|This document is available in two formats: this web page (for browsing content) and PDF (comparable to original document formatting). To view the PDF you will need Acrobat Reader, which may be downloaded from the Adobe site. For an official signed copy, please contact the Antitrust Documents Group.|
Tad R. Callister, Esquire
Dear Mr. Callister:
This letter responds to your request, pursuant to the Department of Justice's Business Review Procedure, 28 C.F.R. § 50.6, for a statement of the Department's enforcement intentions regarding a proposal by five Orange County anesthesia medical groups ("the Groups") to contract collectively with managed-care customers, through a single, price-setting organization, Orange Los Angeles Medical Group, Inc. ("ORLA").(1)
You made this request while the Department's Antitrust Division was conducting an investigation of the Groups' joint venture activities as a result of a complaint by a health benefits plan. Since you assured us that ORLA had taken no steps to implement its business plans, we agreed to provide a business review letter addressing the antitrust issues posed by the proposed joint venture.
After a thorough investigation, and careful consideration of the information you have provided, we conclude that there is a substantial likelihood that the proposed joint venture will harm competition and little prospect that it would yield procompetitive benefits such as improved efficiency or reduced anesthesia rates. Consequently, the Department cannot state a present intention not to challenge ORLA if its proposed operations are implemented.(2)Factual Background
Description of the Parties And The
Each of the five Groups is a large (13-30 member), financially integrated, hospital-based medical practice. Each is the exclusive or dominant supplier of anesthesia services at the hospital or hospitals it serves.(3) According to the managed care payers we interviewed, the hospitals served by the five
ORLA-affiliated Groups ("the Affected Hospitals") comprise six of Orange County's major hospitals, each of which is characterized by a heavy caseload, a wide range of inpatient acute-care services including sophisticated procedures such as open-heart surgery or high-level trauma care, and a high level of demand among managed care health plans.(4) Together, the six Affected Hospitals account for a large proportion of all managed care plans' Orange County hospital expenditures.
Each of the Affected Hospitals uses a "single group" anesthesiology system to obtain dependable, high-quality anesthesia services and to achieve efficient utilization of their operating rooms. Under this system, the hospital delegates extensive or complete responsibility to a single anesthesia group for screening, hiring and monitoring new anesthesiologists; for ongoing peer review; and for the assignment of the hospital's anesthesiologists. The Affected Hospitals switched to the single group system from an "open staff" approach in which independent anesthesiologists were individually admitted to the hospital's active medical staff and assigned to cases by surgeon request or by some type of rotation system. Both the hospital officials and the Group representatives we interviewed agreed that the single group system has achieved better quality control and a substantial improvement in efficiency. They explained that under the single group system the surgeons are willing to relinquish control over anesthesia assignments to the hospital's anesthesia group,(5) and that enables the group to assign anesthesiologists in a manner that efficiently utilizes the hospital's anesthesia staff and operating rooms.
The Proposed Joint Venture
In the proposed joint venture, each of the five Groups would continue to operate as the exclusive or primary anesthesia supplier at the hospital(s) it currently serves. Each one would continue to function independently with respect to traditional indemnity or self-insured patients at its hospital. However, all five Groups would contract with managed care customers exclusively through ORLA. For any such customer, ORLA would negotiate a single agreement covering all five Groups. The customer would pay ORLA for anesthesia services provided by any of the Groups, and ORLA would distribute the proceeds to the Groups that provided services to that customer.
ORLA states that it will offer managed care customers agreements in which it will undertake substantial shared financial risk, and that it will distribute the proceeds of those agreements in a manner that will create a shared incentive among the participating Groups to provide their services efficiently. For managed care contracts, ORLA will have the authority to impose utilization review procedures and practice protocols, and to require efficiency-related and quality-related changes in the Groups' practices.
ORLA cites four principal benefits of the proposed joint venture. It believes its plan to offer a single risk-sharing agreement covering all five groups at a single price will facilitate HMOs' ability to contract directly for anesthesia services on a risk-sharing basis as opposed to obtaining those services through primary provider organizations as part of the bundle of physician services those organizations provide under full risk capitation arrangements. It will reduce the Groups' input costs by jointly purchasing billing, insurance and legal services for all five Groups. It will jointly pursue supplemental sources of income such as conducting area-wide drug trials for pharmaceutical companies. Finally, it will monitor the Groups' efficiency and will adopt efficiency-enhancing procedures such as cost-reducing practice protocols.Rule of Reason Analysis
Based upon your representations, it appears that this joint venture qualifies for rule of reason analysis because its proposed operations entail substantial financial risk-sharing among participating providers in a manner that will create a shared incentive to operate efficiently. See Statements of Enforcement Policy and Analytical Principles Relating to Health Care and Antitrust ("Health Care Antitrust Statements"), issued by the Department of Justice and the Federal Trade Commission on September 27, 1994, pp. 71 and 91.(6) Under a rule of reason analysis, we determine whether the joint venture is likely to have a substantial anticompetitive effect and, if so, whether that potential effect is outweighed by any procompetitive efficiencies resulting from the joint venture.
In analyzing the competitive effects of a joint venture, we first identify the relevant market in which the participants of the joint venture operate. To do this, we assess purchasers' ability and willingness to use alternative providers of the joint venture's service. We identify these alternative providers or substitutes by considering service characteristics (service market) and location (geographic market).
From our investigation, it appears that the Groups participating in ORLA have, primarily, two significant types of price-sensitive customers: (1) primary provider organizations (such as IPAs and large, managed care medical groups) that are responsible for virtually all medical services (including anesthesia services) required by their HMO-insured patients,(7) and (2) managed care health plans (such as PPOs) that contract directly with the Groups for services provided to their enrollees. (Less commonly, the hospital is a price-sensitive customer of its anesthesia group in cases where the hospital, in conjunction with its physicians, offers managed care plans a flat "case rate" for certain procedures.)
If the Group that serves one of the Affected Hospitals demands supracompetitive rates, a managed care customer potentially could respond by redirecting some if not all of its cases to another major Orange County hospital with lower anesthesia rates. Or, an Affected Hospital that is faced with the prospect of losing managed care business to a hospital with lower anesthesia rates could respond by replacing its existing anesthesia Group with a lower priced alternative.
Spokespersons for each of the Affected Hospitals stressed that their hospital generally would not respond to a small but significant price difference by adding individual anesthesiologists -- i.e., by returning to the open staff arrangement that they replaced with the single group system -- or by replacing the hospital's existing Group with an anesthesia group that lacks the requisite management experience as a group, range of experience among its members or confidence of the hospital's surgeons, since either course would require them to forego the substantial quality and efficiency benefits the hospital has derived from the single group system.(8) Instead, each of them stated that, to avoid such supracompetitive demands by their existing Group, they would turn only to a replacement group that is comparable to the hospital's existing Group in size, in its member-anesthesiologists' quality and range of experience, in the group's experience in managing anesthesia services at a comparably large, busy and sophisticated hospital and in its reputation and acceptability to the hospital's surgeons.
In short, the relevant service market definition in this case must reflect the characteristics of the Groups that have enabled the Affected Hospitals to achieve the substantial quality and efficiency benefits they derive from the single group system. Hence, we conclude that the relevant service market for assessing the proposed joint venture is managed anesthesia services provided by adequately sized, financially integrated anesthesia medical groups that have a reputation and range of experience comparable to the existing Groups and are known and acceptable to the surgeons at the Affected Hospitals.
The hospital and other industry representatives we interviewed also told us that to be an adequate substitute for the Groups at the Affected hospitals, it is important that an anesthesia group be located close enough to the hospital to provide call-duty coverage(9) and to be familiar enough to the hospital surgeons to give them confidence that the group's anesthesiologists are able to meet the surgeons' particular needs for anesthesia services.(10)
Consequently, we have concluded that, for both travel time and surgeon familiarity reasons, the relevant geographic market is no larger than Orange County and perhaps even a smaller area.(11)
In an "Orange County" geographic market, six firms meet the service market definition: the five Groups and South Coast Anesthesia Consultants ("SCAC"), which is the exclusive group at Saddleback Memorial Hospital and which recently was brought in to serve Monarch IPA at Mission Hospital. If the Groups proceed with the joint venture, there would be only two competitors (ORLA and SCAC) available in the market. If we were to assume smaller relevant geographic markets on the basis of the travel time requirement, the joint venture's impact on the number of competitive alternatives available to managed care customers would be as great as, or greater than, it would be in an Orange County market.(12)
We also considered whether a number of national firms that provide anesthesia management services for hospitals are good substitutes for the Groups. Although those national firms typically serve smaller and less sophisticated hospitals, they assert that they are capable of managing anesthesia services at hospitals like the Affected Hospitals.
However, spokespersons for each of the Affected Hospitals stated, emphatically, that they would not turn to one of those national firms as a substitute for their existing Group because even the best-known of those firms (such as Premier Anesthesia) are viewed by their surgeons and medical staffs as "unknown quantities."(13) They stated that replacing their existing Group with one of the national firms would jeopardize the important quality and efficiency benefits their hospitals have achieved under a single group system, using a group that has acquired the confidence of the hospital's surgeons.(14)
Our extensive investigation of this matter leads us to the conclusion that ORLA's proposed operations are likely to cause serious anticompetitive effects. In Orange County, ORLA will include five of the six substantially-sized, financially integrated anesthesia groups that are able to provide the kind of sophisticated managed anesthesia services required by the Affected Hospitals, leaving only a single group (South Coast Anesthesia Consultants) as a competitor. Such market dominance, further strengthened by ORLA's exclusive nature, creates a real risk that ORLA could successfully demand supracompetitive rates and other contract terms.(15) In smaller relevant geographic markets, the joint venture may create a monopoly with respect to at least some of the Affected Hospitals. Spokespersons for the health plans, IPAs and Affected Hospitals were unanimous in their concern that the proposed joint venture would cause substantial anticompetitive harm.
The joint venture's substantial reduction in the number of existing market participants would provide less cause for concern if it appeared that new competitors would likely be drawn into the market if ORLA demanded supracompetitive terms. However, under current market conditions, entry by anesthesia groups that are realistic substitutes for ORLA is not likely to occur on a scale that would offset the joint venture's substantial reduction in the number of existing competitors.(16)
The final step in the rule of reason analysis is to weigh the prospect of a significant anticompetitive outcome against any substantial efficiencies that are likely to result from the formation of ORLA and could not be achieved without it.
We considered possible efficiencies arising from implementation of ORLA's risk-sharing proposal, but concluded that any such efficiencies would be minimal. Each of the Groups is able to offer risk-sharing agreements individually,(17) and the proposed combination of those five competing Groups into a single, price-setting joint venture would not materially augment their ability to provide such agreements.(18)
We also considered ORLA's contention that it will be able to reduce the input costs of the Groups by such means as seeking volume-purchase discounts for services they use or by undertaking area-wide drug trials. However, the Groups could pursue those objectives through cooperative ventures that do not entail the formation of a competition-reducing joint venture to collectively set the price at which they will provide their services to managed care customers.
In addition, we considered the Groups' intention to undertake jointly such efficiency-enhancing activities as the development of cost-reducing practice protocols. Joint pricing action is not necessary for the groups to pursue those cost-reducing activities. Moreover, the information provided by the relevant hospital officials indicates that joint action may hamper rather than enhance their ability to do so.(19) In fact, one of those hospital officials observed, it is the advent of competition among financially integrated physician organizations (such as IPAs) for the business of managed care customers that has prompted physicians to reexamine their existing practice methods, and to adopt cost-reducing protocols and utilization review procedures -- not the formation of collective bargaining units that encompass all or most of their managed care customers' competitive alternatives.Conclusion
For the reasons explained in this letter, we conclude that the proposed joint venture poses a substantial risk of competitive harm, and that it will yield no procompetitive benefits that outweigh that risk. Thus, on the basis of the market conditions revealed by our investigation, we cannot state a present intention not to challenge it. If the Groups proceed with the joint venture, the Department would decide whether to challenge it on the basis of then-existing market conditions.
This statement is made in accordance with the Department's Business Review Procedure, 28 C.F.R. § 50.6, a copy of which is enclosed. Pursuant to its terms, your business review request and this letter will be made publicly available immediately.
Your supporting documents will be publicly available within 30 days of the date of this letter unless you request that any part of the material be withheld in accordance with Paragraph 10(c) of the Business Review Procedure.
1. The Groups are: Allied Anesthesia Medical Group (30 members), serving St. Joseph's Hospital (and Children's Hospital of Orange County, which uses the operating rooms of St. Joseph's Hospital); Newport Harbor Anesthesia Consultants (27 members), serving Hoag Memorial Hospital; Mission Anesthesia Medical Association or "MAMA" (14 members), serving Mission Community Hospital; Fullerton Anesthesia Medical Group, Inc. (15 members), serving St. Jude's Hospital; and, Independent Anesthesiology Medical Association or "IAMA," serving Western Medical Center. (ORLA lists IAMA as having 13 members; IAMA supplied a list of 18 members. See also n. 3 below.)
ORLA also lists a sixth participating Group, South Coast Orange County Anesthesia Medical Group, Inc., which has only two members, who serve at South Coast Medical Center (one of many smaller Orange County hospitals offering a more limited range of acute-care services than the six hospitals cited above) along with at least four other anesthesiologists who are on staff at that hospital. (When ORLA submitted its request for a business review, it listed that group with six members, but a spokesperson for the group told us the group had been dissolved.) We do not view the participation in ORLA of two South Coast Medical Center anesthesiologists as significant to our analysis of the proposed joint venture.
2. The Antitrust Division's investigation included a review of documents submitted by ORLA, each of the ORLA Groups, and South Coast Anesthesia Consultants ("SCAC"), a large, financially integrated Orange County anesthesia group that is not participating in ORLA. We interviewed spokespersons for ORLA, for each Group, for each of the major Orange County hospitals that are served by the ORLA Groups, for SCAC, and for two anesthesia groups based at hospitals located outside Orange County that ORLA identified as existing or potential competitors. Finally, we interviewed approximately 15 physician organizations (i.e., IPAs and large medical groups) and health benefits plans that account for a large proportion of the managed care business of the six major Orange County hospitals served by the five ORLA Groups.
3. MAMA is the dominant anesthesia supplier at Mission Community Hospital. MAMA was the exclusive anesthesia supplier at Mission (with the exception of a single non-MAMA staff member) until recently, when Mission permitted an IPA to contract with another anesthesia group (South Coast Anesthesia Consultants, based at nearby Saddleback Memorial Hospital) to serve it at Mission. Each of the other four Groups is, either formally or in practice, the exclusive anesthesia supplier at the Affected Hospital(s) it serves.
ORLA asserts that IAMA, the Group at Western Medical Center, has only 13 members, comprising only about half of Western Med's anesthesiologists, and that the other half are free to contract with the hospital's managed care customers in competition with IAMA. However, a Western Med. spokesperson stated that IAMA is responsible for, and supervises, all anesthesia services at the hospital, and that only IAMA may contract with the hospital's managed care customers.
4. The payers identified eight major Orange County hospitals. Their assessment of the hospitals in Orange County is supported by such data as inpatient surgical minutes, acute-care patient days, average daily census, staffed acute-care beds and tertiary-level services such as high-level trauma care or open-heart surgery with bypass. A ninth Orange County hospital had similar data, but, according to the payers we interviewed, it does not share the eight major hospitals' reputation or level of demand among managed care payers.
Two of the eight major hospitals are not served by an ORLA-affiliated Group: the University of California, Irvine, Medical Center ("UCI") and Saddleback Memorial Hospital. UCI is served by a staff of employee anesthesiologists (largely faculty members). Saddleback is served by the only large, financially integrated Orange County anesthesia group (South County Anesthesia Consultants or "SCAC") that is not participating in the joint venture. The ninth hospital is not served by a single, financially integrated anesthesia group.
5. The surgeons are willing to relinquish control over scheduling to the group for two reasons: First, by carefully screening and monitoring new hires, and closely controlling their assignment to particular cases, the group in effect "certifies" a new anesthesiologist as appropriate for any case to which he or she is assigned. Second, by gradually integrating new hires into the anesthesia schedule, the hospital's surgeons become familiar with each of them before they are assigned to difficult cases.
In addition, a single, financially integrated anesthesia group can make income adjustments (such as spreading the cost of unpaid cases) which minimizes disputes among the anesthesiologists over assignments, and reduces the pressure to use assignments as a means of achieving a fair distribution of income.
6. We have not received ORLA's final organizational documents relating to risk-sharing, but for purposes of this letter we assume that ORLA's planned operations would meet the requirements described in the Health Care Antitrust Statements.
7. Such primary provider organizations take many different forms, including very large professional corporations, Medical Foundation/MSOs, and Independent Practice Associations ("IPAs"). However, for simplicity, we will use here the term "IPA" as a representative example of a primary provider organization that serves HMOs under full risk capitation arrangements.
8. Mission Community Hospital has allowed a second anesthesia group to practice at the hospital for the limited purpose of serving the Monarch IPA there. (Monarch accounts for about 20 percent of Mission's patients.) A hospital representative told us that even this limited modification of the single group system has created serious difficulties for the hospital.
9. When on call duty, an anesthesiologist must be able to reach the hospital within 30 minutes of being called. Although, occasionally, an anesthesiologist has satisfied that requirement by staying at the hospital (or renting a room nearby) whenever he or she is on call duty, our interviews indicate that an alternative group is unlikely to undertake such an assignment unless most of the members who will be assigned to work at the Affected Hospital can reach it from their homes within the required 30-minute response time. In addition, if, as is likely, those members continue to work at the group's base-hospital as well as the Affected Hospital, they generally must be able to travel between the two hospitals within about the same 30-minute response time.
10. Officials at St. Joseph's Hospital, in North Orange County, emphasized the importance of surgeon familiarity with an anesthesiology group. They told us that if St. Joseph's existing Group demanded terms that were supracompetitive by a small but significant amount, they would not replace it with the comparable group at Long Beach Memorial Hospital -- a comparable hospital located in south L.A. County -- because the Long Beach Memorial group is "unknown" to St. Joseph's surgeons. In addition, an official of the firm that manages the Long Beach Memorial anesthesia group told us the group would welcome an opportunity to serve the principal IPA at St. Joseph's, but they have been told that their group is not a viable contestant for that work because it is unknown to the IPA's member-surgeons. Those officials told us they have concluded that to compete effectively for the anesthesia work at St. Joseph's, their group would have to merge or affiliate with an Orange County anesthesia group that is known and acceptable to St. Joseph's surgeons.
11. Since we have concluded that the relevant geographic market for managed anesthesia services of the kind provided by the ORLA Groups is no larger than Orange County, and since ORLA will include about 30 percent of all Orange County anesthesiologists, the Safety Zone for exclusive physician networks described in Statement 8 of the Health Care Antitrust Statements does not apply in this matter.
12. For example, because of travel time constraints, the competitive alternatives available to Hoag Memorial and Western Medical Center, which are located in north Orange County, may be limited to the comparable anesthesia groups that are based at the four comparable north Orange County hospitals (i.e., the Groups that currently serve Hoag, Western Med., St. Joseph's and St. Jude's). The joint venture would reduce those four market participants to one (ORLA). St. Jude's and St. Joseph's (and therefore CHOC, which shares St. Joseph's operating rooms) may also be able to turn to the Group at their sister hospital in south Orange County (Mission Community), despite the additional travel time, because of their greater ability to coordinate anesthesia scheduling at two sister hospitals. Nonetheless, for St. Jude's and St. Joseph's/CHOC, the joint venture would reduce the number of market participants from five (the Groups at St. Jude's, St. Joseph's/CHOC, Hoag, Western Med. and Mission) to one (ORLA).
13. That concern is supported by the experience of Western Medical Center, which decided, several years ago, to replace its existing Group ("IAMA") with Premier Anesthesia because IAMA was unwilling to agree to rates that were acceptable to one of Western Med.'s large managed care customers. Western Med.'s administrators were forced to back down and retain IAMA at significantly higher rates than Premier had offered because of strong opposition to Premier by the hospital's surgeons and medical staff.
14. ORLA points out that Sharp Health Care, which operates three San Diego hospitals, seriously considered replacing its existing anesthesia group ("ASMG") with Premier because of complaints by three of its largest IPA customers about ASMG's rates. The individual who supervised Sharp's search for a lower priced alternative confirmed that assertion. However, he told us that Sharp's first choice as an alternative to ASMG would have been the only comparable anesthesia group in San Diego County, but that group declined to bid because of the distance between Sharp's San Diego hospitals and that group's base hospitals in north San Diego County. (Sharp considered all other substantial-sized, financially integrated southern California anesthesia groups -- such as the anesthesia groups based at the major Orange County hospitals -- to be too far away.)
Moreover, that source pointed out, Sharp elected to retain ASMG, even though Premier's bid was lower than ASMG's by a small but significant amount, because Premier was not acceptable to many of Sharp's surgeons and other medical staff members. The three complaining IPAs acquiesced in that decision, and have absorbed the difference between ASMG's and Premier's rates.
15. In addition to eliminating each of the Affected Hospitals' ability to replace its existing Group with another Group that offers lower rates, ORLA's planned operations also will eliminate the prospect that a managed care customer that is confronted with supracompetitive anesthesia rates at one of the Affected Hospitals might redirect cases to one of the other Affected Hospitals if its Group offered lower rates.
16. Spokespersons for the Affected Hospitals and their IPA customers expressed strong views that a newly-formed group of anesthesiologists who have not previously provided anesthesia management services as a group, at a comparable hospital, would not be an acceptable substitute for any of the Groups.
It is conceivable that the employee anesthesiologists at UCI could form such a group, or that a comparable group located outside Orange County could become an acceptable substitute by merging with a smaller Orange County anesthesia group. However, under the current market conditions revealed by our investigation, it appears that such entry is unlikely to occur on a sufficient scale to offset ORLA's reduction in the number of existing competitors.
17. For example, South Coast Anesthesia Consultants ("SCAC"), which is comparable to each of the Groups, regularly enters into risk-sharing agreements with the IPAs it serves. In fact, SCAC is the only Orange County anesthesia group that does so. (Notwithstanding each Group's acknowledged ability to enter into such arrangements, each IPA we interviewed stressed that each of the five Groups has thus far successfully resisted its entreaties to enter into such arrangements. Indeed, some of those sources believe that the Groups formed ORLA primarily to bolster their resistance to meaningful, competitive risk-sharing arrangements.)
18. Each of the health plan officials we interviewed told us that they would not enter into capitation agreements directly with ORLA for anesthesia services, and "carve-out" those services from their existing capitation agreements with IPAs. They prefer the beneficial service capitated IPAs provide as "general contractors" that arrange for and supervise the services of numerous types of specialists and subspecialists, including anesthesiologists. (That response by the health plans was consistent with the statement of one Group's representative that ORLA's organizers had been told by a consultant that area HMOs "would have no interest" in entering into such "carve-out" capitation agreements with ORLA.) Further, each of the health plan officials stated that in circumstances in which it is beneficial for a health plan to contract directly with anesthesia suppliers, ORLA's intention to negotiate a single agreement for all five Groups would provide only small transactions cost savings, which, they asserted, would be far outweighed by the Groups' ability to impose supracompetitive terms by bargaining collectively through ORLA.
19. For example, administrators of the Affected Hospitals pointed out that practice protocols would have to be approved by each hospital's medical staff, and they suggested that it would be far less cumbersome (and more likely to be successful) for each Group to propose such protocols to its own hospital's medical staff than for ORLA to attempt to obtain a consensus regarding a single set of practice protocols among all five Groups and all of the Affected Hospitals' medical staffs.