Delaware Health Care Inc. v. MCD Holding Co., 957 F. Supp. 535 (D. Del. 1997)

*536 *537 Christopher J. Curtin, of Erisman & Curtin, Wilmington, Delaware; of counsel: Joanne Zack, Michael J. Boni, and Craig W. Hillwig, of Kohn, Swift & Graf, P.C.; Philadelphia, Pennsylvania, for plaintiff.

Stephen E. Herrmann, of Richards, Layton & Finger, Wilmington, Delaware; of counsel: Brian M. Peters, Kathleen M. Chancler, Jonathan B. Sprague, Mark L. Mattioli, Mark L. Mattioli, and Katherine Layman, of Post & Schell, P.C., Philadelphia, Pennsylvania, for defendants.

MURRAY M. SCHWARTZ, Senior District Judge.

This lawsuit alleging violations of federal antitrust laws was brought in 1994 by Delaware Health Care, Inc. ("DHC"), a provider of home health care. See Docket Item ("D.I.") 1. Defendants are MCD Foundation ("the Foundation"), its wholly owned subsidiaries, The Medical Center of Delaware ("MCD"), MCD Holding Company ("MCD Holding") and the Visiting Nurse Association of Delaware ("VNA"), and related entities Infusion Services of Delaware ("ISD"), and the Christiana and Wilmington Hospitals. Id.; A-81; A-527-611. The two hospitals constitute MCD, which the parties agree is Delaware's largest hospital system. D.I. 121, at 4; D.I. 126, at 2.

Before the Court is the defendants' motion for summary judgment on Counts I, II and III of the plaintiff's complaint, and accompanying motion to dismiss the plaintiff's state claim for lack of supplemental subject matter jurisdiction. In opposition papers, the plaintiff informed the Court it was abandoning its claims under Counts I and III of the complaint; remaining for consideration is Count II, alleging "Attempted Monopolization, Monopolization and Predatory Conduct" under Section 2 of the Sherman Act. See D.I. 1, at 11-13; D.I. 126, at 31 n. 15. Jurisdiction is based upon 28 U.S.C. § 1337 and 15 U.S.C. §§ 15, 26. D.I. 1, at 2.

DHC was established by Lawrence and Margaret ("Peggy") Carroll in 1990 as a home health care agency designed to deliver infusion and other health care services to medical patients in their homes. A-1086-88.^[1] Home infusion services include delivery of pharmaceutical or nutritional products to patients as well as related nursing care. A-913; B-100. DHC supplied both the pharmaceutical product and the nursing service components of home infusion services. B-99-101.

Home health care services are sought largely by individuals who have been discharged from hospitals yet remain in need of *538 continuing medical care. A-913. Home health care providers often rely on referrals and contractual positioning (hereafter referred to collectively as "referrals") as sources of business; such referrals come from hospitals, physicians, insurance companies, health maintenance organizations ("HMO's"), nursing homes, state contracts and other patients. A-246; A-1454. Home infusion therapy is considered by some to be the most profitable of home health care services. A-770.

DHC originally operated as a corporate agent of Norrell Healthcare, Inc., a temporary staffing agency. A-1088; A-1428. In November 1993, Norrell's home health care division was sold to Nurse's House Call ("NHC"), and DHC did business under NHC's name. A-1089; A-1428. At the end of 1994, DHC and NHC terminated their relationship and DHC became known as Delaware Home Care, Inc. Id.; B-101.

Defendant ISD, an unincorporated division of defendant MCD Holding, was established in 1990, also with the goal of supplying home infusion pharmaceutical products. A-725. Defendant VNA, in turn, was to provide the home nursing services necessary to administer ISD's pharmaceutical product. A-1407. The relationship among ISD, VNA and MCD all owned by MCD Foundation and the resulting referral of MCD patients to ISD/VNA form the basis of this lawsuit.^[2]

Plaintiff asserts prior to ISD's formation, discharge planners recommended home health care providers to patients on an informal rotating basis.^[3]See e.g. B-35. This rotation process was dismantled at the same time ISD was formed, plaintiff says, and in its place a directive was issued to refer patients only to ISD/VNA. See e.g. B-180-182; B-612; B-800. In practice, plaintiff says, patients were channeled to ISD even when they specifically requested a different provider. See e.g. B-345-52; B-429.

In addition to patient channeling, plaintiff charges defendants permitted ISD and ISD alone access to patients in their hospital rooms. This practice allowed ISD to offer superior services which in turn led to even more referrals. While all the other home care providers were barred from soliciting patients in their hospital rooms, plaintiff proffered deposition testimony that ISD's administrative director Mary Fitzpatrick often met with patients in their hospital rooms, both before and after a formal relationship between the patient and ISD had been established. B-44; B-52; B-141; B-185-86. Plaintiff has adduced evidence Mary Fitzpatrick met with patients in need of home care and taught them about home health care services before they left their hospital beds. B-185-86; B-200. Plaintiff contends this access is a selling point for ISD. D.I. 126, at 12-14.

Since its inception, ISD quickly gained market share. According to plaintiff's experts Dr. Solow and Dr. Black, ISD served 44.9% of the New Castle County home infusion therapy market in 1993.^[4] B-505; B-540. In 1994, the figure was 59%, and in 1995 it was 49.4%. B-505; B-540. The rest of the home infusion therapy market was divided among some 14 other providers.^[5] B-542. ISD was even more successful among the group of home infusion therapy patients who were discharged from MCD. Of those patients, ISD cared for 60.2% in 1993, 76.4% in 1994, and 77.1% in 1995. B-505; B-535. ISD earned gross revenues of $4,585,066 in 1994 and $5,822,456 in 1995. B-674-75.

DHC on the other hand cared for 28 patients, or 3.6% of the New Castle County home infusion therapy market, in 1993. See B-542. In 1994, it cared for 35 patients, or 3.5% of the market. Id. In 1994, it earned gross revenues from its infusion products of $186,511. B-676; B-704. It appears it *539 earned no income from infusion products in 1995. Id.^[6]

Summary judgment is governed by Federal Rule of Civil Procedure 56, and will be granted when, on the record before the Court, "there is no genuine issue as to any material fact and ... the moving party is entitled to a judgment as a matter of law." Fed.R.Civ.P. 56(c). On this summary judgment motion, "[t]he evidence of [the nonmovant] is to be believed, and all justifiable inferences are to be drawn in [their] favor." Eastman Kodak Co. v. Image Technical Servs., Inc., 504 U.S. 451, 456, 112 S. Ct. 2072, 2076, 119 L. Ed. 2d 265 (1992) (citations omitted).

The moving party "bears the burden of proving that no genuine issue of material fact is in dispute." Ideal Dairy Farms, Inc. v. John Labatt, Ltd., 90 F.3d 737, 743 (3d Cir. 1996) (citation omitted). Once the moving party has carried its burden, the nonmoving party "must come forward with `specific facts showing that there is a genuine issue for trial.'" Id. "The non-movant must present concrete evidence supporting each essential element of its claim." Id.

In Count II of the complaint, plaintiff alleges: "Defendants Medical Center and the VNA possess the monopoly power in the hospital and home health industry, respectively, which was wilfully acquired and maintained to effectuate predatory and unreasonable exclusionary acts in violation of Section 2 of the Sherman Act, 15 U.S.C. § 2. The said Defendants have attempted monopolization and have monopolized home health care...." D.I. 1, at 12. In its brief opposing summary judgment, plaintiff limited its Sherman Act Section 2 claims to attempted monopolization.^[7] D.I. 126, at 31.

The elements of a claim of attempted monopolization are "(1) ... the defendant has engaged in predatory or anticompetitive conduct with (2) a specific intent to monopolize and (3) a dangerous probability of achieving monopoly power." Spectrum Sports, Inc. v. McQuillan, 506 U.S. 447, 456, 113 S. Ct. 884, 890, 122 L. Ed. 2d 247 (1993). As in all antitrust cases, plaintiff here must additionally prove antitrust injury. Mathews v. Lancaster Gen. Hosp., 87 F.3d 624, 641 (3d Cir. 1996) (Sherman Act § 1 conspiracy claim); Schuylkill Energy Resources, Inc. v. Pennsylvania Power & Light Co., No. 95-4885, 1996 WL 284994, at *2 (E.D.Pa. May 21, 1996) (Sherman Act § 2 attempted monopolization claim).

Plaintiff has alleged two specific methods by which defendants have worked to achieve their attempted monopoly: "leveraging," and denial of access to an "essential facility."^[8] With respect to the former, plaintiff says MCD Foundation "leveraged its monopoly power, in the hospital market, to extend its monopoly in the home health market through Defendant Hospitals." D.I. 1, at 12-13. Plaintiff's essential facilities argument is:

Defendant Hospitals denied the Plaintiff any access to home care patients already discharged, or about to be discharged, from Defendant Hospitals, and access to the patients' records; but allowed its subsidiary, Defendant VNA, and a related forprofit *540 entity, Defendants Holding and ISD, access to these patients and their medical records.... Defendant Hospitals are the most important referral sources for the Plaintiff, and are, therefore, an essential facility....

Id. at 12.

Illegal leveraging under antitrust laws involves use of monopoly power in one market (the "upstream" market) to gain or threaten monopoly power in another, related market (the "downstream" market). See Fineman v. Armstrong World Industries, 980 F.2d 171, 198, 204-06 (3d Cir. 1992), cert. denied, 507 U.S. 921, 113 S. Ct. 1285, 122 L. Ed. 2d 677 (1993).^[9]

One element of plaintiff's leveraging claim is proof defendant MCD possessed monopoly power in the "upstream" hospital market. See Fineman, 980 F.2d at 198. If there is no monopoly power upstream, there can be no illegal leveraging of the downstream market. Id. at 203. To prove monopoly power in the hospital market, plaintiff must show defendants had the "power to control prices or exclude competition[.]" Eastman Kodak Co., 504 U.S. at 481, 112 S. Ct. at 2090; Fineman, 980 F.2d at 201. Such power does not exist in a vacuum, however, and must be evaluated in terms of a "relevant market." See Spectrum Sports, 506 U.S. at 457, 113 S. Ct. at 891 ("[I]t is beyond a doubt that [a claim of monopolization] requires proof of market power in a relevant market.").

Analysis of the relevant market has two components: a product market and a geographic market. Id. at 459, 113 S. Ct. at 892. Plaintiff bears the burden of defining the relevant market. Pastore v. Bell Tel. Co., 24 F.3d 508, 512 (3d Cir.1994); see also Ideal Dairy Farms, 90 F.3d at 749. Although market definition is a question of fact, Borough of Lansdale v. Philadelphia Elec. Co., 692 F.2d 307, 311 (3d Cir.1982), summary judgment is appropriate if plaintiff fails to present sufficient evidence in support of its proffered market definition. See Ideal Dairy Farms, 90 F.3d at 749; accord Home Health Specialists, Inc. v. Liberty Health Sys., No. 92-3413, 1994 WL 463406, at *2 (E.D.Pa. Aug. 24, 1994), aff'd, 65 F.3d 162 (3d Cir.1995); Miller v. Indiana Hosp., 814 F. Supp. 1254, 1266 (W.D.Pa.), aff'd, 975 F.2d 1550 (3d Cir.1992), cert. denied, 507 U.S. 952, 113 S. Ct. 1366, 122 L. Ed. 2d 744 (1993).

"To define a market is to identify producers that provide customers of a defendant firm (or firms) with alternative sources for the defendant's product or service." 2A Phillip E. Areeda et al. Antitrust Law ¶ 530a, at 150 (1995); see also Eastman Kodak Co., 504 U.S. at 481-82, 112 S. Ct. at 2090 ("The relevant market for antitrust purposes is determined by the choices available to Kodak equipment owners."). The relevant geographic market is "the area in which a potential buyer may rationally look for the goods or services he or she seeks[.]" Tunis Bros. Co., Inc. v. Ford Motor Co., 952 F.2d 715, 726 (3d Cir.1991) (citation omitted), cert. denied, 505 U.S. 1221, 112 S. Ct. 3034, 120 L. Ed. 2d 903 (1992). "[T]he geographic market is not comprised of the region in which the seller attempts to sell its product, but rather is comprised of the area where his customers would look to buy such a product." Id.^[10]*541 Once the relevant geographic market is defined, evidence of dominant market share within that market is a primary, but not sole, determinant of monopoly power. Barr Labs., Inc. v. Abbott Labs., 978 F.2d 98, 112 (3d Cir.1992).

The parties agree the relevant upstream product market in this lawsuit is inpatient hospital services. The geographic market, on the other hand, is hotly disputed; defendants' motion for summary judgment hinges largely on their assertion that plaintiff has "not come close" to meeting its burden of defining the relevant geographic market. D.I. 121, at 17. According to plaintiff, the relevant upstream geographic market is New Castle County, Delaware. D.I. 126, at 32-34. Defendants urge it is an area within a 15 to 20 mile radius of MCD which would include some areas outside New Castle County. D.I. 121, at 18. The implications of such market definition are clear: "A market drawn too tightly creates the illusion of market power where none may exist." Home Health Specialists, 1994 WL 463406, at *2. "The geographic market must be broad enough that consumers would be unable to switch to alternative sellers in sufficient numbers to defeat an exercise of market power." Id.

In this case, potential MCD customers could turn only to one other option for inpatient hospital services within plaintiff's proposed geographic market: St. Francis Hospital.^[11] B-580. On the other hand, between six and 13 hospitals compete within defendants' proposed market comprising a 15 to 20 mile radius of MCD. A-938-40. Under defendants' proposed market, MCD possessed at most a 43% market share during the relevant time period. Id. The Third Circuit Court of Appeals has held a market share of less than 55%, without other evidence tending to show monopoly power, is insufficient as a matter of law to establish that an antitrust defendant had monopoly power. See Fineman, 980 F.2d at 201.

As support for their position plaintiff's market definition is incorrect, defendants argue plaintiff's expert, Dr. Solow, focused on historical patient migration data in constructing its geographic market, without also providing data of what patients would do if faced with an exercise of monopoly power by MCD. D.I. 121, at 21-23.

Specifically, defendants assert, in defining the relevant market, Dr. Solow relied solely on the Elzinga-Hogarty ("E-H") test instead of the "standard methodology" of the U.S. Department of Justice Merger Guidelines ("Merger Guidelines").^[12]Id., at 19-20. The E-H test involves analysis of the flow of consumers in and out of the proposed market. First, a measurement is taken of the percentage of consumers in a proposed geographic *542 market who purchased goods or services produced in the market, as opposed to those who left the market (called "LIFO" Little In From Outside).^[13] B-574. Second, the percentage of goods or services produced in the market that are purchased by residents, as opposed to non-residents, of the proposed market, is measured (called "LOFI" Little Out From Inside).^[14]Id. Both LIFO and LOFI must be considered. Id.

In essence, these calculations are used to test a proposed geographic market. The results can be evidence the proposed market is correctly defined when a large percentage of consumers purchase the goods or services at issue from within the market (LIFO), and when, additionally, a large percentage of people who purchase the goods or services produced in the market are residents of that market (LOFI). See id. A 90% ratio on the LIFO and the LOFI analysis meaning 90% of the consumers in the proposed market stayed in that market to purchase the goods or services, and 90% of the goods or services produced in that market were purchased by residents of that market is considered strong evidence that the proposed market is correctly defined. Id. A 75% ratio is considered weak, but still credible, evidence that the proposed market is correctly defined. Id.

In the present case, plaintiff's experts found in 1992, 90.5% of the residents of New Castle County who used inpatient hospital services, used hospitals in the county. B-576. That corresponds to the LIFO inquiry. Further, the experts found 85% of the patients who used the hospitals in New Castle County in 1992 were residents of the county. B-575. That corresponds to the LOFI inquiry. Accordingly, Dr. Solow opined the proposed New Castle County inpatient hospital services market "clearly passed both the weak LIFO and the weak LOFI tests, and marginally passes the strong LIFO test as well." Id.

Defendants urge the E-H test is insufficient as a tool of market definition. They note critics have opined that the usefulness of the E-H test is limited to determining when a market is not correctly defined. See Dennis A. Yao, The Analysis of Hospital Mergers and Joint Ventures: What May Change?, 1995 Utah L.Rev. 381, 386 (1995) (A-661); see also United States v. Rockford Memorial Corp., 717 F. Supp. 1251, 1271 (N.D.Ill.1989), aff'd, 898 F.2d 1278 (7th Cir.), cert. denied, 498 U.S. 920, 111 S. Ct. 295, 112 L. Ed. 2d 249 (1990). Additionally, defendants point out the antitrust law treatise with which Dr. Solow is associated indicates reliance solely on historical tests like the E-H test "can misstate the size of the market." See 2A Areeda, et al., supra ¶ 550, at 217.

The Merger Guidelines, on the other hand, emphasize what consumers would do if faced with a price increase in the proposed market. Under the Merger Guidelines, the relevant market is defined as the area in which a hypothetical monopolist could profitably impose a price increase:

In determining whether a hypothetical monopolist would be in a position to exercise market power, it is necessary to evaluate the likely demand responses of consumers to a price increase. A price increase could be made unprofitable by consumers either switching to other products or switching to the same product produced by firms at other locations.

*543 A-486 (Merger Guidelines). While Dr. Solow indicates his analysis is consistent with the Merger Guidelines, B-571, defendants argue the analysis lacks the crucial forward-looking component that asks what patients would do in the event of a price increase by MCD. D.I. 121, at 18.

As defendants indicate, there is authority for the proposition that plaintiff's reliance on historical patient migration data in the form of the E-H test, without analysis of where consumers could turn in the event of an exercise of monopoly power, is insufficient to avoid summary judgment. See Home Health Specialists, Inc., 1994 WL 463406, at *3-5; accord Bathke, 64 F.3d at 345. Plaintiff asserts, however, it has provided forward-looking evidence of how patients would respond to a price increase by MCD: "the consumer likely would do nothing." D.I. 126, at 50. That is because plaintiff's expert believes consumers of medical care are driven by factors other than cost in deciding which hospital to use. Dr. Solow states:

Patients generally do not have the sort of information necessary to make informed judgments about what medical procedures are appropriate, which hospital is best able to provide them, or what they will cost. It is often difficult and expensive for them to search for this sort of information, and they are unlikely to be in a position to evaluate the information they do receive. Moreover, patients whose services will be covered by third-party payors have little incentive to be responsive to prices. As a result, they rely on their doctors for advice, often allowing their doctors to select a hospital, and their doctors have little incentive to choose among hospitals so as to minimize the cost of treatment.

B-577-78.

Moreover, plaintiff has provided deposition testimony from doctors stating that they refer patients to hospitals outside New Castle County only infrequently. See B-20; B-158; B-210. This is due to the intersection of state licensure laws for physicians and the necessity for Delaware licensed physicians to have hospital privileges in Delaware. In New Castle County, this means privileges at MCD and/or St. Francis Hospital.

It is this basic fact which ultimately confers market power "the power `to force a purchaser to do something that he would not do in a competitive market[,]'" Eastman Kodak, 504 U.S. at 464, 112 S. Ct. at 2080 (quoting Jefferson Parish Hosp. v. Hyde, 466 U.S. 2, 14, 104 S. Ct. 1551, 1559, 80 L. Ed. 2d 2 (1984)) upon MCD. Here, licensure and privileges define the actual market reality long favored in antitrust law. Id. at 464-65, 112 S. Ct. at 2080-81.

Defendants respond that Dr. Solow's opinion is the result of "outdated" assumptions specific to the health care industry. D.I. 121, at 25. Health care is no longer a localized industry, defendants assert, in fact the opposite is true: the rise of managed care is evidence the relevant market is larger than New Castle County. Id. Managed care entities have every reason to be price conscious, defendants say, and they are not shy about using the power of the purse to influence consumers' choice of hospitals. Id. To buttress their position, defendants point to U.S. Supreme Court and Third Circuit Court of Appeals authority indicating consumer apathy or ignorance is not a source of antitrust market power. See Jefferson Parish Hosp. v. Hyde, 466 U.S. 2, 104 S. Ct. 1551, 80 L. Ed. 2d 2 (1984); Town Sound and Custom Tops, Inc. v. Chrysler Motors Corp., 959 F.2d 468 (3d Cir.), cert. denied, 506 U.S. 868, 113 S. Ct. 196, 121 L. Ed. 2d 139 (1992).

Defendants miss the point. To be competitive, managed care entities must have MCD as an approved provider in New Castle County. While they might bargain for lower prices, managed care entities would not, as a practical matter, refuse to approve MCD in the event MCD raised its prices; even if all physicians had privileges at St. Francis Hospital, that hospital simply does not have anywhere close to a sufficient number of beds for New Castle County admittees. See B-593 (St. Francis has approximately 300 beds, compared to approximately 700 at MCD).^[15]

*544 The market reality, as explained above, is doctors' hospital referrals are limited by hospital privileges and Delaware licensure requirements managed care must work within that framework. For this reason, defendants' argument predicated on Jefferson Parish Hosp. and Town Sound and Custom Tops both tying cases has no relevance to plaintiff's leveraging theory. A reasonable juror could conclude consumers of health care, unlike consumers of shoes for example, would not chose to leave their local hospital market as a result of a price increase. See 2A Areeda, supra § 553, at 242 ("[T]ribunals often find that a single municipality was a relevant geographic market for the provision of hospital services.... Even for hospital services that can be deferred, the hospital is often selected by the patient's doctor, who may be relatively insensitive to hospital prices and who may not have staff privileges in other towns. Hence, most patients would not turn to remote hospitals in response to a local price increase."). Defendants argument that plaintiff failed to correctly define the upstream geographic market cannot serve as the basis for a grant of summary judgment.

Defendants assert even adopting New Castle County as the relevant geographic market, MCD's market share is only 62%, and this is not sufficient to prove monopoly power. D.I. 121, at 28 (citing A-708, A-967). Plaintiff places MCD's market share in 1992 at 71.5%. B-529. Further, plaintiff explains, MCD acquired Riverside Hospital in 1995, increasing its market share to approximately 75%. D.I. 126, at 33 (citing B-581; B-529-31). Using figures from before MCD acquired Riverside Hospital, Dr. Solow found Riverside and St. Francis hospitals combined had insufficient capacity to cover the demand in the event MCD raised its prices. B-581-82. This is further evidence MCD possesses monopoly power, plaintiff says. D.I. 126, at 33 & n. 18.

The Third Circuit Court of Appeals has found as a matter of law that a market share under 55%, in the absence of other factors, does not establish monopoly power. Fineman, 980 F.2d at 201. As by all accounts MCD's market share is greater than 55%, and plaintiff has adduced evidence of other factors contributing to monopoly power, this likewise cannot provide a ground for summary judgment.

A second element of plaintiff's leveraging claim is proof that use of the "upstream" monopoly power has resulted in "actual or threatened" monopoly power in the "downstream" home infusion therapy market. Fineman, 980 F.2d at 206.^[16] Like any other claim of monopolization or attempted monopolization, evaluation of defendant ISD's monopoly power in the downstream market requires a properly defined antitrust relevant product and geographic market, see Spectrum Sports, 506 U.S. at 457-59, 113 S.Ct. at 891-92; once again, defendants urge plaintiff's geographic market definition is flawed.^[17] Plaintiff has asserted the relevant geographic market is New Castle County.

In support of plaintiff's position, Dr. Solow pointed out 75.9% of ISD's patients are residents of New Castle County. B-584. That figure, he said, met the weak LOFI part of the E-H test. Id. He continued, "[E]ven home infusion therapy providers located inside New Castle County have had difficulty in gaining access to Medical Center of Delaware patients, so it is unlikely that providers located further away would be able to serve those patients." Id. As a result, he stated, the LIFO portion of the E-H test also was met. Id.

As to forward-looking analysis, Dr. Solow stated:

*545 In terms of geographic market, we must address whether home infusion patients living in New Castle County would leave that location to receive home infusion therapy, whether patients living outside of New Castle County but being served by New Castle County infusion therapy providers would switch to providers located outside of New Castle County, and whether home infusion providers located outside New Castle County would be able to serve patients located within New Castle County.

B-583-84. Of these possibilities, he opined, "only the second is a reasonable possibility." B-584. Dr. Solow concluded, "It is my opinion that a single seller of home infusion therapy in New Castle County that could restrict access to a large share of the patients referred for home infusion in that area would have market power...." Id.

Plaintiff's market definition, as set forth here, is flawed. Although Dr. Solow purported to rely on the E-H test, he applied the LOFI prong of the test only to ISD, instead of considering all home infusion therapy services produced in New Castle County the proposed geographic market. Information as to the residences of ISD patients does not help define New Castle County as the antitrust relevant geographic market, it is merely evidence of defendant ISD's service area. See Home Health Specialists, 1994 WL 463406, at *4 (fact that vast majority of defendant's customers live in Delaware County, Pennsylvania, does not establish a geographic market; that fact merely points to Delaware County as defendant's service area); see also Hovenkamp, supra n. 10, at 113-14. Dr. Solow completed the LIFO prong of the E-H test with the conclusory statement: "[E]ven home infusion therapy providers located inside New Castle County have had difficulty in gaining access to Medical Center of Delaware patients, so it is unlikely that providers located further away would be able to serve those patients." B-584. This purported application of the E-H test is, to say the least, unconvincing.

Moreover although not adequately addressed by either party it appears to the Court that home infusion therapy presents an atypical market for analysis under the E-H test. The market is unusual because home infusion therapy consumers are not mobile; by definition, they must receive infusion services in their homes. They cannot physically travel to another market to purchase home infusion therapy services in the event ISD raises its prices.^[18] It further appears from deposition testimony in the record that the offices of home infusion therapy providers are merely administrative, while the actual delivery of infusion therapy occurs in the home. B-150-152. The prong of the E-H test that measures the percentage of the goods or services produced outside the market that were purchased by consumers within the market does not aid the analysis, therefore, because the services are always produced in the same location as the consumer's residence.

As noted above, the parties have not weighed in on how this unique aspect of the home infusion therapy market affects the geographic market analysis. Yet there is no dispute that market definition involves identification of "producers that provide customers of a defendant firm (or firms) with alternative sources for the defendant's product or service[,]" 2A Areeda, supra ¶ 530a, at 150. As a matter of logic, therefore, the geographic market inquiry in this case must focus on which providers are willing to provide services to consumers in the potential market. Any home infusion therapy patient, even one already using ISD, is free to switch; if providers are willing to come from the other two counties of Delaware, as well as nearby Pennsylvania, New Jersey and Maryland, there is no reason to limit the relevant geographic market in this case to New Castle County.^[19]

The record is silent on the subject of which providers serve, or are willing to serve, consumers of home infusion therapy located in New Castle County. The only relevant submission *546 by plaintiff is its argument that the referral process at MCD acted to limit choices available to MCD consumers of home infusion therapy. Similar to plaintiff's position with respect to the upstream inpatient services market, it argues because of information costs, consumers of health care often do not possess the full range of choices that seem logically to be available to them. Plaintiff's point is MCD patients were steered to ISD, and therefore, other providers particularly those outside of New Castle County could not have gained access to them to provide home infusion services.

While this is not an implausible argument, it also does not support plaintiff's definition of the relevant geographic market as New Castle County. At most, the argument entitles plaintiff to an inference that a subgroup of New Castle County residents were steered to ISD and thus did not enjoy full choice of home infusion therapy providers. That inference, coupled with the fact approximately 75% of ISD patients reside in New Castle County, is not sufficient to prove the relevant geographic market is New Castle County. Unlike plaintiff's assertion regarding the "upstream" hospital inpatient services market, plaintiff just has not put enough information in the record to support its geographic market definition.

The essential facilities doctrine "is implicated `where a monopolist controls a facility that its competitors need access to if they are to compete effectively.'" Delaware Health Care, 893 F. Supp. at 1287 (quoting Monarch Entertainment Bureau, Inc. v. New Jersey Highway Auth., 715 F. Supp. 1290, 1300 (D.N.J.), aff'd, 893 F.2d 1331 (3d Cir.1989)). The elements of an essential facilities claim are:

(1) control of the essential facility by a monopolist; (2) the competitor's inability practically or reasonably to duplicate the essential facility; (3) denial of the use of the facility to a competitor; and (4) the feasibility of providing the facility.

Id.; accord Ideal Dairy Farms, 90 F.3d 737, 748 (3d Cir.1996). In this case, the claimed essential facility is "the inpatient discharge and referral process at MCD." D.I. 126, at 44.

Defendants make several arguments that summary judgment must be granted against plaintiff. They argue with respect to the first required element, MCD is not a monopolist. D.I. 121, at 44. Earlier in this opinion, it was held an issue of material fact exists with respect to MCD's monopoly power;^[20] accordingly, the Court must reject defendants' same argument directed to essential facilities.

Defendants next urge MCD patients are not, as a matter of law, an essential facility. Id. "An `essential facility' is one which is not merely helpful but vital to the claimant's competitive viability." Cyber Promotions, Inc. v. America Online, Inc., 948 F. Supp. 456, 463 (E.D.Pa.1996) (citation omitted). The denial of access must inflict a "severe handicap" on the competitor, one "that threaten[s] to eliminate competition in the market...." Advanced Health-Care Servs., Inc. v. Giles Mem'l Hosp., 846 F. Supp. 488, 498 (W.D.Va.1994).

The essential facilities doctrine has been traced to the Supreme Court's opinion in United States v. Terminal R.R. Assoc., 224 U.S. 383, 32 S. Ct. 507, 56 L. Ed. 810 (1912). See Areeda, et al., supra ¶ 736.1b, at 620 (1995 Supp.). In Terminal Railroad, the Court ordered an association controlling all three means of crossing the Mississippi River into St. Louis to grant reasonable access to all competing railroads. Although the Court did not use the term "essential facility," it is clear the river passages were considered essential to competition among the railroads because it was impossible to enter or pass through St. Louis without crossing the river. See 224 U.S. at 397, 32 S. Ct. at 510. The Court stated:

The physical conditions which compel the use of the combined system by every road which desires to cross the river, either to server the commerce of the city or to connect with lines separated by the river, *547 is the factor which gives greatest color to the unlawfulness of the combination as now controlled and operated.

Id. at 398, 32 S. Ct. at 511.

A leading antitrust treatise often quoted in essential facilities opinions urges the essential facilities doctrine should "at most" extend to "facilities that are a natural monopoly, facilities whose duplication is forbidden by law, and perhaps those that are publicly subsidized and thus could not practicably be built privately." Areeda et al. supra ¶ 736.2b, at 645 (1995 Supp.); see Cyber Promotions, 948 F. Supp. at 460; Monarch Entertainment Bureau, 715 F. Supp. at 1300; but see Alaska Airlines, Inc. v. United Airlines, Inc., 948 F.2d 536, 544 n. 10 (9th Cir.1991) (suggesting this strict limitation on the essential facilities doctrine is not supported by Supreme Court case law).

Defendants argue the MCD patient discharge process is not an essential facility because as DHC principal Lawrence Carroll conceded referrals for patients come from a variety of sources including physicians, insurers, case managers, and other patients. A-1454.

The Western District of Virginia's Advanced Health-Care is factually similar to this one. Plaintiff, in the business of durable medical equipment ("DME") sued Giles Memorial Hospital and Medserv, another DME provider.^[21] The two defendants had formed a joint venture to provide DME to residents in the service area of the hospital. Under the agreement Medserv rented office space in the hospital and did business under the name Home Connections. A strict policy was maintained under which patients were advised of all DME suppliers in the community, no particular provider was recommended, and the patient was required to pick the provider. 846 F. Supp. at 495.

The court rejected plaintiff's essential facilities argument. It stated: "Access to patients was not essential to competition in Giles County, nor do Giles Memorial's policies completely deny [plaintiff] patient access, nor is it feasible to permit [the plaintiff] to solicit business directly in patient's hospital rooms." Id. at 498. As proof access to Giles Memorial patients was not essential, the court relied on market information and found, "[t]hroughout Home Connections' existence, rival DME companies continued to do substantial business and ... were consistently gaining market share." Id.

The reasoning of Advanced Health-Care is equally applicable here. DHC's essential facilities claim must fail because the patient referral and discharge process at MCD is not an essential facility. First, similar to Advanced Health-Care, the market data does not support a finding that MCD was an essential facility during the period ISD was gaining market share. At best, ISD only possessed a 59% market share of home infusion therapy patients in New Castle County a figure which dropped below 50% in 1995. B-505; B-540. That leaves open approximately 50% of the New Castle County home infusion therapy market for ISD's rivals.

Additionally, following the reasoning of Advanced Health-Care, plaintiff's statistics demonstrate other home health care providers maintained, or increased, their market share during the relevant time. For example, Brandywine had a market share of 15.4% in 1993, 12.4% in 1994, and 17% in 1995. See B-542. Apria/Homedco maintained a 2.6% market share in New Castle County for 1993 and 1994, and in 1995, its market share jumped to 4.9%. Id.

More importantly, however, plaintiff has not limited itself to seeking referrals from hospitals in New Castle County.^[22] Plaintiff *548 defined its service area as the entire state of Delaware. A-1179. Plaintiff also stated it has solicited hospitals in three states surrounding Delaware Pennsylvania, New Jersey and Maryland for referrals of the business of Delaware residents. A-1178-80. Plaintiff's statistics indicate 24,239 Delaware residents used hospitals in Delaware's other two counties, Sussex and Kent counties. B-533. For comparison, 30,361 Delaware residents used MCD in 1992, and 11,376 Delaware residents used St. Francis Hospital. Id. While it is likely only a percentage of all hospital patients require home infusion therapy, the approximately 25,000 patients in the other two counties of Delaware, plus the 11,376 patients who used St. Francis, represented a sizeable source of business for plaintiff.

Accordingly, accepting plaintiff's alleged inability to gain referrals for MCD patients as true, other sources of business for plaintiff exist in a sufficient amount that the patient discharge and referral process at MCD cannot be considered an "essential facility." Given these other sources of business to plaintiff within plaintiff's service area, it can not be said access to MCD's patient discharge process is "vital to [DHC's] competitive viability"; nor can it be said denial of such access inflicts a "severe handicap ... that threatens to eliminate competition in the market." See Cyber Promotions, 948 F. Supp. at 463; Advanced Health-Care, 846 F. Supp. at 498. Plaintiff has not put forth sufficient evidence here to survive summary judgment on its essential facilities claim.^[23] Accordingly, summary judgment will be granted for defendants on plaintiff's federal claims.

Having granted summary judgment for defendants on all plaintiff's federal claims, the Court is left to consider whether, under 28 U.S.C. § 1367(c) (3), it should decline to decide the state law claim. Section 1367(c) allows a federal district court, in its discretion, to decline to exercise supplemental jurisdiction over state law claims when the court has dismissed all claims over which it had original jurisdiction. The Third Circuit Court of Appeals, finding section 1367(c) a codification of "preexisting pendent jurisdiction law," has opined, "[W]here the claim over which the district court has original jurisdiction is dismissed before trial, the district court must decline to decide the pendent state claims unless considerations of judicial economy, convenience, and fairness to the parties provide an affirmative justification for doing so." Borough of West Mifflin v. Lancaster, 45 F.3d 780, 788 (3d Cir.1995).

The parties have not raised any considerations that would cause the Court to deviate from the general rule and instead exercise supplemental jurisdiction over plaintiff's state law claims. Accordingly, supplemental jurisdiction will not be exercised and plaintiff's state law claims will be dismissed.

In conclusion, the Court grants defendants' motion for summary judgment on both of plaintiff's antitrust claims and dismisses plaintiff's state law claim for tortious interference with contract.

[1] Citations to documents with the prefix "A" refer to documents found in defendants' appendices. Citations to documents with the prefix "B" refer to documents found in plaintiff's appendices.

[2] Plaintiff maintains no claims against VNA directly; VNA is a defendant only by virtue of its participation in the alleged leveraging scheme.

[3] Discharge planners are hospital staff members who assist patients in obtaining services for post-discharge.

[4] Both hospitals comprising MCD, Wilmington and Christiana hospitals, are located in New Castle County, Delaware.

[5] An approximate number will be used in this opinion because the number fluctuated during the relevant time period.

[6] In addition to revenue from infusion pharmaceutical products, DHC also earned revenues from nursing services. In 1993, gross revenue from nursing services was $339,570; in 1994, it was $348,782, and in 1995, DHC earned $355,224 from nursing services. B-704. These figures are not relevant in comparison to ISD's revenues because ISD only provided infusion pharmaceutical products and did not provide nursing services. As noted above, plaintiff no longer asserts any direct claims against VNA, the entity that provides nursing services in conjunction with ISD products. See supra n. 2.

[7] Section 2 of the Sherman Act sanctions, "[e]very person who shall monopolize, or attempt to monopolize, or combine or conspire with any other person or persons to monopolize any part of the trade or commerce among the several States...." 15 U.S.C. § 2.

[8] In its complaint, plaintiff also included allegations regarding vertical integration. Plaintiff informed the Court at oral argument it does not wish to pursue this claim.

[9] As noted in an earlier opinion in this case, the Court is aware the Third Circuit Court of Appeals, in Advo, Inc. v. Philadelphia Newspapers, Inc., 51 F.3d 1191, 1203 (3d Cir.1995), indicated leveraging claims were under attack for being "economically groundless." See Delaware Health Care, Inc. v. MCD Holding Co., 893 F. Supp. 1279, 1288 n. 2 (D.Del.1995). Nevertheless, the Court does not read Advo to repudiate all leveraging claims. Id.

[10] This distinction is most vividly described in Herbert Hovenkamp, Federal Antitrust Policy § 3.6d, at 113-14 (1994) (footnotes omitted) (quoted in Bathke v. Casey's Gen. Stores, Inc. 64 F.3d 340, 346 (8th Cir.1995)):

Fifteen miles outside of City A is a small town, Town B, which contains a single shoe store, Smith's Clothing. The only people who ever shop in Smith's Clothing are residents of Town B. When Smith's is accused of monopolization, the plaintiffs argue that Town B defines the relevant geographic market, since all of the store's customers come from there. In that case, Smith's market share is 100%.

But further inquiry shows the following. Last year 800 residents of Town B purchased shoes. 400 of them purchased from Smith's Clothing, and the other 400 purchased from the numerous shoe stores in City A. Note that this conclusion is absolutely consistent with the proposition that Smith's "trade area" is Town B. The only customers Smith's had were from Town B, assuming no one drove from City A to buy shoes at Smith's. In fact, Smith's is probably in close competition with the City A shoe stores. To the extent Town B residents go into City A to work or to shop, they regard the City shoe stores as interchangeable, and would respond to any local price increase by purchasing even more of their shoes in City A.

In sum, "trade area" considers the extent to which customers will travel in order to do business at Smith's. "Relevant market" considers the extent to which customers will travel in order to avoid doing business at Smith's. Unfortunately, this means that using discovery to obtain the addresses of a seller's customers seldom provides us with useful information about geographic market. What we really need to know is the extent to which people from the immediate area can readily turn to alternative sellers....

[11] There are eight hospitals in New Castle County. Three are psychiatric hospitals, and thus compete with MCD but only to the limited extent MCD offers psychiatric inpatient hospital care. A fourth, the Veterans Affairs Medical Center, was not considered by plaintiff's expert to compete with MCD because it serves a different population. Of the remaining three, two are affiliated with MCD A.I. DuPont Institute, a children's hospital, and Riverside Hospital, which was acquired by MCD in 1995. The remaining hospital is St. Francis. See B-580 (Solow Decl.).

[12] The Guidelines "outline the present enforcement policy of the Department of Justice and the Federal Trade Commission ... concerning horizontal acquisitions and mergers.... They describe the analytical framework and specific standards normally used by the Agency in analyzing mergers." A-482 (Merger Guidelines).

[13] The label "Little In From Outside" may seem somewhat of a misnomer when what is measured is actually the percentage of customers who remain in the market or leave the market to purchase goods or services. However, the aim of "Little In From Outside" is to calculate the proportion of sales of goods or services purchased in the provisional market that were produced by suppliers located outside the provisional market which may be accomplished by measuring the percentage of customers who left the market when they purchased the goods or services at issue. See B-574.

[14] Similar to the LIFO label, "Little Out From Inside" seems the wrong designation for this calculation. However, its aim is to calculate the percentage of goods or services produced within the provisional market that are sold to consumers residing in the market versus those consumers who reside outside the provisional market. That calculation may be achieved by measuring the percentage of consumers who entered the market to purchase services. See id.

[15] The figure for MCD includes beds at the A.I. DuPont Institute, an MCD affiliate, and Riverside Hospital, which was acquired by MCD in 1995. See supra n. 11; B-580.

[16] In requiring proof of this element, the Third Circuit Court of Appeals differs from the Second and Sixth Circuits, which require a lesser showing of "competitive advantage" in the "downstream" market, see Kerasotes Michigan Theatres, Inc. v. National Amusements, Inc., 854 F.2d 135 (6th Cir.1988); Berkey Photo, Inc. v. Eastman Kodak Co., 603 F.2d 263 (2d Cir.1979), cert. denied, 444 U.S. 1093, 100 S. Ct. 1061, 62 L. Ed. 2d 783 (1980).

[17] The parties agree the product market is home infusion therapy. D.I. 121, at 31-32; D.I. 126, at 26.

[18] Of course, patients are free to switch from one home infusion therapy provider to another.

[19] The Court takes judicial notice that any geographic point in New Castle County is no more than 30 minutes by car from the Pennsylvania or Maryland borders.

[20] See supra pp. 540-44.

[21] DME refers to durable medical equipment and includes "canes, crutches, oxygen equipment, wheelchairs, walkers, hospital beds and other items used by persons recuperating at home from an illness or accident." Advanced Health-Care Services, 846 F. Supp. at 492 n. 1.

[22] The fact that hospitals outside New Castle County are a potential source of business for plaintiff is not inconsistent with the Court's earlier holding a question of fact exists as to whether New Castle County is the relevant geographic market for inpatient hospital services in this case. The relevant geographic market measures where consumers could look for services; however, DHC does not face similar constraints in where it can market patients.

[23] Because of the Court's holdings in this opinion granting summary judgment for defendants, it is not necessary to address defendants' other arguments that plaintiff has failed to introduce sufficient evidence of anticompetitive conduct and antitrust injury.