Rozema v. the Marshfield Clinic

977 F. Supp. 1362, 1997 U.S. Dist. LEXIS 15258, 1997 WL 611783

• Court: District Court, W.D. Wisconsin
• Decided: October 2, 1997
• Docket: 96-C-592-C, 96-C-916-C and 96-C-730-C
• Status: Published

Opinion

ORDER AND OPINION

CRABB, District Judge.

This is a civil antitrust action for monetary, declarative and injunctive relief brought pursuant to the Sherman Act, 15 U.S.C. § 1, and Wis. Stat. §§ 133.03 and 133.14. Plaintiffs contend that defendants violated the federal and state statutes by entering into a continuing contract, combination or conspiracy to divide the market for all physician services within an eight-county area in north central Wisconsin. Plaintiffs have reached a preliminary settlement agreement with defendants North Central Health Plan of Wisconsin and Rhinelander Medical Center, S.C. Now before the court is the summary judgment motion of defendants Marshfield Clinic and Security Health Plan of Wisconsin, Inc.

The plaintiff class consists of all purchasers of physician services from defendants residing in the eight-county area covering Clark, Price, Lincoln, Oneida, Marathon, Taylor, Portage and Wood Counties and purchasing physician services in that area after July 24, 1992. Plaintiffs assert that defendants and other co-conspirators engaged in a single conspiracy to divide the market for all physician services and that this conduct led to supra competitive prices for physician services in the eight-county class area. Defendants challenge plaintiffs’ market allocation claim on the following grounds: plaintiffs who are HMO subscribers lack antitrust standing because they purchase physician services directly; plaintiffs’ evidence does not show that alleged co-conspirators Rice Clinic and Wausau Medical Center participated in a market allocation scheme that constituted a per se violation of the Sherman Act; and plaintiffs’ evidence that supra-competitive prices resulted from the challenged conduct is insufficient because it is not tied to properly defined markets, fails to account for the effects of legal conduct and relies on impermissible economic assumptions.

I conclude that defendants’ motion will be denied in all but one respect: plaintiffs have not adduced evidence from which a jury could conclude that Rice Clinic participated in a conspiracy to allocate markets. Plaintiff HMO enrollees are not deprived of standing under the direct purchaser rule because, whether they bought physician services in indirect or direct form, they did so from a defendant co-conspirator. The evidence implicating Wausau Medical Center in the market allocation scheme includes the same documents the Court of Appeals for the Seventh Circuit held supported a finding of market allocation in Blue Cross & Blue Shield v. Marshfield Clinic, 65 F.3d 1406, 1416 (7th Cir.1995), cert denied, — U.S. -, 116 S.Ct. 1288, 134 L.Ed.2d 233 (1996). Finally, plaintiffs’ expert evidence of supra-competitive prices, although not compelling, is sufficient to support a jury finding that defendants’ challenged conduct caused an antitrust injury. It is supported by evidence of properly defined markets, is not invalidated by evidence of legal, non-disaggregated price ef fects and is based on permissible assumptions.

From the parties’ proposed findings of fact, I conclude that the following facts are not in dispute.

UNDISPUTED FACTS

Defendant The Marshfield Clinic is a physician-owned, non-profit multi-specialty clinic based in Marshfield, Wisconsin. It has more than 400 physicians who practice at its main facility in Marshfield and its regional centers in 14 counties throughout north central and north western Wisconsin. The defendant clinic has a reputation for high quality. It has entered into various vertical and horizontal relationships with other health care entities in order to establish a regionally integrated system of health care. Defendant Security Health Plan of Wisconsin, Inc. is a Wisconsin health maintenance organization that is wholly owned and controlled by defendant The Marshfield Clinic. Defendant North Central Health Protection Plan is a health maintenance organization with its principal place of business in Wausau, Wisconsin. Defendant Rhinelander Medical Center, S.C. is a not-for-profit corporation organized under Wisconsin law with its principal place of business in Rhinelander, Wisconsin.

A. Market Allocation

1.Freeflow agreement background,

Beginning in 1978 and continuing through the present, defendants Marshfield Clinic and North Central Health Protection Plan have maintained a “free flow” agreement permitting enrollees in defendant Security Health Plan’s Greater Marshfield HMO and the North Central HMO to see participating physicians in either plan without first obtaining a referral. Regardless who provides services for an enrollee, the contract in force for that enrollee determines the payment received by the treating physician for that service. In the year ending June 30,1982, 1,194 North Central patients were seen by Greater Marshfield physicians and 5,584 Greater Marshfield patients received services from North Central physicians. During the early years of the agreement, North Central required its participating physicians to sign an agreement of joint affiliation with the Greater Marshfield HMO. At a 1981 meeting of the North Central board of directors, it was noted that an advantage of the free flow agreement is that physicians are reimbursed directly for care provided to enrollees of the other plan. Without the agreement, physicians would receive lesser payments for referral activity.

2. Limits of freeflow coverage

In 1985, a group of North Central affiliated oral surgeons from the Wausau area sought to establish a practice in Marshfield. Defendant Marshfield Clinic took the position that the free flow agreement “did not support that activity” and refused to treat the practice as an affiliated provider. Put another way, the Greater Marshfield HMO refused to cover treatment provided its enrollees in Marshfield by those Wausau physicians. Ultimately, the oral surgeons did not set up a practice in Marshfield.

Before 1993, several rheumatologists employed by defendant Marshfield Clinic practiced in Wausau and were reimbursed by North Central for treatment of its enrollees. In 1993, the Wausau Medical Center hired its own rheumatologist and North Central determined that the Marshfield physicians would no longer be reimbursed as affiliates of the North Central Plan. The Marshfield Clinic rheumatologists continue to see patients in Wausau a few days a month on an outreach basis but the Greater Marshfield Plan does not promote rheumatological services in Wausau as a satellite office. Also in 1993, North Central began marketing its plan in Taylor County. Because the benefits of the free flow agreement with Greater Marshfield were not available in that area, North Central had a more difficult time marketing its plan there.

3. Marketing plans outside service area service area expansion

As part of the free flow arrangement, North Central and Marshfield had an unwritten agreement not to market in each others service areas. At the time free flow was established, the parties decided not to draw up dear descriptions of their respective service areas in order to minimize antitrust risks. However, the two plans competed in expanding them service areas, each seeking to be the first to establish a presence in formerly neutral territories by affiliating with established local providers. When the free flow agreement was formalized in 1982, North Central terminated its existing affiliations with all non-Marshfield providers located outside its service area of Lincoln, Langlade and Marathon Counties. In a letter to a terminated provider, North Central’s executive directory explained that the purpose of the action was “to further define the geographic service area for marketing, provider relations, and plan control.”

In 1985, a consultant retained by North Central to review its operations observed that the North Central Health Protection Plan had been “relatively untouched by outside'competition” historically, but that considerable fear existed in the community that the Greater Marshfield plan could expand and reduce North Central’s service base. The consultant noted that the efforts at controlling costs within North Central had not been as rigorous as within a “true HMO.” In December 1985, the North Central board of directors met and discussed the advantages and disadvantages of expanding its service area. A handout prepared for that meeting listed as disadvantages to expansion the encouragement of “friendly” competition with Marshfield and the promotion of price competition.

Beginning in 1987, North Central made several efforts to expand its service area, with limited results. That year, North Central attempted to expand to Oneida County by seeking to enlist the Rhinelander Medical Center as a participating provider. The Rhinelander Medical Center declined the invitation to affiliate with North Central because the medical center had recently established its own HMO through Security Health Plan of Wisconsin and did not want to compete with itself in Oneida County. North Central continued its efforts to establish a presence in Oneida County and was finally successful in 1995, after the Rhinelander Medical Center agreed to become a participating provider.

Also beginning in 1987, North Central explored expanding its service area to Portage County. The plan contacted a group called the Portage County HMO Search Committee, but was told that the committee was discussing setting up an HMO through Security Health Plan and had'agreed not to negotiate with any other HMO’s during these discussions. The discussions did not bear fruit, however. Security Health Plan was not approved to market in Portage County until 1995. Again in 1994, a North Central attempt to expand to Portage County stalled when the Rice Clinic, located in Stevens Point (Portage County), refused to join as an affiliated provider. Eventually, North Central began marketing in Portage County in 1995.

4. Wausau Medical Center

The Wausau Medical Center, in Wausau, Wisconsin (Marathon County), is an affiliate provider for the North Central Health Protection Plan and therefore a joint affiliate provider for Greater Marshfield under the free flow agreement. Wausau Medical Center and Security Health Plan formed a Medicare supplement HMO called “Security 65” that was introduced in the Wausau area in 1987. Speaking to a meeting of the Marathon County Medical Society in January 1987, Security’s medical director stated that the plan was not designed to draw patients from Wausau providers. Only patients requiring tertiary care were to be referred to the Marshfield Clinic, and only with the approval of the plan’s director.

As part of the arrangement establishing the Security 65 plan, the Wausau Medical Center agreed not to participate in any similar competing programs, including the “65 Plus” plan established subsequently by North Central. The Wausau Medical Center remained a participating provider for North Central’s general HMO. The provider listing for the Security 65 plan did not include the Marshfield Clinic’s regional centers in Marathon County or the outreach rheumatology services performed by Marshfield in Wausau. Certain affiliated providers from outside Wausau were limited to locations at which they could provide treatment under the plan.

In 1987, officials of the Marshfield Clinic and the Wausau Medical Center met and discussed numerous subjects, including physician recruitment, the medical center’s willingness to enter into arrangements with other HMO’s, physician and group management and compensation plans. An internal memorandum summarizing the meeting drafted by Marshfield Clinic official Robert De Vita included the following:

We agreed I should respond to any Met-Life representatives who call us. MetLife talked with the Medical Center. This is a very complicated issue. Entails a third HMO besides NCHPP [North Central] and Security Sixty-Five. The question is still open as to whether the Medical Center would agree to expanding Security Health Plan to sell group business in the Wausau Region. We should decide whether we want to do this.

The memorandum indicated that the parties agreed to discuss how they should deal with the University of Wisconsin-Wausau Hospital relationship. “We acknowledged that one way would be to enter into projects which provide the same thing except with the value added of Marshfield Clinic. That is, services are provided locally by subspecialties.”

The relationship between the Marshfield Clinic and the Wausau Medical Center was discussed in a March 1989 meeting of Marsh-field’s executive committee. Minutes from that meeting note that the following questions were raised:

• Should we continue our current cooperative efforts with the Wausau Medical Center?

• Should we provide oncology services to Wausau via the Wausau Medical Center?

• Should we begin a single specialty group in Wausau?

• Should we take a competitive stand with Wausau, in general?

• What services should we provide? Part of the discussion alluded to the fact that if we do place certain specialists in Wausau, this would take patients from Marshfield where these specialists ordinarily refer to our own surgeons, radiology, and other services. This poses a significant problem.

Goal — We should continue our current strategy in Wausau working with the Wausau Medical Center and the insurance plans in the area.

At the same meeting, the executive committee agreed that Marshfield’s primary competition came from Minneapolis, the Mayo Clinic, Madison and Milwaukee.

On October 19, 1994, almost two months before the trial began in Blue Cross (see below), representatives of the Marshfield Clinic, the Wausau Medical Center and the Rhinelander Medical Center met.

5. Rhinelander Medical Center

As stated above, the Rhinelander Medical Center and Security Health Plan formed the Security Northcare HMO in Oneida County in 1986. The parties’ agreement did not prevent Rhinelander from affiliating with other HMO plans. Initially, the medical center agreed to serve as a provider for North Central as well. However, to avoid competing with itself in the Security Northcare plan, in 1986, the Rhinelander Medical Center decided not to participate with North Central in Oneida County. Physicians practicing out of the Rhinelander Medical Center’s satellite clinic in Tomahawk, Wisconsin, in Lincoln County did become affiliated providers with North Central. As a result, the Tomahawk office was listed as a joint affiliated provider for the Greater Marshfield plan under the free flow agreement but the Rhinelander Medical Center was not. In 1994, the Rhine-lander Medical Center agreed to affiliate fully with the North Central Health Protection Plan, which began marketing in Oneida County in 1995.

At a meeting of the Marathon County Medical Society in January 1987, Security’s medical director explained that under the Security Northcare plan, primary care was to be provided in the Rhinelander area with tertiary care referrals to the Marshfield Clinic by approval only. In a 1991 meeting of officials of the Marshfield Clinic and the Rhi nelander Medical Center, the recent expansion of Marshfield’s offices in Woodruff (Oneida County) was discussed. In a subsequent internal memo, a Marshfield official who attended that meeting described the discussion as follows:

We said that our relationship with the Rhinelander medical group is very valuable to us. We do not want to see ourselves “knocking heads” for the same services for the same patient population. I indicated Marshfield Clinic’s northern region market is north and west, while [Rhinelander Medical Center’s] market is north and east. We informed him our increase in size was not just attributable to primary care but more so to specialty care. I also mentioned that we do not market actively in the Rhinelander area and appreciated that the Rhinelander Medical Center didn’t actively market in the Lakeland area.

6. Ministry Corporation

The Ministry Corporation owns St. Joseph’s Hospital in Marshfield, St. Michael’s Hospital in Stevens Point, St. Mary’s Hospital in Rhinelander, Sacred Heart Hospital in Tomahawk and, jointly with Marshfield Clinic, Flambeau Hospital in Park Falls. Dr. Wesbrook of the Marshfield Clinic serves on the board of directors of the Ministry Corporation. In an April 22, 1987 letter to the president of Ministry Corporation, the president of the Marshfield Clinic discussed a number of possible cooperative efforts between Ministry-owned hospitals and the different Security HMO’s. The president proposed an arrangement in Rhinelander in which “Sacred Heart-St. Mary’s Hospital would take a position on the Operating Committee in return for financial arrangements between Security Health Plan and Sacred Heart-St. Mary’s Hospital for the North-care service area.” Regarding the Stevens Point area, Marshfield’s president proposed the following:

The third step would be an arrangement between St. Michael’s Hospital and the Stevens Point area physicians and Security Health Plan. We suggest the possibility' of patterning that proposal after the arrangement for Rhinelander. Obviously we are dealing with different types of corporations in each of these cases so that the arrangements might be somewhat different. As you know, there is a separate corporation in Portage County, developed by St. Michael’s and the physicians, to deal with organizations such as Security Health Plan. The common denominator is the fact that the plans will be managed by the Marshfield Clinic and all of the hospitals involved are part of the Ministry Corporation.

In the late 1980’s, the Marshfield Clinic and St. Joseph’s Hospital reached an agreement whereby, during certain hours of the day, the clinic would provide urgent care treatment and the hospital’s emergency room would treat only true emergency cases. This arrangement addressed an overcrowding problem at the emergency room.

In 1991, the consulting firm Ernst & Young prepared a strategic plan for Ministry Corporation. A draft synopsis of the plan viewed by officers of the Marshfield Clinic included the following passages:

Another key factor is elimination of competition among SSM-MC [Sisters of the Sorrowful Mother-Ministry Corporation] hospitals and key physician allies.

Core Strategy: Major strategic alliance with Marshfield Clinic

• Create a formal vehicle for SSM-MC/Clinic joint marketing/program planning and execution.

• Support the major strategies of other partner.

• Work toward selected joint ventures.

• Agreement to not harm each other.

• Provide rights of first refusal on program/market initiatives.

Basic Principles

* * * * * *

• There should be recognition of interdependence. Inappropriate competition should be reduced.

• SSM-MC can influence Marshfield Clinic relationships/behaviors in key local markets and with key MD groups.

Beginning in 1991, officials of the Marsh-field Clinic and Ministry Corporation began meeting regularly to determine ways in which they could cooperate to produce a “seamless system of care” in which hospitals, clinics and doctors would work together. Also in 1991, the president of Ministry wrote a letter to the president of the Rhinelander Medical Center proposing a cooperative effort to expand cardiology services in Rhine-lander, relying on the Marshfield Clinic as the “key provider of tertiary clinical expertise in the accomplishment of this objective.” The letter included the following statement: “We believe a long-term relationship between Ministry Corporation and the RMC is essential to achieving quality health care in the Rhinelander-Tomahawk Community for the future.” A copy of this letter was forwarded to officials of the Marshfield Clinic, along with a cover memorandum in which Ministry’s president assured Marshfield that “we see you as our partner when it comes to clinical aspects.” Later in 1991, Ministry Corporation, the Marshfield Clinic and the Rhinelander Medical Center reached an arrangement whereby Ministry lent Rhinelander $4.5 million for building expansion costs, Rhinelander rented to Ministry the newly constructed space for radiation therapy and Marshfield recruited and employed the physicians providing radiation therapy and cardiology services at the facility.

7. Rice Clinic / Portage County

In 1987, a letter of intent was drawn between defendant Marshfield Clinic and Portage County physicians for “new plan development.” A new HMO did not result from that letter. In approximately 1993, Marsh-field Clinic began working with the Central Wisconsin Health Alliance, a group of employers from the Stevens Point area attempting to organize a network of health care for their employees. As of September 1994, a system had not been developed but the Alliance and Marshfield had entered into a letter agreement whereby Marshfield would help develop the system and provide services at a discount. Marshfield had no control over which other providers would be signed up for the network Also involved with the alliance was Dr. Tom O’Malley of the Rice Clinic.

Rice Clinic is located in Stevens Point, Wisconsin, in Portage County. At a June 1992 meeting of Marshfield Clinic and Ministry Corporation officials, discussion of the Rice Clinic followed directly after discussion of the Rhinelander Medical Center radiation therapy and cardiology project. At an August 1992 meeting of Ministry, Marshfield and Rice Clinic representatives, Ministry offered to build a facility for Rice Clinic adjacent to St. Michael’s Hospital. At a similar meeting in November 1992, a representative from Ministry stated that his corporation’s offer to fund the construction was conditioned on a strong affiliation with the Marsh-field Clinic. Notes from the November meeting show that the Ministry representative informed the group that “the Ministry Corporation’s strategic plan called for an alliance with Marshfield Clinic and plans to develop this relationship insofar as possible where the Ministry Corporation has its hospitals.” The minutes from a September 1993 meeting of Ministry and Marshfield officials reflects that “[b]uild[ing] relationships with Rice Clinic and independents” was considered a Marshfield priority and “[b]uild[ing] linkages with Rhinelander Medical Center” was a priority for both Marshfield and Ministry. A list of priority projects composed for discussion at that September meeting listed under the heading for Wausau: “Building relationships with local medical community (impacts Stevens Point, Medford, Kronenwetter and Rhinelander).”

In December 1994, Rice Clinic and Marsh-field Clinic entered into an agreement whereby Marshfield Clinic provided certain management and consultative services to Rice Clinic for a fee. One reason Rice Clinic entered into the agreement was its belief that Marshfield would help establish a better working relationship between Rice Clinic and St. Michael’s Hospital. In discussions leading up to the management services agreement, the parties considered the option of Marshfield Clinic’s purchasing the Rice Clinic. Because preparing for such a transaction would require extensive work by Marshfield’s legal and accounting staff and because that staff was occupied with the Blue Cross litigation at the time, Marshfield indicated that it could not commit to such a proposal until January 1995. Rice Clinic’s agreement to purchase management services from Marsh-field was seen as a step toward acquisition.

Under the management services agreement, Marshfield’s director of operations, Dennis Blanchard, devoted one day a week to providing administrative and managerial services to Rice Clinic. The agreement provided for a consultation committee and a Marsh-field Clinic administrative team that worked directly with Rice Clinic Staff. Rice Clinic asked Blanchard to evaluate a wide range of management functions, including information systems, computerized patient records, personnel and staffing policies, medico-legal issues and marketing and public relations. Additionally, Blanchard and Marshfield were asked to advise Rice Clinic in its efforts to build new facilities. Rice Clinic’s on-site manager had a reporting relationship with Blanchard and attended Marshfield Clinic management meetings to observe its administrative practices and resources. Rice Clinic terminated the management services agreement in January 1996.

At a February 7, 1994 joint meeting of the Rice Clinic stockholders and the Rice Real Estate and Equipment Management Committee, Dr. Bergin of Rice Clinic instructed physicians to assess the patient care services offered at Rice Clinic and apprise him of services offered at Marshfield Clinic that could be integrated at Rice Clinic, as well as the Marshfield physicians performing them. At the same group’s December 15, 1994 meeting, a discussion whether the Rice Clinic would affiliate with North Central’s HMO included mention that Wausau Medical Center and Marshfield Clinic were providers under the plan. In 1995, Rice Clinic adopted a vision statement containing the following passage:

The Rice Clinic is the key participant within a system of integrated health care that includes St. Michael’s Hospital and the Marshfield Clinic. Our cohesive integration with other participants is philosophical, functional and physical.

In 1995, defendant Security Health Plan was authorized by the Wisconsin Office of the Commissioner of Insurance to market in all of Portage County. Before this time, Security’s approved marketing area included only small portions of western Portage County. In 1996, Rice Clinic affiliated with Security as a provider for its Greater Marshfield HMO.

8. Clinic locations

The provider listings for the Greater Marshfield Health Plan and the North Central Health Protection Plan list no North Central affiliated physicians practicing in Marshfield and no Marshfield Clinic physicians practicing in Wausau. The Greater Marshfield listing does include radiologists, numerous non-physician services (chiropractic, mental health, oral surgery, etc.) and outreach services in Wausau. The Wausau Medical Center does not employ any physicians in Marshfield and, other than outreach rheumatologists discussed above, the Marsh-field Clinic employs no physicians in Wausau. The Marshfield Clinic has regional centers in Stevens Point (Portage County), providing oral and maxillofacial surgery, and Rhinelander (Oneida County), providing neurology and radiation therapy. Rice Clinic, of Stevens Point, employs no physicians north and west of Portage County and employs no oral surgeons.

9. Barriers to entry

Physicians employed by The Marshfield Clinic control the credentials committee at St. Joseph’s Hospital in Marshfield and only Marshfield-employed physicians have practice privileges there. Several independent physicians have requested and been denied privileges to practice at St. Joseph’s. In the eight-county area, physician referral networks are heavily dependent on HMO affiliation. Independent physicians have had difficulty gaining affiliation with Security’s Northcare HMO. The Marshfield Clinic has attained a reputation among consumers as the main tertiary care facility in north central Wisconsin. Physicians who practice independently in rural areas require cross-coverage or someone who will treat their patients when they are not available. Physicians have a more difficult time obtaining cross-coverage in north central Wisconsin than they do in the rest of the state.

B. Plaintiffs’ Expert Evidence

1. Market definition

In their expert reports, both Dr. Freeh and Dr. Beyer define a product market that includes all physician services. Dr. Freeh defines three distinct geographic markets within the eight-county class area outlined in the complaint: a five-county market including Oneida, Price, Taylor, Clark and Wood counties; a two-county market including Lincoln and Marathon Counties; and a one-county market consisting of Portage County. Dr. Beyer defines a single thirteen-county geographic market for physician services that includes the entire eight-county class area. None of plaintiffs’ experts defines a geographic market for HMO services.

2. Market share

Dr. Freeh computed market share on the basis of the number of doctors and charges for physician services in each of his three geographic markets. From the number of doctors, he concluded that defendants’ and their co-conspirators’ market share was 63.4 percent in Lincoln and Marathon Counties, 70 percent in Oneida, Price, Taylor, Clark and Wood Counties and 44.1 pei'cent in Portage County. From charges for physician services, Dr. Freeh concluded that defendants’ and their co-conspirators’ combined market share was 72.6 percent in Lincoln and Marathon Counties, 85.4 percent in Oneida, Price, Taylor, Clark and Wood Counties and 64.8 percent in Portage County. Dr. Freeh considered the group of defendants and co-conspirators to include The Marsh-field Clinic, Wausau Medical Center, Rhine-lander Medical Center, North Central Health Protection Plan and Rice Clinic. Dr. Freeh computed the market share of each defendant or co-conspirator in all three markets. Neither of plaintiffs’ experts considered alternative forms of health care financing as competitors of defendants and their co-conspirators in computing market power.

3.Injury

Drs. Freeh and Beyer each concluded that class members paid increased prices for physician services as a result of defendants’ and the co-conspirators’ anti-competitive conduct. Dr. Freeh concluded that defendants’ conduct had a common impact on all class members. Dr. Freeh reached his conclusions without independently analyzing the prices charged by defendants. Dr. Beyer compared the prices charged by defendant Marshfield Clinic with those charged by the Mayo-Midelfort Clinic in Eau Claire, Wisconsin, a clinic Beyer determined was similar to Marshfield in size, variety of care offered and relative market position. Marshfield charged higher prices than Mayo-Midelfort Clinic for 39 of the 43 medical procedures looked at by Beyer, with an average higher charge of 8.5 percent. Additionally, Beyer compared the premiums charged by defendant Security Health Plan with those charged by Compcare (an HMO subsidiary of Blue Cross) in southeastern Wisconsin; Security’s premiums were higher by an average of 15 percent.

Plaintiffs submitted the testimony and expert report of Dr. Robert Tollison to prove injury and damages. In his expert report, Tollison calculated damages from defendants’ alleged market allocation of the eight-county class area, which he referred to as the “North Central Wisconsin market.” Tollison analyzed records covering 50 million different physician service transactions of Blue Cross subscribers in Wisconsin between 1988 and 1996. For each class of medical procedure offered, he calculated an average, or indexed, price across that time period. Using these indexed prices as his base, Tollison calculated per capita physician charges both inside and outside the eight-county region and concluded that the average annual per capita charge for the class area was $58 (or 6.9%) higher than for the rest of the state. Of this amount, Tollison attributed $26 to increased utilization of physicians (more total visits or more visits involving expensive procedures) and $32 to higher prices for comparable procedures. Before reaching these figures, Tollison performed a regression analysis on the data that tracked, among other things, population density, per capita income, market structure, age and sex. This regression analysis included a HHI (Herfendahl-Hirshman Index) variable that measures the relative distribution of market shares within each county. Tollison did not address any conduct or characteristics specific to defendant Marsh-field Clinic, but rather assessed all providers within the class area equally.

Additionally, Tollison concluded that HMO subscribers of defendants Security Health Plan and North Central Health Protection Plan paid increased premiums that were directly attributable to the increased cost of physician services in the class area. To determine the amount the premiums were increased, Tollison calculated what percentage of each plans premiums went to physician services costs, on average, between 1988 and 1996. Tollison assumed that if the cost for physician services had been 6.9% less for each year, that savings would have been passed directly on to the consumer.

In his expert report, Tollison asserts that his calculation procedure attributes the 6.9% “overcharge” solely to defendants’ market allocation conduct and not to other aspects of defendants’ operating policies or market position. In describing that market allocation conduct, Tollison includes the actions of The Marshfield Clinic, Wausau Medical Center, Rhinelander Medical Center, Ministry Corporation, Security Health Plan and North Central Health Protection Plan to maintain a division of service areas and customers within the class area. The conduct served to allocate portions of the market to select providers and included “cooperation among the competitors to share facilities and physicians in support of their rival HMO plans.” Additionally, Tollison noted that the defendants took actions to foreclose the market entry of potential competitors.

4. Increased utilization

In an expert report and affidavit, Dr. Freeh states that health care markets are subject to economic forces different from other industries. It is his opinion that most Americans consume a higher than optimal level of health care because of the moral hazard created by health insurance. When market competition and managed care are operating properly, they reduce the utilization of physician services. According to Freeh, anticompetitive conduct leads to higher utilization.

C. Blue Cross Litigation

I take judicial notice of the following facts. On February 16, 1994, Blue Cross & Blue Shield United of Wisconsin and Comp-care Health Services Insurance Corporation filed claims in this court against The Marsh-field Clinic and Security Health Plan, Inc. alleging that defendants engaged in monopolization, price-fixing and market allocation activities in violation of §§ 1 and 2 of the Sherman Act and Wis. Stat. § 133.03. Blue Cross & Blue Shield United of Wisconsin sought significant damages for overcharges and equitable relief enjoining The Marshfield Clinic from future anti-competitive conduct. At trial, which began on December 13, 1994, the jury found for the plaintiffs on all claims and concluded that The Marshfield Clinic and Security Health Plan, Inc. had entered into a contract, combination or conspiracy to allocate customers, territories and product or service markets in violation of § 1 of the Sherman Act and Wis. Stat. § 133.03.

The Honorable Judge Shabaz denied post-trial motions relating to the § 1 violations in Blue Cross v. Marshfield Clinic, 883 F.Supp. 1247 (W.D.Wis.1995), determining that there was evidence that defendants allocated markets and territories with Rhinelander Medical Center, S.C., North Central Health Protection Plan, Wausau Medical Center and Wausau area physicians. The court entered an injunction that included provisions voiding agreements and understandings to allocate customers, territories, and product and service markets among The Marshfield Clinic, Security Health Plan, Inc., Rhinelander Medical Center, S.C., Wausau area physicians, North Central Health Protection Plan, Ministry Corporation, Gunderson Clinic and Rice Clinic.

On April 21, 1995, The Marshfield Clinic appealed the verdict. Although the Seventh Circuit reversed the verdict as to most violations, it affirmed the liability finding relating to the § 1 conspiracy to divide markets and remanded the case for a new trial to determine damages resulting solely from the division of markets. See Blue Cross, 65 F.3d at 1416. The court of appeals directed the district court to rewrite the injunction but held that the provisions forbidding defendants from allocating customers, territories and product and service markets with Wausau Area Physicians, North Central Health Protection Plan, Rhinelander Medical Corporation, S.C. and Ministry Corporation must stand. See id. On remand to this court, defendants were granted summary judgment because the plaintiffs were unable to demonstrate any injury tied specifically to the market allocation verdict. Blue Cross, Opinion and Order (April 7,1997).

OPINION

Section 1 of the Sherman Act, 15 U.S.C. § 1, makes illegal “[ejvery contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce among the several States.” Wisconsin’s antitrust law, Wis. Stat. § 133.03, contains similar language and provides coverage parallel to that of the federal act. State v. Waste Management of Wisconsin, Inc., 81 Wis.2d 555, 569, 261 N.W.2d 147 (1978). Because § 1 concerns itself with contracts, combinations and conspiracies, it does not proscribe actions that are taken unilaterally. Copperweld Corp. v. Independence Tube Corp., 467 U.S. 752, 768, 104 S.Ct. 2731, 2740, 81 L.Ed.2d 628 (1984). Further, the Sherman Act does not bar all restraints on trade, but only those that are unreasonable. NCAA v. Board of Regents of the University of Oklahoma, 468 U.S. 85, 98, 104 S.Ct. 2948, 2958-59, 82 L.Ed.2d 70 (1984).

Generally, courts consider conspiracies to allocate markets to be per se illegal under § 1 of the Sherman Act. Polk Bros., Inc. v. Forest City Enterprises, Inc., 776 F.2d 185, 188 (7th Cir.1985). However, per se illegality is limited to “naked” restraints on trade and does not cover agreements that facilitate productive activity. Blackburn v. Sweeney, 53 F.3d 825, 828 (7th Cir.1995); Polk Bros., 776 F.2d at 188. Market allocations that accompany and promote the success of larger endeavors are considered “ancillary” trade restraints and warrant more in-depth analysis under the Rule of Reason. Polk Bros., 776 F.2d at 190-91. Such an analysis focuses on the market power of the cooperating entities and their ability to raise prices, id., and turns ultimately on whether the overall effect of the challenged restraint is to enhance or reduce competition. NCAA, 468 U.S. at 104, 104 S.Ct. at 2961-62. In distinguishing between per se and Rule of Reason cases, courts must consider whether the anti-competitive agreement promoted enterprise and productivity at the time it was adopted. Polk Bros., 776 F.2d at 189. When the cooperative agreement contributes to productivity through integrated efforts, the Rule of Reason is the norm. Id. at 188.

Whether asserting a per se or Rule of Reason violation, private plaintiffs alleging market allocation must show that they have suffered an antitrust injury resulting directly from defendants’ unlawful conduct. Israel Travel Advisory Service, Inc. v. Israel Identity Tours, Inc., 61 F.3d 1250, 1257 (7th Cir.1995); MCI Communications v. American Telephone and Telegraph Co., 708 F.2d 1081, 1161 (7th Cir.1983). This requirement is often referred to as causation of damages and is distinct from a proof of the amount of damages. MCI, 708 F.2d at 1161. To establish causation of damages for market allocation, plaintiffs must demonstrate that the alleged co-conspirators possessed market power sufficient to raise prices. See Rebel Oil Co. v. Atlantic Richfield Co., 51 F.3d 1421, 1434 (9th Cir.1995) (“[i]n order unilaterally to raise prices above competitive levels, the predator must obtain sufficient market power”); William M. Landes & Richard A Posner, Market Power in Antitrust Cases, 94 Harv. L.Rev. 937, 938 and 955 (1981) (noting private antitrust plaintiff must prove effect on prices or exercise of market power). In turn, a finding of market power requires definition of relevant product and geographic markets of which the co-conspirators possess a substantial share. Id.

The numerous arguments raised by defendants in their motion for summary judgment fall into three categories: objections to the standing of HMO enrollee plaintiffs; objec tions to plaintiffs’ proof of an illegal conspiracy to allocate markets; and objections to plaintiffs’ proof of causation of damages.

A. Standing of HMO Enrollees

The class of plaintiffs is defined as including “all purchasers of physician services from defendants The Marshfield Clinic, Security Health Plan, Inc., Rhinelander Medical Center, S.C. and North Central Health Protection Plan.” In an opinion and order dated July 18, 1997 (dkt.# 187), I determined that HMO subscribers of defendants North Central and Security Health Plan would be included in the class as purchasers of physician services. This determination was based on plaintiffs’ theory that the defendant HMO’s were merely a means for marketing and selling physician services and on the fact that Security was wholly owned by defendant Marshfield Clinic. At the time of the ruling, I noted that this conclusion would be revisited on summary judgment if it were not supported by the evidence.

Taking their cue, defendants contend that plaintiff HMO subscribers are barred from asserting an antitrust claim under the direct purchase rule established in Illinois Brick Co. v. Illinois, 431 U.S. 720, 736, 97 S.Ct. 2061, 2069-70, 52 L.Ed.2d 707 (1977). In Illinois Brick, the Supreme Court held that antitrust relief was limited to individuals who purchase supra competitively priced products directly from a violating party, indirect purchasers who buy from a downstream middleman (usually a wholesaler or distributor) have no claim. Defendants argue that HMO subscribers are indirect purchasers of physician services because their plans purchased the physician services directly and resold them to plaintiffs. Even if this characterization of HMO’s is correct and the subscriber plaintiffs did buy “resold” products, the direct purchase rule does not apply because those products were sold by alleged co-conspirators. See In re Brand Name Prescription Drugs Antitrust Litigation, 123 F.3d 599, 604-05 (7th Cir.1997) (observing that rule inapplicable when middleman wholesaler also a conspirator).

B. Conspiracy to Allocate Markets

Plaintiffs contend that defendants conspired to allocate markets for all physician services within the eight-county class area, arguing that defendants’ conduct constitutes a per se violation of § 1 of the Sherman Act. In support of their claim, plaintiffs have introduced evidence of numerous vertical and horizontal arrangements between defendant Marshfield Clinic and alleged co-conspirators that plaintiffs assert reflect the existence of one overriding agreement to allocate the physician services market. On this motion for summary judgment, defendants contend that plaintiffs’ evidence does not support a finding that two of the alleged co-conspirators, Wausau Medical Center and Rice Clinic, were involved in an agreement to allocate markets. According to defendants, removing either of these alleged conspirators from the mix thwarts plaintiffs’ ability to demonstrate causation of damages and thus dooms plaintiffs’ entire claim.

In attempting to demonstrate the illegal market allocation activities of both Rice Clinic and Wausau Medical Center, plaintiffs seek to avail themselves of the tenet that minimal evidence is required to connect a particular participant to an established conspiracy. See United States v. Consolidated Packaging Corp., 575 F.2d 117, 126 (7th Cir.1978). This argument assumes that a market allocation conspiracy has been established. Plaintiffs advance two justifications for this presumption: that a market allocation conspiracy was found in Blue Cross, 65 F.3d at 1415-16, and that defendants have not challenged the overall conspiracy on this motion for summary judgment, thereby conceding its existence for the purpose of this motion. However, I have already ruled that the Blue Cross rulings have no preclusive effect in this litigation. See Opinion and Order, August 26, 1997 (dkt.# 247). Furthermore, defendants’ summary judgment challenge to the conspiracy placed the onus on plaintiffs to introduce evidence from which a reasonable jury could conclude that Rice Clinic and Wausau Medical Center agreed to allocate markets; if plaintiffs’ ability to prove these issues requires proof of the existence of a broader conspiracy, it was their burden to introduce evidence supporting such a finding. Accordingly, in considering plaintiffs’ evidence of market allocation I will not presume the existence of any overall conspiracy that is not supported by the record.

Similarly, plaintiffs make much of the Supreme Court’s admonition in Continental Ore Co. v. Union Carbide & Carbon Corp., 370 U.S. 690, 699, 82 S.Ct. 1404, 1410-11, 8 L.Ed.2d 777 (1962) (citations omitted), that a conspiracy should not to be judged “by dismembering it and viewing its separate parts, but only by looking at it as a whole.” This is another attempt by plaintiffs to avail themselves of the unavailable presumption that an overall conspiracy exists. Again, if plaintiffs’ proofs as to Rice Clinic and Wausau Medical Center depend on the existence of such a conspiracy, they were obligated to come forward with appropriate evidence. I will not assume the evidence exists.

I note that plaintiffs’ conspiracy evidence is unlike the acts the Supreme Court required to be viewed as a whole in Continental Ore but rather lends itself to more individual analysis. In Continental Ore, the collection of events the Supreme Court refused to “dismember” was innocuous when the events were viewed individually but supported a finding of an illegal conspiracy when viewed in the aggregate. Id. at 698-99, 82 S.Ct. at 1409-11. Here, plaintiffs’ evidence of a “single conspiracy” consists of several geographic sets of agreements, each involving defendant Marshfield Clinic and one or more other providers. The success and legality of the agreements in one area are not dependent on arrangements elsewhere, as they would be in a boycott or price-fixing conspiracy. The only permissible inference of a unified plan that may be drawn is that, in initiating and developing these individual relationships, defendant Marshfield Clinic intended that the different regional efforts would result in a fully allocated eight-county area; this unilateral intent does not affect the legality of the individual relationships. In any event, plaintiffs’ conspiracy proof as to each co-conspirator is unlike the evidence in Continental Ore, 370 U.S. at 698-99, 82 S.Ct. at 1409-10, and will be analyzed independently unless plaintiffs show otherwise.

Similarly, because the existence of a single conspiracy vel non does not affect the legality of the relationships between individual providers and defendant Marshfield Clinic, plaintiffs’ inability to establish a uniform plan does not end their claim. Plaintiffs may prevail by establishing several independent illegal market allocation agreements and proving that those agreements caused them to suffer antitrust injury.

1. Rice Clinic

Plaintiffs’ evidence showing that Rice Clinic agreed to allocate markets for physician services falls into two categories: evidence that Rice Clinic and The Marshfield Clinic have not established competing practices in each other’s home towns and evidence that the two clinics engaged in several cooperative ventures not related to market allocation for physician services. In determining whether a party’s evidence of an antitrust conspiracy is sufficient to survive summary judgment, a two-part inquiry is appropriate:

“(1) is the plaintiffs evidence of conspiracy ambiguous, i.e., is it as consistent with the defendants’ permissible independent interests as with an illegal conspiracy; and, if so, (2) is there any evidence that tends to exclude the possibility that the defendants were pursuing these independent interests.”

Market Force Inc. v. Wauwatosa Realty Co., 906 F.2d 1167, 1171 (7th Cir.1990) (quoting Gibson v. Greater Park City Co., 818 F.2d 722, 724 (10th Cir.1987)). As to the first inquiry, plaintiffs have not shown that The Marshfield Clinic’s or Rice Clinic’s failure to compete in each other’s home areas was inconsistent with their independent interests. Plaintiffs have not demonstrated that either party normally marketed or provided services in areas already occupied by an established clinic or that competition existed between the clinics before the alleged agreement. The proffered evidence reveals no agreement, implied or otherwise, by either clinic not to market in the other’s area. See Dreiling v. Peugeot Motors of America, Inc., 850 F.2d 1373, 1381 (10th Cir.1988) (concluding that first part of two-part Gibson inquiry not met where evidence reflected no expressed intent to suppress competition), cited favorably in Market Force Inc., 906 F.2d at 1171.

As to the second inquiry, plaintiffs’ evidence does not tend to exclude the possibility that, in failing to compete in each other’s areas, the two clinics acted independently. Plaintiffs have introduced numerous documents showing that, beginning in 1992, The Marshfield Clinic, with the aid of the Ministry Corporation, sought to build a “strong affiliation” with the Rice Clinic. These documents do not specify the nature of this affiliation and give no indication that it relates to market allocation. The clinics’ 1994 management services arrangement reflects another joint venture unrelated to market allocation. The existence of these cooperative arrangements does not tend to exclude the possibility that the parties’ failure to compete in each other’s areas resulted from independent decisions. In Dreiling, 850 F.2d at 1380, the court rejected an antitrust conspiracy claim supported with evidence similar to that offered here by plaintiffs. Dreiling involved an alleged conspiracy between Chrysler and Peugeot to refrain from producing competing cars. The court determined that the proffered evidence showed merely that the companies did not produce competing cars and that they had entered into an unrelated joint venture regarding distribution; it did not tend to exclude the possibility of independent action and thus failed to meet the second prong of the two-part Gibson inquiry. Id.

Plaintiffs’ remaining arguments regarding Rice Clinic are not supported by the evidentiary record. Although plaintiffs contend that Rice Clinic and Security Health Plan maintained an ongoing exclusive agreement to develop an HMO in Portage County that barred North Central from entering the area, their contention is unsupported by any evidence and is contradicted by the undisputed fact that Rice Clinic did not become an affiliated provider for Security until 1996. Even if this factual allegation were supported, plaintiffs do not explain how preventing North Central from establishing an HMO in Portage County affected the market allocation of physician services.

Finally, plaintiffs submit a 1993 letter to a Rice Clinic official from (presumably) another Portage County doctor. The letter describes a “realistic concern” among unidentified local physicians that they establish referral relationships with defendant Marsh-field Clinic or face a competing Marshfield facility in Stevens Point, but it is inadmissible hearsay. See Winskunas v. Birnbaum, 23 F.3d 1264, 1267-68 (7th Cir.1994) (requiring plaintiff to oppose summary judgment motion with admissible evidence). Plaintiffs argue that the letter was kept as a business record of defendant Marshfield Clinic, apparently trying to avail themselves of the Fed. R.Evid. 803(6) hearsay exception for business records. Even if this letter qualified for that exception, which it does not, the unidentified physicians’ “realistic concern” provides an additional layer of hearsay to which no exception applies.

Construing the evidence in the light most favorable to plaintiffs, I conclude that plaintiffs have failed to adduce sufficient evidence from which a reasonable jury could conclude that Rice Clinic entered into an illegal conspiracy to allocate markets.

2. Wausau Medical Center

Plaintiffs assert that Wausau Medical Center’s participation in an illegal market allocation agreement is demonstrated through its participation in the “Free Flow” agreement and the Security 65 medicare HMO as well as in documents reflecting a cooperative relationship between Wausau Medical Center and The Marshfield Clinic. Defendants contend that the Security 65 plan is an exclusive dealing agreement that must be analyzed under the Rule of Reason and that plaintiffs’ remaining evidence is either ambiguous or inconsistent with the alleged market allocation scheme.

In Blue Cross, 65 F.3d at 1415-16, the Court of Appeals for the Seventh Circuit held that documents presented at trial describing the “Free Flow” agreement between Security Health Plan and North Central were sufficient to support the jury’s finding of a mar ket allocation conspiracy. Those documents described how the agreement did not cover care rendered by North Central affiliated physicians in Marshfield and by Marshfield Clinic physicians competing in Wausau. Despite noting the agreement’s beneficial effect of expanding physician choice for both plans’ enrollees, Judge Posner refrained from affording it Rule of Reason evaluation because defendants Marshfield Clinic and Security Health Plan had provided no argument or evidence showing why the no-compete clause was necessary to the venture. Id. at 1416.

It appears that those same documents, or very similar ones, have been presented here. However, defendants have ignored Judge Posner’s invitation to show that the territorial restrictions of the “Free Flow” agreement were necessary to promote the overall venture and therefore subject to Rule of Reason analysis. They have not adopted the “free riding” justification for the restrictions Judge Posner presumed to exist and have provided no factual submissions concerning the matter. Defendants mention in passing that the pro-competitive effects of “Free Flow” were recognized in Blue Cross, but, as that case demonstrates, pro-competitive effects are irrelevant to a per se determination unless the alleged trade restrictions support their success. See also Polk Bros., 776 F.2d at 188.

Defendants offer no valid reason why plaintiffs’ evidence supported a jury finding of market allocation in Blue Cross but no longer does so. Defendants emphasize that the Security 65 plan, a joint venture between Wausau Medical Center and Security Health Plan, demonstrates that the “Free Flow” arrangement did not reflect a broad agreement to keep Security out of the Wausau area. They miss the point. Judge Posner did not conclude that the “Free Flow” agreement divided markets by keeping Security out of Wausau, but rather by keeping the individual clinics and physicians from opening offices in each other’s territories. Blue Cross, 65 F.3d at 1416. This aspect of the agreement, the allocation of physician services, is of concern here as well. Although defendants note that Wausau Medical Center was not a direct signatory to the “Free Flow” agreement, this fact is irrelevant; Wausau’s awareness and participation in the agreement’s territorial restrictions was sufficient. See Wilk v. American Medical Association, 735 F.2d 217, 219 (7th Cir.1983) (concluding that antitrust liability supported by evidence that party knew concerted action invited and participated in scheme).

I conclude that, in and of itself, the “Free Flow” agreement supports a finding that the Wausau Medical Center participated in an illegal conspiracy to allocate markets in physician services, making it unnecessary to consider the remaining evidence. I note, however, that the arguments of both parties concerning the Security 65 plan reflect misunderstandings of the per se — Rule of Reason distinction that should be corrected before trial. Defendants contend that any competitive restrictions associated with the Security 65 plan are ancillary to a larger productive effort (although again they supply thin evidence of the necessity of those restrictions). Plaintiffs counter that Security 65 is merely a component in an overall conspiracy to allocate markets and that such conspiracies are per se illegal.

Even if plaintiffs were allowed the presumption that an overall conspiracy exists, their assertion that all market allocation agreements are per se illegal ignores the distinction between naked and ancillary restraints. See Polk Bros., 776 F.2d at 188. Market allocations are per se illegal only if they do not facilitate cooperative and productive activity. Id. Here, the challenged arrangement, or at least a portion of that arrangement, is ancillary to a pro-competitive venture. Plaintiffs give no explanation how evidence of a broader market allocation, if it existed, could transform the Security 65 plan from an ancillary to a naked restraint. (Whether it would affect the Rule of Reason analysis is a different issue, although one also untouched by plaintiffs.) Further, it appears from the current record that the lion’s share of plaintiffs’ market allocation evidence is based on ancillary restraints subject to the Rule of Reason. Plaintiff cannot cast all of these as per se violations and therefore disregard their productive benefits simply by contending that they are part of an “overall” naked restraint, the only proof of which is the ancillary restraints themselves.

In turn, defendants assert that because plaintiffs have advanced their claims exclusively under a theory of per se liability, their case fails if part or all of the challenged conduct qualifies for assessment under the Rule of Reason. However, the authority cited by defendants, Polk Bros., 776 F.2d at 191, does not place a strict requirement on antitrust plaintiffs to advance alternative arguments under per se liability and the Rule of Reason. In Polk, the court of appeals rejected the plaintiffs per se challenge to what was in fact an ancillary restraint only after concluding that the record did not contain the evidence necessary to conduct a Rule of Reason analysis (market power and consumer harm). Id. The present record is not subject to that same deficiency.

C. Causation of Antitrust Damages

Evidence of an illegal market allocation scheme is not sufficient to get plaintiffs to trial; additionally, they must demonstrate that there is a jury issue whether they suffered an “injury of the type the antitrust laws were intended to prevent and that flows from that which makes defendants’ acts unlawful.” MCI Communications, 708 F.2d at 1161 (quoting Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 489, 97 S.Ct. 690, 697, 50 L.Ed.2d 701 (1977)). Proof of causation of damages begins with evidence that the alleged co-conspirators possessed market power, which entails the ability to set price above marginal cost. See Landes & Posner, 94 Harv. L.Rev. at 938, 939. Antitrust damage must flow from the exercise of that power.

1. Market power

The standard method of proving market power is to define a relevant market by product and geography, show that the co-conspirators’ share of that market is substantial and show that other suppliers are inhibited from entering or increasing their share in the market. Rebel Oil Co., 51 F.3d at 1434. A product market’s boundaries are determined by “the reasonable interchangeability of use or the cross-elasticity of demand between the product itself and substitutes for it.” Brown Shoe Co. v. United States, 370 U.S. 294, 325, 82 S.Ct. 1502, 1523-24, 8 L.Ed.2d 510 (1962). Plaintiffs’ experts define an appropriate product market that consists of all physician services. Although this market may include several physician specialty sub-markets, this fact does not affect its legitimacy. See id. (antitrust product markets may contain independently sustainable sub-markets).

Defendants assert that physician services do not qualify as a “cluster market,” a term used in antitrust cases to denote a market limited to sellers who provide a collection of goods. See Thurman Industries, Inc. v. Pay ‘N Pak Stores, 875 F.2d 1369, 1377 (9th Cir.1989). A cluster market includes only those sellers who provide the entire array of “clustered” goods, such as hardware centers, and excludes sellers of individual products, such as hammer stores and nail stores. “[A] cluster approach is appropriate where the product package is significantly different from, and appeals to buyers on a different basis from, the individual products considered separately.” JBL Enterprises, Inc. v. Jhirmack Enterprises, Inc., 698 F.2d 1011, 1016-17 (9th Cir.1983) (approving cluster market at wholesale level of suppliers of comprehensive line of beauty products, as distinct from' suppliers of shampoo only). To qualify as a cluster market, the “clustered” goods must be generally purchased as a group, see United States v. Grinnell Corp., 384 U.S. 563, 572, 86 S.Ct. 1698, 1704-05, 16 L.Ed.2d 778 (1966) (sellers who provided both fire and burglar alarms made up one cluster market because products generally purchased together) and the value of the individual services must increase when packaged with the rest, see United States v. Philadelphia National Bank, 374 U.S. 321, 356-57, 83 S.Ct. 1715, 1737-38, 10 L.Ed.2d 915(1963) (commercial banks providing full array of financial services are separate market from institutions providing individual products, such as savings and loans). Defendant contends that different physician services, such as dermatology and oral surgery, are generally purchased individually and are valued the same by the consumer whether purchased from a comprehensive clinic or a sole practitioner.

However, defendants’ assertion that physician services do not constitute a duster market is as irrelevant as it is correct. Plaintiffs claim does not depend on a duster theory; they do not define a market limited to large clinics that provide the full array of physician services. To the contrary, plaintiffs’ market definition includes every physician in the eight-county class area, whether an employee of The Marshfield Clinic behemoth or a self-employed specialist. Defendants’ real objection is that plaintiffs do not analyze physician services as making up several individual markets, but none of the duster market cases cited requires them to do so.

Plaintiffs’ experts have defined the proper geographic markets for physician services in two different ways. Dr. Freeh concluded that three distinct geographic markets existed within the eight-county class area, while Dr. Beyer concluded that the entire class area was subsumed by a larger, thirteen-county market. Defendants argue that plaintiffs’ claim fails because neither expert defined a geographic market coextensive with the class area. Defendants overemphasize the role of market definition, which merely provides a method of determining market power and causation of damages. See Landes & Posner, 94 Harv. L.Rev. at 938. There is no requirement that market power be shown through a single market. As long as plaintiffs show that within each defined geographic market the alleged co-conspirators hold a sufficient share from which market power can be inferred, their market definition is sufficient.

Dr. Freeh has calculated market share by determining in each of his three markets the percentage of doctors affiliated with the alleged eo-eonspirators. This method appears the most reasonable for calculating market share in the physician services market. After discounting from his calculations the Rice Clinic physicians, who can no longer be considered part of a market allocation scheme, the alleged co-conspirators account for 63 percent of all physicians in the Lincoln and Marathon County market, 70 percent in the Oneida, Price, Taylor, Clark and Wood County market and 3 percent in the Portage County market. Plaintiffs concede that, without Rice Clinic as a co-conspirator, they cannot show market power in Portage County. Accordingly, any class members who resided in Portage County or purchased physician services there will be dropped from the plaintiff class.

The co-conspirators’ 63 and 70 percent shares in the remaining two markets must be considered in light of the barriers faced by competitors seeking to enter those markets or expand existing production and any direct evidence that the co-conspirators were able to exert market power. See Wilk v. American Medical Association (Wilk II), 895 F.2d 352, 359-60 (7th Cir.1990) (evidence of 50 percent market share, barriers to entry and anti-competitive effects supported finding of market power); Ball Memorial Hospital, Inc. v. Mutual Hospital Insurance, 784 F.2d 1325, 1336 (7th Cir.1986) (large market share irrelevant when no barriers to entry). Plaintiffs have submitted evidence that physicians who enter the class area and are not affiliated with one of the defendants face difficulties in getting hospital privileges, establishing referral networks, overcoming the Marshfield Clime’s brand recognition and obtaining cross coverage. Although this evidence may not support a monopolization claim against defendant Marshfield Clinic, see Blue Cross, 65 F.3d at 1413-14, it does show that the ability of potential competitors to enter the market was restrained. The co-conspirators’ market shares of over 50 percent in each market, combined with these barriers to entry and evidence of above average physician costs (see below), are sufficient evidence from which a jury could conclude that the defendants possessed market power in the revised seven-county area.

2. Plaintiffs’ injury

Plaintiffs allege that defendants’ market allocation activities led to supra-eompetitive prices for physician services. Dr. Tollison, plaintiffs’ damages expert, has submitted calculations based on data of medical service transactions that occurred between 1988 and 1996 throughout Wisconsin. Tollison’s calculations show that, during that peri od, the average annual per capita cost for physician services in the eight-county area was 6.9%, or $58 higher than it was outside that area. Tollison asserts that his calculations control for legal market factors affecting price. Specifically, he refers to a regression analysis he performed that controlled for market density and contends that the statewide data area, which includes Madison and Milwaukee, is broad and varied enough to account for any price increase associated with reputation or quality.

Although Dr. Tollison’s analysis is not thoroughly convincing, I cannot say that a jury would be unreasonable to conclude from his expert report, combined with findings of a market allocation conspiracy and strong market power, that plaintiffs paid more for physician services because of defendants’ unlawful conduct. Despite extensive briefing on the issue of causation of damages, defendants have cited no case where similar evidence was held insufficient to support a jury finding of antitrust injury. Defendants cite several cases in which the evidence regarding causation of damages was rejected for failing to disaggregate the effects of legal conduct by the defendants. See Schiller & Schmidt, Inc. v. Nordisco Corp., 969 F.2d 410, 415-16 (7th Cir.1992); City of Vernon v. Southern California Edison Co., 955 F.2d 1361, 1371-73 (9th Cir.1992); MCI Communications, 708 F.2d at 1162-64. However, in each of these cases the expert report at issue either relied specifically on conduct determined by the court to be legal, City of Vernon, 955 F.2d at 1372 (expert stated that he could not or would not segregate effects of lawful act); MCI Communications, 708 F.2d at 1163 (lost profits study premised on 22 acts of monopolization did not support causation of damages after 15 acts found legal), or was a thinly veiled attempt to attribute to a minor violation damages that were obviously caused by legal competition, Schiller & Schmidt, 969 F.2d at 415. In contrast, Tollison states specifically that the calculated overcharge is attributable to defendants’ market allocation activities only.

The rejection of Rice Clinic as a co-eonspirator does not undermine plaintiffs’ damage evidence and justify summary judgment, although Tollison’s conclusions will have to be recalculated to exclude Portage County. It would be unreasonable to terminate plaintiffs’ claim on this point, particularly when the evidence indicates that Portage County’s exclusion from the calculation is unlikely to alter the result substantially. Defendants attack Tollison’s report for failing to disaggregate numerous other legal factors that, according to defendants, affected the price of physician services in the class area. These range from natural monopoly to various and unspecified legal cooperative ventures. However, with the exception of evidence regarding defendant Marshfield Clinic’s strong reputation, the parties’ proposed findings of fact do not establish that these factors even exist, let alone that they served to raise the cost of physician services in the area.

Defendants challenge Tollison’s use of an area-wide average overcharge calculation, contending that the degree to which each class member’s annual medical costs differed from the state average would vary depending on the type of services utilized; according to defendants, the mark-up for pediatric care might be lower than that for orthopedic surgery, for example. Tollison’s statement that the alleged market allocation had a uniform negative effect on all class members and the price of all physician services is unsupported by any reference to data analysis and will be disregarded. See Minasian v. Standard Chartered Bank, PLC, 109 F.3d 1212, 1216 (7th Cir.1997) (rejecting expert affidavit containing several conclusions but “devoid of analysis”). However, defendants have cited no authority for their underlying premise, that plaintiffs are required to prove identical injury for all class members. From Tollison’s averaged data, a reasonable jury could not conclude that each class member actually paid $58 more for physicians services each year because of defendants’ conduct. Such a conclusion is not necessary; if it were, averaged data would never be permissible to show antitrust damages. It is sufficient that Tollison’s report supports an inference that defendants’ conduct led to higher prices for physician services in the class area. Defendants have submitted no evidence that would render that inference unreasonable, as their assertion that the relative overcharges varied among practice areas is unsupported by any proposed finding of fact.

Of the $58 annual overcharge Tollison found in the class area, he attributes $32 to increased prices and $26 to increased utilization, meaning either more total visits or a higher occurrence of expensive procedures. Defendants contend that plaintiffs’ claim is doomed by this second conclusion, that patients in the eight-county area received more care than average. As defendants see it, this increased output of physician services proves that plaintiffs were not harmed by anti-competitive conduct: antitrust injuries can result only from an output reduction. Defendants support this position with numerous quotes from opinions requiring that antitrust plaintiffs prove a loss resulting from either reduced output or raised prices. See, e.g., Stamatakis Industries, Inc. v. King, 965 F.2d 469, 471 (7th Cir.1992) (“[t]he antitrust injury doctrine ... ‘requires every plaintiff to show that its loss comes from acts that reduce output or raise prices to consumers’ ”) (quoting Chicago Professional Sports Limited Partnership v. National Basketball Association, 961 F.2d 667, 670 (7th Cir.1992)). Some cases go further: “Unless a contract reduces output in some market, to the detriment of consumers, there is no antitrust problem. A high price is not itself a violation of the Sherman Act.” Chicago Professional Sports Limited Partnership v. National Basketball Association (Chicago Prof Sports Ltd. II), 95 F.3d 593, 597 (7th Cir.1996).

Plaintiffs contend that these holdings are inapplicable to the health care field, where unique market factors and perverse incentives encourage a monopolist or cartel to artificially increase output, This argument is a drastic departure from antitrust economics and unsupported by case law; I would be reluctant to allow a showing of damages derived solely from evidence of increased treatment, or over-utilization. See Blue Cross, Opinion and Order at --- (April 14, 1997). However, defendants’ position that plaintiffs are barred from proceeding because of increased utilization is equally novel and will be rejected. They cite no case denying a similar claim where the evidence showed an increase in both price and output, the existence of which demonstrates that conventional economic principles do not control.

Defendants’ remaining objections to the damages evidence show that Tollison’s report could be more convincing, but do not rob it of its validity. They contend that the statewide data are an improper benchmark, but they do not explain exactly why or provide any evidence to debunk Tollison’s conclusion to the contrary. Defendants do contend that the report’s uniform treatment of the eight-county area is flawed because of the evidence of substantially different overcharge rates between physicians employed by defendant Marshfield Clinic and the other defendants’ physicians. However, this evidence was attached as an exhibit to their reply brief and not introduced through proposed findings of fact, as required by this court’s summary judgment procedures, so I have not considered it.

ORDER

IT IS ORDERED that the motion for summary judgment of defendants Marshfield Clinic and Security Health Plan of Wisconsin, Inc. is GRANTED as to plaintiffs’ contention that Rice Clinic conspired with defendants to allocate markets for all physician services and as to plaintiffs who reside in Portage County or have purchased physician services there; the motion is DENIED in all other respects.