INTRODUCTION

     The economics of health-care delivery are causing an increased consolidation via proliferation of large integrated delivery systems. [FN1] As this consolidation continues, two trends are emerging in health law that apparently will yield a form of greater liability and accountability for organized providers of health care, especially managed care organizations (MCOs).[FN2]
     The first trend arises under the Employee Retirement Income Security Act (ERISA). [FN3] There is an increasing erosion of the ERISA preemption under various theories of ostensible agency and redefinition of language contained in the ERISA statute. [FN4]
     The second trend is the expansion of the doctrine of informed consent from a common-law sense of permission to touch toward a doctrine of knowledgeable choice on the part of patients.[FN5]

I. MANAGED CARE AND THE PHYSICIAN-PATIENT RELATIONSHIP

     The health care delivery system has been undergoing historic change. Multiple new kinds of delivery systems and reimbursement systems have been developed by private sector initiatives and governmental programs. [FN6] A physician and patient previously had an independent fiduciary relationship. Now it is more likely than not that a person receiving medical benefits is in some way dependent upon receiving health care through a panel of physicians selected by a third-party intermediary. [FN7] The third-party intermediary, in the form of a managed care organization, either itself, or on behalf of the employer purchasing benefits for its employees, is likely to have control over physician access to patients as well as the ability to cut off the supply of patients to a physician. [FN8] Managed care health plans involve networks created by the managed care organization through selection and retention of physicians. The focus of this selection or retention process tends to be on design, physician payment, and utilization management and quality improvement. [FN9] Health plans develop a network of the "best physicians," however defined, and limit patient access to "nonnetwork physicians." This is seen as one of the most important aspects of a managed care health plan. [FN10] MCOs tend to develop networks either by recruiting many physicians to the network first and then reducing the total number pursuant to cost criteria developed by the plan, or by carefully selecting physicians at the outset. The latter approach represents the more common of the two strategies. [FN11] Although MCOs maintain criteria for credentialing of plan physicians, this information is not widely available. More commonly known are the specific bases for exclusion, such as use of malpractice history, review of licensure status, and more qualitative information, such as practice style. In this regard, some MCOs may undertake site visits to physicians' offices, while others may try to examine the "actual experience" of physicians. [FN12] These data include, as elements reviewed or verified, the following: (1) credentials and affiliations; (2) National Practitioner Data Bank files; (3) licensure; (4) substance abuse problems; (5) physician office visits; (6) medical records review; (7) physician office facilities; (8) physician practice style; (9) indemnity claims data; (10) state hospital discharge data; and (11) other databases. [FN13] In addition, a majority of MCOs require board certification or board eligibility. [FN14]
     An important development is the orientation of physicians in health plans to the corporate culture of the managed care organization. "At some plans, this process goes well beyond explaining plan administrative procedures to considering approaches to care delivery." [FN15] Although there have been some large scale "deselections" reported, ultimately the extent of, and reasons for, turnover in health plans remains largely obscure. [FN16] This fact, however, leaves open the possibility for abuse by MCOs and, importantly, a great deal of control over incentives and other ways of influencing the decision-making of plan physicians.
     In addition to the credentialing process, MCOs control payments to plan physicians by a variety of means. In fact, the majority of MCOs report that they use payment structures to influence and control physician behavior. A mix of fee for service, capitation, bonuses, and withholds are used. Capitation involves payment of a flat fee per patient, per time period, with the physician then assuming the risk for the cost of all services. A withhold or bonus involves an arrangement where part of the payment to a physician is kept back by the MCO until the end of the year. If the allocation of resources for patient care is lower than anticipated, then physicians in the plan share the money withheld or receive a bonus. If utilization is greater than anticipated, then physicians in the plan may share in the loss. [FN17]

II. LIABILITY ISSUES IN THE MANAGED CARE ENVIRONMENT

     Many managed care arrangements are structured in ways that involve some form of oversight of a physician's conduct or an incentive plan to make the physician conscious of the costs of treatment. MCOs may undertake utilization review of a physician's care in a concurrent or prospective manner or may implement a payment scheme that puts physicians at some form of financial risk for the care delivered. With the actual delivery of medical care still in the hands of individual physicians, but under review by third parties, many of the new managed care methodologies blur the line between providing traditional insurance coverage of payment for medical services and influencing the actual delivery of medical care. Because clinical decision-making is now shared between physicians and MCOs or influenced by contractual obligations or financial incentives, it has been necessary for courts to view malpractice claims in a very different context. A managed care or utilization review organization thus may share liability with a treating physician if the utilization review is administered in a negligent manner. [FN18] It is recognized, however, that a physician must protest and appeal utilization review decisions that are contrary to medical judgment regardless of the utilization reviewer's role. [FN19] Nevertheless, a physician and the utilization reviewer both may be liable for harm resulting from a negligently administered denial of benefits to an insured patient, even if a physician ultimately bears responsibility for medical judgments. [FN20]
     Courts have begun to recognize that managed care arrangements create issues of vicarious liability and ostensible agency. Many courts are beginning to look carefully at the agency relationships between physicians and MCOs. Thus, courts have been willing to conclude that physicians are agents of MCOs for malpractice purposes when patients have no choice in selecting physicians except as provided through the MCO. [FN21] But the courts have been reluctant to extend this principle where the organizational structure may be complicated. In Chase v. Independent Practice Association, [FN22] malpractice was committed by a physician employed by a group practice that contracted with an Independent Practice Association to deliver care to the patients of an MCO. Because of the numerous intermediaries, the agency link between the physicians and the MCO was seen as too tenuous to support a claim of vicarious liability.

A. The Role of ERISA in Changing the Nature of Liability

     Most recipients of nongovernmental health care benefits receive their health care in some way through plans that are under the jurisdiction of ERISA. [FN23] These plans take the form of insurance policies that are purchased by the employer for its employees or self-insured plans, where the employer sets up a form of trust that pays for employee health benefits pursuant to a third-party, contractually administered, health plan, with a form of stop loss insurance as back up. [FN24] Such federally regulated health plans account for a majority of all private health-care funding. [FN25] Between 1987 and 1994, the proportion of workers enrolled in HMOs and managed care generally more than doubled. [FN26] Thus, the significance of regulation through ERISA has far-reaching consequences, and the interpretation of ERISA by courts becomes all the more important in determining how medical care is delivered in the United States and under what system of checks and balances. Much of the reorganization of health delivery from a cottage industry into large integrated groups or managed care groups has been made possible by the ERISA law. Congress passed ERISA in an effort to prevent abuse and mismanagement in the private pension plan system. [FN27]
     ERISA establishes uniform federal requirements for employee welfare benefit plans, including health plans. ERISA preempts the states from regulating such plans while confirming the states' powers to regulate the business of insurance. The federal courts have exclusive jurisdiction over most ERISA violations, except for violations that are subject to concurrent jurisdiction between federal and state courts, as follows: Except for actions under subsection (a)(1)(B) of this section, the district courts of the United States shall have exclusive jurisdiction of civil actions under this subchapter brought by the Secretary or by a participant, beneficiary, or fiduciary. State courts of competent jurisdiction and district courts of the United States shall have concurrent jurisdiction of actions under subsection (a)(1)(B) of this section. [FN28]
     All other ERISA claims are within the exclusive jurisdiction of federal courts. [FN29] Actions brought under section 1132(a)(1)(B) are personal in nature, unlike other ERISA claims that allege violations of a particular provision of ERISA or an ERISA plan. Plaintiffs suing under section 1132(a)(1)(B) seek: (1) to declare the plaintiff beneficiary's rights under the plan; (2) to recover benefits personally due to the plaintiff beneficiary; or (3) to enforce personal rights to a plan. [FN30]
     ERISA is largely deemed to be a pension law and thus is subject to equitable interpretation under the law of trusts. As such, the main purpose of ERISA is to preserve the funds within the trust. [FN31] ERISA does not mandate benefits, but rather regulates the provision of benefits administered by private employers in commerce and labor unions. [FN32] Furthermore, ERISA provides specific remedies against those who are fiduciaries under ERISA, but does not provide remedies against those who are not fiduciaries. Hence, it is essential for courts adjudicating claims to determine whether they fall within the jurisdiction of ERISA. [FN33]
     ERISA is the governing law when there is a conflict between a state law and ERISA. The law specifically "saves" state insurance regulation but exempts employee benefit plans from regulation as insurance. [FN34] Thus, in Shaw v. Delta Airlines, Inc., [FN35] a state law regulating civil rights of employees was held preempted by ERISA. In Metropolitan Life Insurance Co. v. Massachusetts, [FN36] an insurance contract purchased for employees for health benefits, as opposed to the employee benefit plan itself, was seen as being within the regulatory orbit of state government as a reflection of traditional insurance regulation. Several important case decisions have begun to set the way in which ERISA affects health-care liability. Essentially, ERISA has been interpreted by courts to preempt state-based tort liability for tortious administration of employee benefits under certain circumstances. In Pilot Life Insurance Co. v. Dedeaux, [FN37] for example, the Supreme Court held that a common-law cause of action arising from "improper processing of a claim for benefits" was preempted. [FN38] This interpretation was followed and extended to claims involving medical benefits determinations in the case of Corcoran v. United Healthcare, Inc. [FN39] There, the court applied ERISA to preempt a state- based claim of tort and limit a mother's recovery for the death of her newborn to the costs of the care that had been recommended by her physicians and denied by United Healthcare (acting as utilization reviewer for a Blue Cross plan).
     The Corcoran case set a standard of sorts for several years, with Kuhl v. Lincoln National Health Plan [FN40] as the most extreme reach of this reasoning. In Kuhl, the plaintiff was denied bypass surgery as medically unnecessary. The claim was adjudicated, and the plaintiff won benefits but had progressed to the point of needing a transplant. The plaintiff died while adjudicating that issue. The plan was found liable only for costs of care denied. While state-based tort suits for medical malpractice against individual physicians are not preempted by ERISA, the law came to be recognized as a basis for snuffing out state-based tort claims against administrators of health plans and delivery systems that were based on wrongful denial of benefits or wrongful interference with benefits. Once a cause of action is based upon a denial of benefits for a health plan beneficiary, it is adjudicated under ERISA. [FN41] If the claim for denial of benefits rests upon a state law that "relates" to or is "connected" to an employee benefit plan, then the state claim is preempted and only the ERISA claims and remedies are available, [FN42] unless the law is "saved" from preemption as insurance. [FN43]

B. Retreat from Extreme Preemption

     While the ERISA preemption became one of the most sweeping in federal law, there has been a considerable refinement of thinking in this area. In New York State Conference of Blue Cross and Blue Shield Plans v. Travelers Insurance Co., [FN44] the Supreme Court upheld a New York State rate scheme that imposed a tax on hospital usage for the purposes of funding a Blue Cross rate program. This decision was based upon the fact that the taxes existed independently of the employee benefit plans and were not targeted at the employee benefit plans. Indeed, to read the preemption provision as displacing all state laws affecting costs and charges on the theory that they indirectly relate to ERISA plans that purchase insurance policies ... would effectively read the limiting language in § 514(a) out of the statute, a conclusion that would violate basic principles of statutory interpretation ... "[p]reemtion does not occur ... if the state law has only a tenuous remote, or peripheral connection with covered plans ...." [FN45] The Travelers court especially focused on the underlying purpose of ERISA to provide for uniform benefit administration and differentiated direct versus indirect regulation of employee benfit plans:
     [A]n indirect economic influence does not bind plan administrators to any particular choice and thus function as a regulation of an ERISA plan itself.... Nor does the indirect influence of the surcharges preclude uniform administrative practice or the provision of a uniform interstate benefit package if a plan wishes to provide one.... It is an influence that can affect a plan's shopping decisions, but it does not affect the fact that any plan will shop for the best deal it can get. [FN46] This redefinition of the term "relates" as used in section 514 of ERISA has now been given added weight. In the case of California Division of Labor Standards Enforcement v. Dillingham, [FN47] the Supreme Court unanimously ruled that an apprentice wage scheme in California was seen as not relating specifically to employee benefit plans, even though the overwhelming majority of such plans are employee benefit plans, and not subject to the jurisdiction of ERISA. Thus, the Dillingham court further emphasized the difference between state laws having an economic impact on an ERISA plan and state laws that determine the structure and choices of a plan. We could not hold preempted a state law in an area of traditional state regulation based on so tenuous a relation without doing grave violence to our presumption that Congress intended nothing of the sort. We thus conclude that California prevailing wage laws and apprenticeship standards do not have a "connection with" and therefore do not "relate to" ERISA plans. [FN48]
     Thus, the ERISA preemption covers laws that single out ERISA plans and dictate benefits. In a concurring opinion by Justices Scalia and Ginsberg, the proposition was advanced that the broad sweep of ERISA should be curtailed in favor of a more traditional form of conflict preclusion as articulated in prior cases such as Silkwood v. Kerr-McGee Corp., [FN49] which would narrowly "preclude" only those state laws that quite specifically were targeted at areas specifically reserved to the federal government by federal statutes. I think it would greatly assist our function of clarifying the law if we simply acknowledged that our first take on this statute was wrong; that the "relate to" clause of the pre-emption provision is meant, not to set forth a test for pre-emption, but rather to identify the field in which ordinary field pre-emption applies .... [FN50] The Court has recently underscored the fact that state laws of general application are not preempted by ERISA. In De Buono v. NYSA-ILA Medical & Clinical Services Fund, the Court specifically rejected a defense that ERISA preempted laws that would have the consequence of depleting fund assets. A consideration of the actual operation of the state statute leads us to the conclusion that the HFA is one of "myriad state laws" of general applicability that impose some burdens on the administration of ERISA plans but nevertheless do not "relate to" them within the meaning of the governing statute. [FN51] Similarly, an intermediate state court in Pennsylvania has ruled that, because Travelers stands for the proposition that any indirect economic regulation of an ERISA plan is not prempted, and torts are an indirect economic regulation, torts generally are not preempted. [FN52] This case has been argued before the Pennsylvania Supreme Court without a ruling as of this writing. A favorable ruling would push persuasive authority on this issue to its most extreme view of nonpreemption, and the Travelers line of cases might harbinger a similar ruling now by the United States Supreme Court if it were to grant certiorari. Along these lines, in 1997, the state of Texas enacted a change in the law to subject all utilization review to an ordinary standard of care in a neutral way that would seem to be the kind of law of general application contemplated by this line of cases. [FN53] Additionally, the scope of claims covered by the jurisdiction of ERISA has been under review in lower federal courts. In many of the newer managed care schemes, the division between providing care and arranging payment for the provision of care has been blurred and has forced courts to determine what is under review in a given case. [FN54] Claims that a physician may be seen as an ostensible agent of an entity where the benefit is membership in a plan but the medical service is seen as separate from the benefit have served as a basis for allowing state-based torts claims to proceed in state court rather than be subject to ERISA preemption. Thus, vicarious liability against managed care plans, triggered by the conduct of treating physicians who are seen as agents of the plan, have been remanded to state courts as not being preempted in the third, seventh, and tenth circuits. [FN55] The third circuit has introduced a more subtle distinction over interpreting the meaning of section 502 of ERISA in Dukes v. U.S. Health Care. [FN56]
     Looking at the literal meaning of denial of benefits, the court reasoned that, in a medical malpractice case, there is not necessarily a claim for denial of benefits, or clarification of a right as covered by section 502(a)(1)(B). Rather, there is a question regarding whether the benefits delivered are somehow inadequate. [FN57] Looking at this distinction between quality of benefits and quantity of benefits, the court reasoned that a vicarious or corporate liability claim made directly against the managed care plan would not be subject to the jurisdiction of ERISA, and therefore such a claim would not be preempted. [FN58] In this holding, the third circuit has addressed the worry of the fifth circuit in Corcoran that there might be a wrong without a remedy when benefits determinations are intermingled with medical care. Under the Dukes reasoning, an incentive scheme that causes a physician to compromise the delivery of care would be a source of negligence on the basis, not that the plan has a negligent administrative scheme, but that the plan's agent has delivered the benefit with inadequate quality. Therefore, the plaintiffs were asserting a claim that was not preempted by federal occupation of the field and removal to federal court was improper. Ironically, the changing interpretation of ERISA has now led one managed care plan to assert that it had denied required benefits specifically in an attempt to circumvent the reasoning in Dukes. [FN59] Although this defense ultimately was deemed to focus on the quality, not quantity, of benefits, for purposes of ruling on removal, it demonstrates the fluidity of thinking on this issue. There are certainly some important risks in raising this defense, as the defense raises the notion that plan administrators have violated their fiduciary duties. [FN60] Additionally, the difference between "conflict preemption" of section 514 and "claim preemption" under section 502 demonstrates the care with which claims must be pleaded or attacked. The seventh circuit allowed the vicarious liability claim in Rice v. Panchal [FN61] to be remanded as not preempted, as it was deemed outside the scope of section 502, but it also ruled that claims for denial of benefits really were ERISA claims in disguise in Jass v. Prudential Health Care Plan, Inc. [FN62] The United States District Court for the District of Maryland similarly has divided cases along the same lines. [FN63] Separately, a post-Travelers attempt to assert that mere employment by a health maintenance organization, related to an employee benefit plan, was deemed too tenuous and the case was remanded. [FN64]

III. INFORMED CONSENT

     Informed consent as a doctrine also has been evolving. [FN65] Originally, informed consent was very simple. A physician had to obtain permission from a patient to treat. If the physician failed to obtain such permission, then any treatment would be considered a battery. [FN66] Schloendorff v. Society of New York Hospital [FN67] is an early case in which the court laid out several important doctrines, one of which is an outline of early informed consent thinking. The patient was seen as autonomous: "Every human being of adult years and sound mind has a right to determine what shall be done with his own body." [FN68] But to violate the standard articulated in Schloendorff, a patient actually would have to object, and the physician would have to proceed in spite of this objection, for a battery to occur. In practice, this is an extremely rare occurrence and most states treat inadequate informed consent as negligence. [FN69] Generally, informing a patient in a manner consistent with professional standards was considered adequate. If other physicians did not disclose, or disclosed in a manner similar to the physician defendant, then there was no breach of duty. [FN70] But in the case of Canterbury v. Spence [FN71] a new standard was articulated. The standard was based on the patient's informational needs, not the duty of disclosure established by the profession. "A risk is ... material when a reasonable person, in what the physician knows or should know to be the patient's position, would be likely to attach significance to the risk or cluster of risks in deciding whether or not to undergo the proposed therapy." [FN72]
     In Moore v. Regents of the University of California, [FN73] the informed consent standard again was advanced to include a need for disclosure of conflicts of interest. In this case, a physician harvested spleens from leukemia patients without revealing a research project that utilized those spleens. He went on to patent a cell line from one of the patients' spleens. The issue was not competence of care. Instead, the issue was the physician's use of the plaintiff for the physician's monetary and pecuniary benefit under the guise of helping the plaintiff. Even if the splenectomy had a therapeutic purpose, it does not follow that [the physician] had no duty to disclose his additional research and economic interests. As we have already discussed, the existence of a motivation for a medical procedure unrelated to the patient's health is a potential conflict of interest and a fact material to the patient's decision. [FN74] The lack of disclosure triggers liability because the patient cannot trust what the physician says if the patient is unaware of the physician's hidden pecuniary interest. The court emphasized "that a physician who is seeking a patient's consent for a medical procedure must, in order to satisfy his fiduciary duty and to obtain the patient's informed consent, disclose personal interests unrelated to the patient's health, whether research or economic, that may affect his medical judgment." [FN75] The court, therefore ruled that the plaintiff "has stated a cause of action for breach of fiduciary duty, or lack of informed consent, based upon the disclosures accompanying that medical procedure." [FN76] It is reasonable to see that there is an absolute duty to disclose conflicts of interest or any consent obtained is not truly informed. Because a patient in a fee-for-service setting has an understanding of the conflicts, but may not understand some of the new payment schemes, there are some important implications for physicians in the managed care context. But how does one separate the duty not to benefit from the reality of being a physician today, where everything said to a patient is colored by what the physician can afford to do and what the patient can afford to have done? At what point is there a specific duty and when is it breached? Is it a line drawn at harvesting body parts or is it not telling a patient that, as a managed care physician paid by capitation or under a withhold, the physician may stand to gain from not recommending a specialist or a procedure? Is that worse than a fee-for-service arrangement where physicians gain by recommending more treatment? Under the Moore and Canterbury approaches, it would seem that a physician would be liable for failing to obtain an adequate informed consent if the physician failed to explain to a patient that the physician's financial compensation may be based on incentives to deny or restrict care. The financial gain from limiting patient options in a setting where the average patient may be ignorant of modern health payment complexities certainly would fit the Moore model.
     The courts in the Canterbury and Moore cases seem to be moving in a direction of personal autonomy as almost a civil right, while erosion of the ERISA preemption through various reinterpretations of personal state-based rights appears to signal a sensitivity on the part of the courts to preserve personal rights in the face of a legislative scheme designed to preserve a more collective right in the form of trust fund integrity. Given the trend toward allowing a vicarious liability suit against the parent managed care organization, even in the context of an ERISA benefit plan, these lines of cases combined would seem to herald a floodgate of new tort litigation based upon probable widespread nondisclosure or inadequate disclosure of the financial incentives for physicians in many managed care plans.

IV. THE PHYSICIAN AS A FIDUCIARY

     There is an emerging twist on this notion of nondisclosure. A misrepresentation regarding the nature or operation of a health plan that might otherwise create a tort is preempted under ERISA as it relates to the plan. So while there is a trend emerging in tort law toward greater disclosure to patients, this may be viewed as administrative in nature and thus, under ERISA, the tort effectively may be snuffed out. The United States Supreme Court, in Varity Corp. v. Howe, [FN77] has determined that disclosure is an absolute fiduciary duty owed to the beneficiaries of an ERISA employee benefit plan. Under ERISA, a fiduciary is determined by the extent to which the individual in question (i) exercises any discretionary authority or discretionary control respecting management of such plan or exercises any authority or control respecting management or disposition of its assets, (ii) he renders investment advice for a fee or other compensation, direct or indirect, with respect to any moneys or other property of such plan, or has any authority or responsibility to do so, or (iii) he has any discretionary authority or discretionary responsibility on the administration of such plan. [FN78] Thus, the eighth circuit, in Shea v. Esensten, [FN79] decided that this preemption of the state-based tort nonetheless opens the health plan to claims under sections 1002(21) and 1140(a)(1) of ERISA as a fiduciary violating a duty to the beneficiary if the undisclosed details are material. In Shea, a patient with chest pain saw his primary care physician but was denied a referral to a specialist. The physician told the patient that he was too young to have heart disease. Even though the patient offered to pay out of pocket for the specialist care, no referral was made. The patient died a few months later from heart failure. At no time was the patient told by the physician or managed care plan that there was a capitation funding scheme in place. A lawsuit was filed by the deceased patient's estate in state court, alleging fraudulent nondisclosure by the managed care plan (Medica) about the terms of its physician incentive plan. Thus, the case started in state court on the basis of an allegation of an undisclosed benefit limitation. When characterized as a claim for wrongful denial of benefits, the lawsuit became a routine version of an intermingling of benefits determination and medical decision-making similar to that in Corcoran, and thus subject to removal to federal court and dismissal for failure to state a claim under ERISA. But Shea amended the pleading and alleged breach of fiduciary duty by Medica. The court responded by ruling that the information not disclosed was material. Further, the court ruled that there was a nondelegable duty by Medica to disclose the nature of the capitation arrangement with physicians to plan members. Withholding such information constituted breach of a fiduciary duty. The court relied on the Supreme Court's analysis in Varity that ERISA requires discharge of duties solely in the interest of beneficiaries under section 1104, and relied on the history that Congress structured ERISA so courts would apply the common law of trusts. Thus, in adopting the analytic position of the common law of trusts, the court determined that, when administering a plan for beneficiaries, the plan must tell those beneficiaries how benefit decisions are to be made and that this is the core of the fiduciary duty of loyalty. A physician incentive plan, therefore, is a material fact for purposes of disclosure. The financial arrangement goes right to the heart of how a physician will deliver care, particularly in a situation where the physician is encouraged to withhold care. Because this was a material fact, the patient's estate had stated a valid claim under ERISA, and because there existed a breach of trust, the plaintiff was eligible for damages. Thus, Shea stands for the proposition that, under ERISA, the mere presence of an incentive plan is not a breach, but that the quality, or sufficiency, of disclosure is a measure of fiduciary duty. "Any other result would reward Medica for giving its preferred physicians an incentive to make more money by delivering cheaper care to the detriment of patients ... and ERISA should not be construed to permit the fiduciary to circumvent [its] ERISA-imposed fiduciary duty in this manner." [FN80]
     Significantly, the court left unaddressed the consequences of language in the ERISA statute that included "discretionary authority ... on the administration of such plan." [FN81] The scope of such discretion would seem to open the door to finding that what might otherwise be negligent utilization review in tort law could now be a breach of fiduciary duty in trust law. The "discretionary authority" language is broad enough to raise interesting questions of whether or not treating physicians are fiduciaries in complex structured plans where the medical decision-making is deemed not to be preempted, but rather an administrative breach of fiduciary duty. This issue has not yet been adjudicated. [FN82] In addition, the mandatory disclosure requirements accrue to the administrator of the plan, whoever is so designated. [FN83] This may be problematic for managed care organizations that operate through proprietary protocols. Must proprietary clinical guidelines and treatment protocols, as well as physician selection criteria, now be disclosed as material facts pursuant to fiduciary duties?

V. ANALYSIS

     Given the mix of decisions on ERISA as outlined in this article, one can see that there is a trend to hold utilization reviewers and MCOs accountable for the consequences of their incentive plans. This is a complicated endeavor in view of the sweeping nature of the governing federal law and its rather vague language. It is not always clear who actually holds the fiduciary duty. If the plan is a self-insured plan, then the employer is a fiduciary under Varity. If the plan is an insured plan, then the MCO, as the utilization reviewer, would seem to be the fiduciary under Shea. In multilayered arrangements, one may analogize to Chase and find the fiduciary duty further down the line. Presumably, if an MCO's role in an employee benefit plan is to function only as an administrative services organization, then the holder of the plan assets at risk, not the MCO, may bear the duty. [FN84] Some of this is determined by ERISA itself, through contractual designations. [FN85] What happens when the new arrangements of health care delivery leave capitated physicians' networks holding the assets at risk? This question is not yet answered by the courts but should be of concern to physicians as they organize networks through capitation. So far, we see an increasing trend toward nonpreemption of state-based tort claims under ERISA. We also see a trend toward allowing ostensible agency and vicarious liability actions to proceed against parent MCOs. There is also potential direct corporate liability for MCOs. What happens when these trends converge? There may be some paradoxical results. The underlying theme of all of these trends seems to be the creation of greater tort rights for patients and more widespread liability. It seems more likely than not that the defenses proferred by managed care organizations actually will be strengthened by much of what has happened and that physicians in their integrated networks will incur increased liability.
     As discussed above, there is a movement toward increasing consolidation of practices into provider networks that, in turn, contract their services to other organizations. Chase stands for the proposition that increased complexity breaks the chain of causation. The more an arrangement shifts decision-making to the physician and the more the managed care organization takes on the character of simply arranging for payment for services, the more the changing notions of ERISA preemption will not matter.
     Thus, liability, whether under tort law or under an ERISA-based fiduciary duty, will continue to remain primarily with the treating physician or the treating physician's practice. We may find a temptation, because of the ERISA preemption of state-based torts, for physicians to seek a contractual remedy for their malpractice liability. A network of physicians, organized as a Preferred Provider Organization, could seek to negotiate a contract for medical services with a self-insured employer. The contract could contain provisions to determine the scope of utilization review, decisional liability of parties, limits on recovery and the like: If the physician/patient contract is not specific as to the physician's duties, then the norms and standards of the profession as a whole should be read into the agreement by implication. But if the parties can fairly be said to have chosen a different regime of responsibilities and rights, courts might then forswear their customary regulatory stance and allow the parties' choice to control. [FN86] Because this would be a contract for benefits covered by ERISA, so long as it was in keeping with the provisions of ERISA, it would seem to preclude tort liability, including liability based on fraud and misrepresentation of benefits and inadequate delivery of care by the treating physicians. One commentator admits that it is impossible to write a contract that could account for all medical contingencies and these currently still rely on professional norms. [FN87] Further, courts traditionally construe such contracts for health services as contracts of adhesion. [FN88] Furthermore, these contracts tend to be negotiated by employers with second parties to deliver benefits to the plan beneficiaries without the input of the beneficiaries. As such, these would be third-party benefits contracts that abrogate beneficiaries' legal rights without their consent, and thus subject to being declared unconscionable. [FN89] ERISA lays out its purpose as dictating a duty to benefit plan beneficiaries. Contracts that limit beneficiaries' rights for the purpose of saving money for the employer who has established the plan would seem to conflict with the purposes of ERISA. [FN90] It is especially problematic to negotiate contingencies given the high error rate reported for medical care [FN91] and the current infancy of quality measures and the varying roles and responsibilities of those trying to craft meaningful practice guidelines for the delivery of medical care. [FN92] Because of Shea, we see a new twist that would bring back analogous responsibilities by the parties under the fiduciary duties imposed by ERISA. Thus, we move from a tort-based liability to a trust-based liability that becomes effectively the same at the end. ERISA now can seemingly be interpreted not to preempt certain liability that accrues through ostensible agency to those who may administer the benefits, as seen in the Dukes line of cases. On the other hand, where more traditional tort liability is seen as preempted regardless of agency relationships, the liability now accrues to those who are fiduciaries as defined in the body of ERISA and as applied in Shea and Varity. [FN93] The implication here is that the individual who holds the actuarial risk, under either interpretation, may bear the ultimate liability. As providers are increasingly organizing themselves into networks and other forms of integrated, risk-sharing arrangements, not only is there an increased need for disclosure, but the more actuarial risk they take, and the more they take over the final decision-making discretion, the more the liability risk accrues to them under either interpretation. [FN94] Such disclosure is especially important because the degree of consumer unhappiness with managed care has begun to show signs of meaningful patient backlash, as seen in such measures as the number of complaints received by the California Department of Corporations. [FN95] It may, however, take a while before employers understand the degree of liability they have under either theory, as there seems to be a focus on cost and only a less meaningful focus on true quality. This was demonstrated in a 1996 survey of 351 large employers about their health plans, which revealed an overwhelming preoccupation with cost over quality and concern over government regulations. [FN96] The employers' apparent focus on cost alone may not be justified any longer on the basis of squeezing out unnecessary and wasteful care on behalf of the beneficiaries of their health plans as they contract with MCOs. This is because the market has now changed so dramatically. It is ironic that employers are taking an activist and pennypinching stance when annual increases are already far more reasonable than in the past. In any case they are initiating the demand for lower prices in many cases. In other cases, insurers and HMOs are making the demand, in order to increase their market share, augment their profits, or both. [FN97] There may be a dawning implication that there is some increased concern that ERISA makes the employers liable as plan fiduciaries, especially after the Varity ruling. [FN98]
     For the foreseeable future, employers who have established self-insured health benefit plans may incur a tremendous liability risk. It is not likely that such risk will go unnoticed for long, however. Because employers have discovered new and increasing bargaining power in this area, [FN99] the ability of employers to bargain and shape contracts may allow them to force upon the emerging physician networks both actuarial risk and fiduciary as well as traditional malpractice liabilities. This situation would mean that, by contracting to limit liability rather than by contracting to improve quality of benefits for plan beneficiaries, physician networks would turn some predictions upside-down by leaving physicians with increased exposure rather than limiting their exposure. [FN100] Ultimately, it is possible to designate by contract who will serve as a plan administrator and with what discretionary authority. [FN101] However, with this discretion and authority comes fiduciary responsibility and liability, and so it may be possible that, by shifting increased discretion to networks of physicians while simultaneously shifting actuarial risk, physicians may take on a new form of liability in addition to that of traditional tort law. While some of the contractual penalties incurred might be more tolerable to physicians than some tort liabilities, the possibility of removal as fiduciaries in a competitive market may mean death to a physician network.

CONCLUSION

     The impact of the changing jurisprudence governing ERISA will have an ever-increasing impact on the final shape of the health delivery system in the United States. Increasingly, new and different liabilities should be fostering certain directions for these changes. Those most affected by this jurisprudence, especially employers and physicians, as yet seem unaware of it and its paradoxes. This poses both an opportunity and a challenge for health attorneys.

[FNa1]. Practicing psychiatrist; immediate past president and current board of trustees chair of the Medical Society of the District of Columbia; former vice chair of the American Medical Association Section Council on Psychiatry. Address correspondence to Dr. Blackmon at 4501 Connecticut Ave., N.W., Suite 104, Washington, D.C. 20008.

[FN1]. K.P.M.G. PEAT MARWICK LLP, INTEGRATED PATIENT CARE: MANAGING HEALTH CARE COSTS (1996).

[FN2]. See generally Kinnard, Judicial Refusal to Apply ERISA preemption to Tort Actions Against Health Care Providers, THE BRIEF, Winter 1997, at 24.

[FN3]. 29 U.S.C. § 1001.

[FN4]. Id.

[FN5]. See generally J. MALCOLM, TREATMENT CHOICES AND INFORMED CONSENT (1988).

[FN6]. See generally PHYSICIAN PAYMENT REVIEW COMMISSION, 1995 ANNUAL REPORT TO CONGRESS (1995). This volume contains the results of several surveys specifically commissioned for the P.P.R.C. See also PHYSICIAN PAYMENT REVIEW COMMISSION, 1997 ANNUAL REPORT TO CONGRESS (1997) (for a less comprehensive follow up of some of these issues).

[FN7]. K.P.M.G., supra note 1, at 31-53.

[FN8]. ANNUAL REPORT (1995), supra note 6, at ch. 10.

[FN9]. Id. at 220.

[FN10]. Id. at 221.

[FN11]. Id.

[FN12]. Id. at 222.

[FN13]. Id. at 223.

[FN14]. Id. at 224.

[FN15]. Id.

[FN16]. Id. at 226.

[FN17]. Id. at 229.

[FN18]. Wickline v. State of California, 239 Cal. Rptr. 810 (Cal. App. 1986) (utilization review scheme was negligently administered to deny benefits for a full hospital stay, leading to the premature discharge of a patient with resulting harm to the patient).

[FN19]. Id. at 820.

[FN20]. Wilson v Blue Cross of Southern California, 271 Cal. Rptr. 876 (Cal. App. 1990); Hughes v. Blue Cross of Northern California, 263 Cal. Rptr. 850 (Cal. App. 1989).

[FN21]. See, e.g., Boyd v. Albert Einstein Medical Center, 547 A.2d 1229 (Pa. Super. 1988).

[FN22]. 583 N.E.2d 251 (Mass. App. 1991).

[FN23]. UNITED STATES GOVERNMENT ACCOUNTING OFFICE, HEALTH INSURANCE: HOW HEALTH CARE REFORM MAY AFFECT STATE REGULATION (Nov. 5, 1993). The exact number of workers covered by ERISA is actually a matter of some dispute. The G.A.O. has not issued an updated estimate since 1993, and the United States Department of Labor has never officially published its own estimate and has used a different, but comparable, figure in a press release. See UNITED STATES DEPARTMENT OF LABOR, LABOR DEPARTMENT ANNOUNCES INITIATIVES TO IMPLEMENT PRESIDENT CLINTON'S CONSUMER BILL OF RIGHTS DIRECTIVE, USDL No. 98-69 (Feb. 20, 1998).

[FN24]. ANNUAL REPORT (1995), supra note 6, at 249-52.

[FN25]. HEALTH INSURANCE, supra note 23.

[FN26]. ANNUAL REPORT (1995), supra note 6, at 186, fig. 4-4.

[FN27]. For a discussion of ERISA in general, see B. COLEMAN, PRIMER ON ERISA (3d ed. 1989); J. LANGBEIN & B. WOLK, PENSION AND EMPLOYEE BENEFIT LAW (1990). For a discussion of the legislative history of ERISA, see Gordon, Managed Care, ERISA Preemption, and Health Reform-The Current Outlook, BNA HEALTH CARE POLICY AFFAIRS (Apr. 17, 1995).

[FN28]. 29 U.S.C. § 1132(e)(1).

[FN29]. See Livolsi v. Ram Constr. Co., 728 F.2d 600, 602 (3d Cir. 1984). According to ERISA's legislative history, Congress granted the federal courts exclusive jurisdiction because of the interstate character and nature of most employee benefit plans and Congress believed there was an essential need to provide for uniformity in administration of these plans. H.R. REP. NO. 533, 93d Cong., 2d Sess. 17 (1973), reprinted in 1974 U.S.C.C.A.N. 4639, 4655. See also Gordon, supra note 27.

[FN30]. ERISA section 1132(a)(1)(B) provides that "[a] civil action may be brought by a participant or beneficiary ... to recover benefits due to him under the terms of the plan, or to clarify his rights to future benefits under the terms of the plan." According to ERISA's legislative history, suits brought under section 1132(a)(1)(B) may not involve a specific title I provision of ERISA. See H.R. REP. NO. 1280, 93d Cong., 2d Sess. 327 (1974), reprinted in 1974 U.S.C.C.A.N. 5038, 5107.

[FN31]. Varity Corp. v. Howe, 516 U.S. 489 (1996).

[FN32]. 29 U.S.C. § 1003(a).

[FN33]. Mertens v. Hewitt Assoc., 508 U.S. 248 (1993).

[FN34]. 29 U.S.C. § 1144.

[FN35]. 463 U.S. 85 (1983).

[FN36]. 471 U.S. 724 (1985).

[FN37]. 481 U.S. 41 (1987).

[FN38]. See also Mackey v. Lanier Collection Agency and Servs., Inc., 486 U.S. 825 (1988). Importantly, "run of the mill" tort claims against nonfiduciaries, such as treating physicians, are not preempted and, therefore, malpractice claims against physicians, regardless of whether they are the product of a managed care incentive scheme, generally can proceed.

[FN39]. 965 F.2d 1321 (5th Cir. 1992).

[FN40]. 999 F.2d 298 (8th Cir. 1993).

[FN41]. 29 U.S.C. § 1132(b)(1). This is the so-called "claim preclusion."

[FN42]. Id. § 1144(a).

[FN43]. Id. § 1144(b)(2)(A).

[FN44]. 514 U.S. 645 (1995).

[FN45]. Id. at 661.

[FN46]. Id. at 659.

[FN47]. 117 S. Ct. 832 (1997).

[FN48]. Id. at 842.

[FN49]. 464 U.S. 238 (1984).

[FN50]. Dillingham, 117 S. Ct. at 843.

[FN51]. 117 S. Ct. 1747, 1752 (1997).

[FN52]. Pappas v. Asbel, 675 A.2d 711 (Pa. Super. 1996).

[FN53]. Tex. Civ. Prac. & Rem. Code § 88.002(a) (1997). A health insurance carrier, health maintenance organization, or other managed care entity for a health care plan has the duty to exercise ordinary care when making health care treatment decisions and is liable for damages for harm to an insured or enrollee proximately caused by its failure to exercise such ordinary care.

[FN54]. Haas v. Group Health Plan, Inc., 875 F. Supp. 544 (S.D. Ill. 1994).

[FN55]. Pacificare of Oklahoma v. Burrage, 59 F.3d 151 (10th Cir. 1995); Rice v. Panchal, 65 F.3d 637 (7th Cir. 1995); Dukes v. U.S. Health Care, 57 F.3d 350 (3d Cir. 1995).

[FN56]. 57 F.3d 350 (3d Cir. 1995).

[FN57]. Id. at 357.

[FN58]. Id.

[FN59]. Kampmeier v. Sacred Heart Hosp., 1996 U.S. Dist. LEXIS 5739 (E.D. Pa.).

[FN60]. See Varity Corp. v. Howe, 516 U.S. 489 (1996); Shea v. Esensten, 107 F.3d 625 (8th Cir. 1997).

[FN61]. 65 F.3d 637 (7th Cir. 1995).

[FN62]. 88 F.3d 1482 (7th Cir. 1996).

[FN63]. Chaghervand v. Carefirst, 909 F. Supp. 304 (D. Md. 1995) (claims asserting direct liability of an HMO for malpractice due to denied benefits are preempted); Prihoda v. Shpritz, 914 F. Supp. 113 (D. Md. 1996) (claim alleging vicarious liability for poorly delivered benefits is not preempted).

[FN64]. Edelen v. Osterman, 943 F. Supp. 75 (D.C. Cir. 1996).

[FN65]. J. MALCOLM, supra note 5, at 59, 63.

[FN66]. Id. at 62.

[FN67]. 105 N.E. 92 (N.Y. 1914).

[FN68]. Id. at 93.

[FN69]. J. MALCOLM, supra note 5, at 64.

[FN70]. Id.

[FN71]. 464 F.2d 772 (D.C. Cir. 1972).

[FN72]. Id. at 786-87.

[FN73]. 271 Cal. Rptr. 146 (Cal. 1990).

[FN74]. Id. at 153. In fact, the court notes in footnote 11 that the medical care received was necessary.

[FN75]. Id. at 150.

[FN76]. Id. at 152.

[FN77]. 516 U.S. 489 (1996).

[FN78]. 29 U.S.C. § 1002(21)(a).

[FN79]. 107 F.3d 625 (8th Cir. 1997).

[FN80]. Id. at 628.

[FN81]. 29 U.S.C. § 1002(21)(a)(iii).

[FN82]. 29 U.S.C. § 1109 covers liability for breach of fiduciary duty and imposes personal liability for losses due to breach of fiduciary duty and even removal of a fiduciary.

[FN83]. 29 U.S.C. § 1024(b). The plan "administrator" must furnish copies of any "contract, or other instrument under which the plan is established or operated."

[FN84]. Id. § 1002(21)(a).

[FN85]. Id. § 1002.

[FN86]. Havighurst, Practice Guidelines as Legal Standards Governing Physician Liability, 54 LAW & CONTEMP. PROBS. 87, 108 (1991).

[FN87]. Id. at 110.

[FN88]. Id.

[FN89]. Id. This is a problem that Havighurst advocates could be solved by contracting for practice guidelines. However, courts traditionally have been reluctant in the health care arena to permit such contracting of rights as against public policy. For a discussion of this tradition, see Tunkl v. The Regents of the Univ. of California, 383 P.2d 441 (Cal. 1963).

[FN90]. 29 U.S.C. § 1001. "It is ... the policy of this chapter to protect ... the interests of participants in EBPs and their beneficiaries ... well-being and security of ... employees ... improving the equitable character ... of such plans ... desirable to increase the likelihood that full benefits will be paid."

[FN91]. Leape, Error in Medicine, 272 J.A.M.A. 1851 (1994); Brennan, Leape, Laird, Hebert, Localio, Lawthers, Newhouse, Weiler, & Hiatt, Incidence of Adverse Events and Negligence in Hospitalized Patients: Results of the Harvard Medical Practice Study I, 324 NEW ENG. J. MED. 370 (1991).

[FN92]. Blumenthal & Epstein, Quality of Health Care, Part 6: The Role of Physicians in the Future of Quality Management, 335 NEW ENG. J. MED. 1328 (1996).

[FN93]. It is unclear how the "restitution" assessed against the fiduciary in Varity would work in a health plan setting, and we must await the final outcome of Shea following remand.

[FN94]. 29 U.S.C. § 1002(21)(a)(iii).

[FN95]. MacStravic, Price Wars Are No-win Games for Health Care Systems, 13 HEALTH CARE STRATEGIC MGMT. 5 (May 1, 1996) (now at 1,500 calls per month on its consumer HMO complaint line); Church, Backlash Against HMO's, TIME, Apr. 14, 1997, at 32 (in a poll conducted by the nonpartisan National Coalition on Health Care, 80% of respondents said they believed the quality of medical care is often compromised by insurance companies to save money).

[FN96]. Lippman, Can Employers See Beyond Price? A Survey of the Opinions of Employers with 20 Employees or More, on the Quality of Health Care Programs, and the Costs, 14 BUSINESS & HEALTH 44 (Dec. 1996).

[FN97]. MacStravic, supra note 95, at 5.

[FN98]. Lippman, supra note 96.

[FN99]. Findlay & Adams, The Coalition Movement Grows Up: Employer Coalitions and Health Purchasing Cooperatives HPC, 14 BUSINESS & HEALTH 16 (July 1996).

[FN100]. Id.

[FN101]. 29 U.S.C. § 1002(16)(A). ERISA defines "administrator" as "the person specifically designated by the terms of the instrument under which the plan is operated."