Risk management in an IPA setting - part I.

When physicians were seen as "captains of ships," they were responsible for the acts of other health professionals. Hospital risk management strategies have developed over the past 30 years, as courts have found hospitals liable for acts performed in their facilities. As medical care has shifted from the hospital to ambulatory care, some large ambulatory care facilities, such as multispecialty medical groups, have found a need to develop risk management programs of their own. They usually start with the hospital risk management model, although they deal with a different set of patient mishaps and somewhat different diagnoses. The development and growth of new health care delivery organizations, such as HMOs, have led to new legal strategies to extend liability for negligence. This requires a response by these organizations to manage risk.

Traditional Risk Management Functions

Risk management texts describe risk management processes in detail.[5-7] The basic elements of this process include:

* Risk Identification--identifying

areas of operation that put the

organization at risk. Broadly, these

include professional malpractice,

general liability, workers' compensation,

casualty exposure, hazardous

substance exposure, potential

environmental damage, transportation liability, defamation, embezzlement, antitrust, breach of contract, fraud and abuse, securities violations, and others.

* Risk Measurement--use of probability

data and prior loss history

to estimate the potential losses in

each area of risk identified.

* Risk Financing--accruing or

insuring for identified and measured

risks, or determining to pay

losses out of operating income.

* Risk Avoidance-reducing areas

of identified risk by limiting the

business to less risky ventures.

* Risk Control--methods to minimize

losses due to risk that the

business cannot avoid.

Each of these elements applies to IPA- and network-model HMOs. Senior management usually performs the first four elements. In collaboration with legal counsel, managers review business operations for liability risk. Managers estimate the dollar value of this risk and then decide to insure against losses or to self-insure. They may decide to contract for legal services or to develop an internal legal department. In some cases, the organization may forgo certain parts of the business as too risky. For example, a staff-model HMO may decide not to provide perinatal services (and contract for them instead) because the liability risk is not worth the potential operational cost savings.

Once senior management completes this analysis, the organization will still retain some risk. It may insure itself against that risk. Insurance companies vary their premiums according to the degree to which management controls risk. Organizations that decide to self-insure, or to purchase insurance with high deductibles, will retain more risk. Whatever the insurance decision, most organizations have an incentive to control risk.

Risk Control

Risk control is the function most people associate with risk management. There are several types of risk control:

* Risk Transfer--use of contracts or

other means to shift the financial

burden of losses to another party,

such as a supplier or a customer.

* Risk Prevention--policy and procedure

design, system design, and

educational programs to reduce

occurrences that put the organization

at risk.

* Risk Mitigation--a set of decisions

and activities undertaken,

once an adverse event has

occurred, to reduce the probability

of lawsuits.

* Litigation Management--actions,

taken after an allegedly

injured party has filed a lawsuit, to

lessen the financial, emotional,

and public relations impact of legal

claims. The strategies described above apply to any business. An organization that provides health care has a special risk: medical malpractice. Risk managers in health care may be responsible for all areas of risk, including workers' compensation, general liability, and hazardous substance disposal. Or, medical risk managers may limit the scope of their involvement to malpractice liability.

The lines between these categories of risk sometimes blur. For example, an HMO employee may also be a patient, and workers' compensation, general liability, and malpractice risk may arise from a single injury. In IPA-model plans, attempts to exclude a physician may subject the HMO to antitrust actions, but failing to expel a physician can incur the risk of tort and breach of contract claims. The focus of this article is malpractice risk.

Legal Principles of HMO Liability

In IPA- and group-model plans, the HMO contracts with independent providers to deliver care. When medical errors occur, the direct provider, rather than the HMO, is usually responsible. Yet there are several legal principles that have been or could be invoked to impose liability on the HMO.

The law recognizes three type of business relationships that may arise in the IPA setting, with varying degrees of liability--employer/ employee, agency, and independent contractor. The legal relationship between an HMO and a group or IPA physicians will vary with the details of the specific relationship and with the opinions of the court in the specific case.

In employer/employee relationships, the employer pays wages, withholds taxes, and directs details of work. An employer is legally responsible for most of its employees' negligent acts. Independent contractors are paid to perform a specific task. Once an organization hires an independent contractor, it has no right or ability to "oversee the details of the contractor's work. The institution is liable for the actions of the contractor only if it was negligent in selecting the contractor." In an agency relationship, a principal, such as an HMO, "is held liable for the actions of the agent that are directly related to the duties that the agent performs for the principal."[7] Respondeat superior, ostensible agency, and negligent selection are all agency relationship legal theories.

The respondeat superior principle states that "an employer is liable for the acts or omissions of its employees."[8] This principle is similar to vicarious liability. In Stewart v. Midani,[9] determination of vicarious liability depended on an assessment of a hospital's control over a physician practice. Whether the hospital contracted for a service (implying an employer/employee relationship) or contracted for a specific accomplished task (implying an independent contractor relationship) was part of that assessment. The court also considered the hospital's ability to direct work; its authority to control a physician's time, to inspect work, and to terminate the contract; the skill of the work; and the method of payment.

For HMOs, respondeat superior applies when the HMO employs the providers, as in staff-model programs. Some courts have held HMOs liable for the acts of contract consultants practicing at staff-model centers. The method by which the HMO engaged and selected the consultants, paid their wages, and controlled the physicians' behavior determines the HMOs liability. In theory, the principle could extend to IPA physicians if the HMO controlled enough of the physicians' practices, determined referral patterns, and dictated rules of medical practice.

In Schleier v. Kaiser Foundation Health Plan,[10] the court held a staff-model HMO liable for the negligence of nonemployed consultants because the HMO:

* Selected the physician.

* Paid wages to the physician.

* Retained the power to discharge

the physician.

* Had the power to control the physician's

behavior.

* Used the physician for work that

was part of the regular business of

the HMO. In Sloan v. Metropolitan Health Council,[11] an appeals court found that an HMO could be found vicariously liable for the acts of its employed physicians because they were under the control of a medical director. More recently, the appeals court in Dunn v. Prais[12] found the HMO vicariously liable for the acts of a urologist because it paid the urologist by capitation rather than fee-for-service, the urologist examined patients at the HMO's offices, the urologist could not reject specific patients, and the HMO determined subsequent visits to the urologist.

IPA-model HMOs usually select and pay participating physicians. They retain the power to discharge physicians, albeit through due process mechanisms. They do not generally withhold taxes, but "the legal test of employment, irrespective of a person's tax status, is the degree of control that the employer can exercise over the person's work."[8] A key issue will be whether the HMO has the power to control physician behavior. If the HMO performs QA and UR reviews, enforces documentation standards or practice guidelines, or otherwise controls how its affiliated physicians will practice medicine, the "Schleier test" could be successfully applied and the HMO might be held liable for the negligence of its IPA physicians.

Some HMOs pay their physicians indirectly. For example, the HMO may allocate a portion of revenue to a physician pool. The IPA management, and not the HMO, determines the distribution of that pool to individual providers. The less direct the payments, the weaker the argument that the HMO paid wages to the physicians. Indirect payment schemes therefore may provide a defense against the "Schleier test."

In group-model and network-model HMOs, the group is the employer of the physician. The HMO contracts with the group and may have no relationship with the individual physicians at all. The HMO is often unable to select participating physicians, pay them, or discharge them. The group president or administrator controls physician behavior. It is unlikely that use of the respondeat superior principle to place liability on the HMO in a group model will be successful.

Ostensible agency is a principle that holds one organization responsible for acts by another if a third party would reasonably consider the second organization to be an agent of the first. If patients rely on the HMO to provide care for them and see their physician as an agent of the HMO, the HMO may be held accountable for the acts of the physician. Boyd v. Albert Einstein Medical Center[13] is an example of this principle. In Boyd, the HMO did not employ the physician, but it did use primary care physicians as gatekeepers. Further, the HMO had advertised its process to select physicians who could participate in the HMO. If an HMO advertises that its physicians are competent or carefully selected, the HMO could be held liable for their acts.

In McClellan v. Health Maintenance Org. of Pennsylvania[14] and Decker v. Saini,[15] courts found HMOs could be held liable under the principle of ostensible agency. In both cases, the HMO's motion to dismiss the case was denied by an appeals court. In the former case, the court found that the HMO presented the physician as its agent. The plaintiff claimed that the HMO had stated its physicians were qualified and competent. In the latter case, the primary care physician referred the patient to a nonplan specialist. The plaintiff claimed that they looked to the HMO for treatment, not merely for payment for treatment. The HMO restricted patients' choice of physicians. The patient paid the HMO, not the physicians.

In Raglin v. HMO Illinois,[16] however, the HMO was not liable under ostensible agency theory. The HMO did not assess the accuracy of medical opinions. It specifically advised its members that it did not directly provide medical care. The court found it played only an administrative role in overseeing the delivery of health care.

This principle could be applied to individual IPA physicians, as in the Boyd case, or to groups. The HMO can try to minimize its liability exposure by informing patients that the group is a contractor and not an agent of the HMO. However, this strategy may conflict with a marketing strategy that assures patients that the HMO is managing their health care.

Even if the relationship between an HMO and its participating physicians is deemed that of an independent contractor, the HMO could be liable for injuries to patients under the theory of negligent selection. In Harrell v. Total Health Care. Inc.,[17] a court affirmed the duty of the HMO to ensure the competence of its contractors. In this case, cursory credentialing did not reveal a malpractice history. The court found that the HMO had a duty to select its participating physicians.

Aside from respondeat superior, ostensible agency, and negligent selection, injured patients have used other legal principles against HMOs. The HMO can be held accountable by the theory of corporate liability. The HMO has corporate duties that extend at least to credentialing and performance of legislatively mandated quality assurance activities. The arguments for corporate liability often parallel those of negligent selection. In fact, negligent selection could be considered a subset of the corporate liability principle.

A class action suit alleged violation of the Racketeer Influenced and Corrupt Organizations Act of 1970 (RICO).[18] In this case, the HMO's financial incentive system was alleged to have caused damage. Because promotional materials did not disclose these incentive systems, the HMO was charged with misrepresentation. This argument was rejected by the court--this time.

A breach of warranty occurs when an HMO explicitly or implicitly warrants to its members to provide care of a high level of quality and that level of quality does not ensue. In Pulvers v. Kaiser Foundation Health Plan,[19] the court asserted the basis of this principle, though it found it did not apply in this case. The key determinant was whether the HMO promised a particular result that did not occur.

In the contract between the HMO and the medical group, the medical group may warrant to the HMO a high level of quality of care. Because the patient's written contract is with the HMO, not the medical group, a patient injured negligently could sue the HMO for damages. In theory, the HMO could then sue the medical group to recover those damages. Such countersuits will likely damage the customer-supplier relationship between the HMO and the medical group. Legal fees for such suits could be expensive. This is true even if the HMO is only a conduit for the reimbursement of damages from the medical group to the patient.

Risks in Utilization and Quality Management

A key component the HMO employs in managing health care is utilization review, which carries its own liability. The legal risks inherent in performing utilization review may involve negligence, breach of contract, insurance bad faith, infliction if emotional distress, and others.[20] The Employee Retirement Income Security Act (ERISA) limits potential losses on such cases when they are related to denial of plan benefits. Lawsuits under ERISA do not allow damages for bodily injury, property damage, lost wages, or emotional distress. Several cases[21],[22] have affirmed the preeminence of ERISA over state laws for patients who are members of plans that fall under ERISA. Courts have not ruled medical malpractice and ostensible agency claims preempted by ERISA, but claims of negligent administration, breach of contract, and intentional misrepresentation from interpretation or administration of benefits have been countered by invoking ERISA. Independent enrollees are not constrained from suing on these grounds. ERISA only applies to employer group benefit programs.

Utilization review decisions that adversely affect patient care because of a flawed process can be deemed negligent. Suppose an HMO fails to follow its own procedures or guidelines for denial of a requested procedure or consultation, and the patient is injured. The HMO's utilization review program could be held liable. This liability is separate from any act by a practitioner actually caring for the patient. In essence, utilization review can be construed as practicing medicine, and potential risk follows this activity.

Wickline is the most famous utilization review case.[23] While the utilization review function of the Medi-Cal program was not held liable in this case, the court recognized the potential for liability of negligent administration of cost control programs. Specifically, in this case, the physician did not avail himself of the appeal mechanisms. The court stated that liability could occur if an appeal is filed and not given fair treatment. Other cases have since addressed this issue. The appeal court in Wilson v. Blue Cross of S. Cal.,[24] noting the treating physician's testimony that denial of benefits caused him to discharge the patient from the hospital, overturned a summary judgment for the insurer and sent the case back to trial. The court did not award damages to the plaintiff, but the case still acts as a precedent.

More recently, a California court awarded more than $70 million dollars to a plaintiff for denial of treatment. This case will likely be appealed. Health Net of California denied prior authorization for an autologous bone marrow transplant for a patient with metastatic breast cancer. Other HMOs had a similar policy. Because studies of the effectiveness of that procedure for that disease were inconclusive, Health Net deemed the treatment experimental. A key factor in this court's decision was the inconsistent way in which Health Net applied its authorization guidelines.

Assuring fairness in benefit interpretation decisions, and a defined process for making those decisions, will help avoid such awards. A bioethics decision-making model would likely better meet the fairness test. Bioethics committees use interdisciplinary discussions to reach consensus. Bioethics case reviews identify stakeholders, make each stakeholder's interest clear, include participants from outside the organization, and include a patient advocate.

Specifically defining benefit exclusions at the time of a member's enrollment could also provide some protection. Terms such as "experimental treatments" are subject to differing opinions. Unfortunately, more specific exclusion language may quickly fall out of date with the publication of new evidence on effectiveness.

The more the HMO directs physicians in their medical practice, the more likely that respondeat superior applies. However, key customers, including large employers and federal agencies, expect HMOs to manage the quality and cost of medical care. For example, the National Committee on Quality Assurance, which accredits HMOs, includes implementation of practice guidelines as a standard. It will be difficult to convince these customers that the HMO controls medical care and simultaneously convince a court that the HMO does not control physician behavior.

In Maine, physicians may present their conformance to a practice guideline as a defense in malpractice actions. This experiment may show that guidelines reduce malpractice litigation. The legislation does not protect HMOs, but, if fewer physicians are sued, their HMOs will likely also be sued less. However, if an HMO develops a faulty guideline, it may be held responsible for damages occurring when physicians follow the guideline. For that reason, HMOs should ensure that their guideline development processes meet tests of quality. A good guideline is validated by outcomes or a consensus panel, specifies the patient subset to which it applies, and is updated periodically and whenever new information requires. Assuring physicians have the most recent version is a challenge.

As HMOs measure quality and change systems to improve performance on quality indicators, they take on some risk. A poorly done study may lead to erroneous conclusions. Physicians who rely on those conclusions may implicate the HMO in the process, leading to damages. Every endeavor entails some risk. The best defense is to perform well.

As the cases above illustrate, plaintiffs can successfully use these legal principles to obtain damages from HMOs in IPA settings. HMOs have rarely paid damages. Many cases are dismissed or settled. Yet, even if the HMO is dismissed from a suit, defending such suits can be expensive. For example, pretrial preparation can cost $15,000 or more. Expenses rise if the case is ultimately tried. Thus, IPA- and group-model HMOs may have a financial risk if participating physicians are deemed negligent. How can an HMO apply the principles of risk management described above to minimize this risk in contracted care? That will be the subject of the second part of theis two-part article, in the June 1994 issue of Physician Executive. 11

References

[1.] Johnson, D. Marion Merrell Dow Managed Care Digest, HMO Edition. Chicago, Ill.: Marion Merrell Dow, 1991, p. 5. [2.] Stocker, M. "Quality Assurance in the IPA Setting." HMO Practice 3(5):183-7, Sept.-Oct. 1989. [3.] Tan, S. "Quality Assurance in an IPA Setting." Physician Executive 17(3):43-5, May-June 1991. [4.] Romano, P. "Quality Assurance in a Health Maintenance Organization." FHP Journal of Clinical Research 1(1):24-6, 1990. [5.] Kraus, G. Health Care Risk Management. Owing Mills, Md.: Rynd Communications, 1986. [6.] Monagle, J. Risk Management: A Guide for Health Care Professionals. Rockville, Md.: Aspen Publications, 1985. [7.] Richards, E., and Rathbun, K. Medical Risk Management. Rockville, Md.: Aspen Publications, 1983. [8.] Mulholland, D. "Managing Care and the Risk for Managing Quality." Legal brief, Horty, Springer, and Mattern, P.C., Pittsburgh, Pa. [9.] Stewart v. Midani, 525 F. Supp. 843 (N.D. Ga 1981). [10.] Schleier v. Kaiser Foundation Health Plan, 876 F.2d 174 (D.C.Cir. 1989) [11.] Sloan v. Metropolitan Health Council, 516 N.E. 2d 1104 (Ind. Ct. App. 1987). [12.] Dunn v. Prais, 256 N.J.Super. 180 606 A.2d 862 (1992). [13.] Boyd v. Albert Einstein Medical Center, 547 A.2d 1229 (Pa. Super. 1988). [14.] McClellan v. Health Maintenance Org, of Penn., 604 A.2d 1053 (Pa. Super. Ct. 1992). [15.] Decker v. Saini, No. 88-361768 NH, 1991 WL 277590 (Mich. Cir. Ct. Sept. 17, 1991). [16.] Raglin v. HMO Illinois, Inc., 230 Ill. App. 3d 642, 595 N.E.2d 153 )1992). [17.] Harrell v. Total Health Care, Inc., 781 SW2d 58 (Mo. 1989). [18.] Teti v. U.S. Healthcare, Inc., No. 88-9808 (E.D. Pa. Nov. 20,1989). [19.] Pulvers v. Kaiser Foundation Health Plan, 99 Cal. App. 3d 560,160 Cal.Rptr.392 (1979). [20.] Jespersen, K. "Managed Health Care Programs: Reduce Exposure to Liability." The Brief 21(1):19-39, Fall 1991. [21.] Corcoran v. United Healthcare, Inc., 965 F.2d 1321 (5th Cir. 1992). [22.] Elsesser v. Hospital of the Philadelphia College of Osteopathic Medicine, 795 F. Supp. 142 (E.D. Pa. 1992). [23.] Wickline v. State of California, 183 Cal. App. 3d 1175,228 Cal. Rptr, 661; 231 Cal. Rptr. 560, 727 P.2d 753 (1986); 239 Cal. Rptr. 805, 741 P.2d 613 (1987). [24.] Wilson v. Blue Cross of S. Cal., 222 Cal. Ap. ed 660, 271 Cal. Rptr. 876 (1990).

Jonathan Harding, MD, FACPE, is Medical Director, FHP, Inc., Cerritos, Calif. He is a member of the College's Forums on Bioethics and Quality Health Care and its Societies on Managed Health Care Organizations and Hospitals