic procedures. (Id.).

Following this response, Schwartz wrote to the Grievance Committee of the Board of Directors (the "Committee") to appeal the decision and request a hearing. (Id. Ex. 24). At the hearing, held before the Committee on July 20, 1998, Schwartz presented her grievances to a three-person panel comprised of an Oxford board member, an Oxford staff attorney, and an Oxford physician. (Id. Ex. 25 at 2). None of the panel members answered Schwartz's charges, urging her instead to switch to an in-network provider. (Id.). Schwartz declined to leave Sloan-Kettering. (Schwartz Aff. ¶ 10).

By letter dated July 30, 1998, the Committee upheld the denial of additional reimbursement, explaining its decision as follows:

 After considering all of the information available to it, the Committee stands by the Grievance Review Board's assessment of payment for the claims and question and agrees that no additional payment will be forthcoming. As you chose to utilize your Freedom Plan benefits in seeking care from a non-participating provider, such treatment is subject to any applicable deductible, coinsurance, and UCR. As stated in your Oxford [Handbook], Oxford reimburses 80% of the UCR for a given procedure. If the charges exceed the UCR, the member pays the difference plus the 20% coinsurance. Accordingly, the Committee maintains that the claims were paid correctly.

(Vogel Aff., Ex. 26).

Schwartz subsequently filed a written complaint against Oxford with the New York State Insurance Department. (Id. Ex. 25). On July 1, 1999, the State Insurance Department asked Oxford to voluntarily submit to an external review of Schwartz's grievances, but Oxford refused. (Id. Exs. 28-29).

  H. The Instant Case

Schwartz commenced this suit against Oxford on May 7, 1999, asserting various claims under ERISA. The parties engaged in discovery.

Trial was scheduled to begin on February 24, 2000. On the day before trial was to start, the parties advised the Court that the case had been settled in principle. Thereafter, on February 24th, the Court issued a 30-day order dismissing the case, *588 subject to the execution of a settlement agreement.

The parties were not, however, able to finalize a settlement with respect to all of plaintiff's claims. On April 3, 2000, the Court issued an order reinstating the case.

By stipulation dated April 19, 2000, the parties settled all of plaintiff's claims for reimbursement through January 31, 2000. By a separate stipulation dated April 19, 2000, the parties agreed to the Court's resolution of plaintiff's "going forward" claims, i.e., her claims for reimbursement after January 31, 2000, through a summary bench trial. As the parties' stipulation makes clear, the parties agreed, pursuant to Acuff-Rose Music, Inc. v. Jostens, Inc., 155 F.3d 140 (2d Cir. 1998), to waive their right to a full trial. The parties further stipulated that the record for the summary bench trial would consist of the pleadings and cross-affidavits, in addition to any attached exhibits.

The parties thereafter submitted their materials to the Court for its consideration.


The parties agree that Schwartz is a "beneficiary" under an "employee welfare benefit plan" that provides "medical, surgical, or hospital care or benefits" within the meaning of ERISA. 29 U.S.C. §§ 1002(1), (8). Section 502(a) (1) (B) of ERISA provides that a beneficiary may bring a

 civil action ... to recover benefits due to [her] under the terms of [her] plan, to enforce [her] rights under the terms of the plan, or to clarify [her] rights to future benefits under the terms of the plan.

29 U.S.C. § 1132(a) (1) (B).

The parties disagree as to the standard of review to be applied. Schwartz concedes that an arbitrary and capricious standard of review applies when a court is asked to review a plan administrator's denial of benefits where, as here, the plan gives the administrator discretion to construe the terms of the plan. Schwartz argues, however, that in this case a de novo standard should apply because Oxford was operating under a conflict of interest Oxford is both the plan administrator and the party paying benefits under the plan. (Pl. Br. at 11-12, 14-15) (citing Firestone Tire & Rubber Co. v. Bruch, 489 U.S. 101, 115, 109 S. Ct. 948, 103 L. Ed. 2d 80 (1989), and Sullivan v. LTV Aerospace & Defense Co., 82 F.3d 1251, 1255-56 (2d Cir.1996)).

Oxford argues that the arbitrary and capricious standard applies. Responding to Schwartz's conflict of interest claim, Oxford argues first that Schwartz has failed to demonstrate the existence of any conflict of interest. In addition, Oxford contends that "even a conflicted insurer" is entitled to "the same deference as an independent fiduciary." (Def. Reply Mem. at 16) (quoting Wojciechowski v. Metropolitan Life Ins. Co., 75 F. Supp. 2d 256, 263 (S.D.N.Y.1999)).

The answer lies somewhere in between the parties' two positions. As the Second Circuit has held, the standard in these cases is arbitrary and capricious, but with a caveat:

 [I]n cases where the plan administrator is shown to have a conflict of interest, the test for determining whether the administrator's interpretation of the plan is arbitrary and capricious is as follows: Two inquiries are pertinent. First, whether the determination made by the administrator is reasonable, in light of possible competing interpretations of the plan; second, whether the evidence shows that the administrator *589 was in fact influenced by such conflict. If the court finds that the administrator was in fact influenced by the conflict of interest, the deference otherwise accorded the administrator's decision drops away and the court interprets the plan de novo.

Sullivan v. LTV, 82 F.3d at 1255-56; accord Whitney v. Empire Blue Cross & Blue Shield, 106 F.3d 475, 477 (2d Cir. 1997).

Here, Schwartz has shown that Oxford had a conflict of interest. Oxford was both the plan administrator as well as the entity paying the benefits. Oxford had a financial interest in denying benefits, as any claims would have to be paid out of its own assets. It also had a financial interest in encouraging beneficiaries to use in-network HMO providers, as out-of-network providers cost Oxford more money. In determining its UCR for various procedures and pharmaceuticals, Oxford had an incentive to use methodology that would reduce the UCRs at the expense of beneficiaries, as that would reduce the benefits that Oxford would have to pay. Significantly, the three members of Oxford's Grievance Committee that heard Schwartz's appeal of the denial of her benefits consisted of an Oxford board member, an Oxford staff attorney (whose "client," presumably, is Oxford), and an Oxford physician.

Consequently, I apply the two-prong approach set forth in Sullivan. I will do so twice: first, with respect to medical services, and second, with respect to pharmaceuticals.

  B. Medical Services

Oxford determined the UCR for medical services based on data contained in the medical and surgical (but not the hospital) editions of the PHCS, based on fees charged by individual physicians for services provided in their private offices rather than on the basis of fees charged by hospitals or other cancer treatment centers for outpatient services.

  1. Was Oxford's Determination Reasonable?

I conclude that Oxford's method for determining UCR in this case was not reasonable, for the following reasons.

First, the controlling language in the plan is the phrase "usual, customary, and reasonable." Once Schwartz reached a maximum of $2,300 in deductibles and co-insurance, Oxford was obligated to reimburse her for the full cost of her medical expenses, as long as they did not exceed the usual, customary, and reasonable rates charged for the services and procedures in question. Oxford did not have absolute discretion to choose how much of Schwartz's medical expenses to reimburse; rather, it was required to cover usual, customary, and reasonable charges.

Second, Oxford used data based on rates charged by individual physicians for services provided in their offices. This decision was not reasonable, for Schwartz was not receiving medical care at the hands of an individual physician in a private office. Rather, her cancer required out-patient hospital services and procedures that are usually and customarily provided at comprehensive cancer centers like Sloan-Kettering. There is nothing in the record to suggest that Schwartz could have received comparable services in the private office of an individual physician. Hence, in Schwartz's case, it would have made much more sense for Oxford to have based its determination of the UCR rates on what other comprehensive cancer centers in New York City charged for similar outpatient services. Indeed, as the other three comprehensive cancer centers in New York were part of Oxford's network, *590 their fee schedules were readily available to Oxford, and it would only have been reasonable for Oxford to look at what these other facilities charged.

Third, Oxford did not even use the most applicable PHCS data. Oxford chose to use data for individual physicians, even though there was hospital data available, at least through 1998. Even if Oxford had wanted to limit itself to PHCS or HIAA data, there was such data available for hospitals (at least until 1998), which Oxford chose to ignore.

Fourth, throughout 1995, 1996, and the first half of 1997, Oxford routinely reimbursed Sloan-Kettering for the cancer treatment Schwartz received there. During this period, Oxford never complained that Sloan-Kettering's charges were excessive or above UCR rates. These actions certainly suggest that Oxford believed Sloan-Kettering's rates were usual, customary, and reasonable.

In short, I do not believe it was reasonable for Oxford to determine that Sloan-Kettering's rates exceeded UCR rates based on a comparison of Sloan-Kettering's rates, for services provided by a comprehensive cancer center, with rates charged by individual physicians for services provided in their private offices. The rates charged by individual physicians were simply not the usual or customary rates charged by facilities comparable to Sloan-Kettering, where Schwartz could have obtained treatment and services comparable to what she received at Sloan-Kettering.

  2. Was Oxford's Determination Influenced by the Conflict?

I also conclude that Oxford's conflict of interest adversely affected its determinations.

First, throughout 1995, 1996, and the first half of 1997, Oxford routinely reimbursed Sloan-Kettering for the cancer treatment Schwartz received there. In mid to late 1997, however, Oxford started publicly announcing substantial operating losses; at that point Oxford started substantially reducing Schwartz's benefits. It also simultaneously attempted to convince Schwartz to leave Sloan-Kettering for an in-network provider. This combination of circumstances strongly suggests that Oxford's determinations were influenced by the conflict of interest presented by its dual role of plan administrator and payor of benefits.

Second, that Oxford's decisions were influenced by its conflict of interest is also demonstrated by the weakness of its position. Oxford has simply not been able to make a credible argument to justify its use of fees charged by individual physicians for services rendered in their private offices to measure the reasonableness of the fees charged for out-patient hospital services provided at comprehensive cancer centers.

Third, the grievance proceedings also strongly suggest that Oxford's decisions were influenced by its conflict of interest. At a minimum, Schwartz had a colorable grievance, as Oxford had paid her Sloan-Kettering bills without any controversy for some two-and-a-half years. Despite the meritoriousness of her claim, the Committee denied Schwartz's grievance without any meaningful attempt to explain its decision. Instead of explaining how Oxford had calculated its UCR rates, the Committee simply reiterated what was not in dispute that Schwartz's choice of an out-of-network provider meant that she was subject to a deductible and coinsurance payments. The letter was not even accurate as it ignored the $2,300 cap. More significantly, it made no effort to address the critical issue of how UCR was determined. Indeed, the Committee attempted to persuade *591 Schwartz to switch from Sloan-Kettering to an in-network provider.

  3. The Final Analysis

In the end, because I conclude that there was a conflict of interest and that Oxford's determination was influenced by that conflict, I review Oxford's decision de novo. For the reasons set forth above, I conclude that Oxford's determination of the UCR rates in this case was incorrect and inconsistent with the terms of the plan. Oxford used rates that were not usual, customary, or reasonable for the services and procedures in question. In addition, even assuming an arbitrary and capricious standard were to apply, my decision would stand: Oxford's decision to use rates of individual physicians charged for services performed in their private offices instead of rates charged for comparable out-patient hospital services provided by comparable comprehensive cancer centers in New York City was arbitrary and capricious and not reasonable.

Oxford argues that adopting the valuation methods proposed by Schwartz would "rewrite the [Handbook] and ignore the means by which HMO's across the country do business." (Kilcoyne Aff. ¶ 3). This position, however, overlooks the fact that Schwartz's Freedom Plan option for out-of-network providers is not an HMO but an indemnity insurance plan.

  C. Pharmaceuticals

Starting in the second half of 1997, Oxford reimbursed Schwartz for pharmaceuticals at 110% of the average wholesale price. As a consequence, Schwartz was only reimbursed for a fraction of her pharmaceuticals expenses.

  1. Was Oxford's Determination Reasonable?

Oxford's decision to set the UCR on the basis of 110% of the average wholesale price for pharmaceuticals was not reasonable.

First, there is nothing in the Handbook to support the use of 110% of the average wholesale price for pharmaceuticals. The Handbook does not expressly address pharmaceuticals and both sides seem to assume that pharmaceuticals are to be reimbursed (subject to the deductible and coinsurance) at a maximum of the UCR, but Oxford has not pointed to any language in the Handbook that supports its assertion that the UCR is 110% of the average wholesale price.

Second, the use of wholesale prices as a base makes no sense. Schwartz was not a wholesaler; like other beneficiaries under the plan, she was required to purchase medications at retail, not wholesale. Nor is there anything in the record to suggest that 10% above wholesale is the usual and customary price charged at retail for prescriptions. I would be surprised if retail prices were only 10% above wholesale, but Oxford, in any event, has not submitted any evidence to show that rates at 110% of average wholesale prices are usual, customary, and reasonable charges for patients who must purchase at retail. See Priority Solutions, Inc. v. CIGNA & Air France Medical Plan, No. 98 Civ. 4499(BSJ), 2000 WL 64884, **2-3, 2000 U.S. Dist. LEXIS 705, at **7-8 (S.D.N.Y. Jan. 12, 2000) (use of average wholesale prices as basis for determining health plan member's reimbursement for pharmaceuticals violated terms of plan); Priority Solutions, Inc. v. CIGNA & Price Waterhouse Health Plan, No. 98 Civ. 4336(MBM), 1999 WL 1057202, *2, 1999 U.S. Dist. LEXIS 17605, at *6 (S.D.N.Y. Nov. 8, 1999) (same).

Third, PHCS data was available for pharmaceuticals; yet, Oxford chose to ignore it. The fact that it would selectively *592 use or not use PHCS data underscores the arbitrariness of its decisions.

Fourth, Oxford also unilaterally reduced the dosage of drugs for which it would reimburse Schwartz. For example, although Schwartz received 90 milligrams of pamidronate on March 10, 2000, Oxford reimbursed her for only 30 milligrams. Oxford has not even attempted to justify these decisions, which can only be described as arbitrary.

Accordingly, I conclude that Oxford's decision to ignore the actual retail cost of pharmaceuticals, to ignore other available data regarding the cost of pharmaceuticals, and to use instead an arbitrary figure of 10% above wholesale was not reasonable.

  2. Was Oxford's Determination Influenced by the Conflict?

For the reasons set forth above with respect to medical services, I find also that Oxford's determination with respect to pharmaceuticals was influenced by its conflict of interest. I add that Oxford's selective use of data also demonstrates that its decisions were influenced by its conflict of interest; when use of the data benefitted Oxford financially, it would use it, and when use of the data disadvantaged Oxford financially, it would choose not to use it.

  3. The Final Analysis

In the end, reviewing Oxford's decision with respect to pharmaceuticals de novo, I conclude that Oxford's decision to reimburse Schwartz for only up to 110% of the average wholesale price of her medications was unreasonable and inconsistent with Oxford's obligation to reimburse her up to a maximum of usual, customary, and reasonable rates. In addition, even assuming an arbitrary and capricious standard were to apply, my decision would stand: Oxford's decision to use 110% of the average wholesale price to determine its UCR for pharmaceuticals was arbitrary and capricious and not reasonable.


For the foregoing reasons, judgment will be entered in favor of plaintiff against Oxford on plaintiff's "going forward" claims. The Clerk of the Court is directed to enter judgment declaring that Oxford's disallowance of Schwartz's claims for reimbursement of her cancer treatment was unlawful and in violation of its obligations under the Freedom Plan and requiring Oxford, subject only to the deductible and coinsurance provisions of plaintiff's Supplemental Certificate of Coverage, to reimburse her for all of Sloan-Kettering's charges for medical, surgical, and diagnostic procedures, as well as chemotherapy drugs, pharmaceuticals, and facilities, unless Oxford certifies that any such charges exceed the usual and customary rates charged by other comprehensive cancer centers in New York City for the same procedures, drugs, pharmaceuticals, and facilities administered to their similarly situated non-HMO patients,[10] subject to Schwartz's right to challenge the accuracy of Oxford's certifications. This Court will retain jurisdiction to resolve any disputes that might arise.



[1] A "comprehensive cancer center" is a medical facility designated by the National Cancer Institute as qualified to participate in nationwide clinical studies and equipped to provide the most up-to-date treatments for advanced cancer. (Vogel Aff., Exs. 17-19).

[2] Under the terms of the Oxford Freedom Plan, a "Network Provider" is a physician, hospital, or other health professional "under contract" with Oxford to provide plan participants with covered services. A "Non-Network Provider" is a "[p]rovider who has not contracted with Oxford." (Freedom Plan Summary of Benefits, Supplemental Certificate of Coverage, and Member Handbook ("Handbook"), attached as Ex. 20 to Vogel Aff., at 26).

[3] Notably, Oxford's HMO reimburses its in-network provides based on fee schedules also derived from HIAA and RBRVS charge data. (HMO Handbook Addendum at 1).

[4] The PHCS was sold by HIAA to a company called Ingenix in 1998, but is still known throughout the health care industry as the "HIAA database." (Kilcoyne Aff. ¶ 7 n. 1).

[5] CPT codes are "standardized numerical codes universally used in the health care industry, assigned by the American Medical Association to individual medical and surgical procedures for reimbursement purposes." (Kilcoyne Aff. ¶ 8).

Charge percentile refers to the percentage of providers who submit a charge above a stated amount for a particular procedure. (Id. ¶ 9).

[6] HIAA ceased publication of its hospital edition in January 1998. (Kilcoyne Reply Aff. ¶ 11).

[7] J-codes are assigned by the federal government to injectable pharmaceuticals. The Medispan Database is created by First Databank and is "used throughout the insurance industry." (Kilcoyne Aff. ¶ 13).

[8] Schwartz attributes the change in its relationship with Oxford to the confluence of two events in late 1997. At that time, doctors discovered that Schwartz's cancer had progressed further, and placed her on a more intensive regimen of treatments to stabilize her condition. (Schwartz Aff. ¶ 8). This development coincided with Oxford's public announcement of substantial operating losses in October 1997. (Id.; Vogel Aff., Ex. 21 (copies of press reports)). As a consequence, Oxford began to seek a substantial reduction in its Freedom Plan reimbursement expenses and thus cut Schwartz's benefits. (Schwartz Aff. ¶ 8).

[9] For reasons unclear to the Court, Schwartz states that the charges for that billing period totaled $8295.26. (Schwartz Reply Aff. ¶ 11).

[10] Oxford may not use the reduced rates it pages to the other comprehensive cancer centers as a basis for determining the UCR; it must rely on the fees charged by the centers to patients for whom the centers are not "in-network." In other words, Oxford may not use the HMO methodology to determine the fees for a participant in an indemnity insurance plan.