by Guy O. Farmer, II and John H. Douglas

Page 37

In May 1975, when anesthesi ologists protested the spiraling malpractice premiums of the day by refusing to assist in elective surgeries for four weeks, half of the hospitals beds in San Francisco emptied and hospital financial losses soon ran into the millions. Though such action was probably an illegal “group boycott” then—and would almost certainly attract the wrath of the Federal Trade Commission or Justice Department today—the result in 1975 was the passage of legislation capping damage awards to medical malpractice plaintiffs in California and, at least temporarily, the pacification of many physicians.

Enter “managed care,” a phenomenon responsible, by some estimates, for reducing physician incomes 44 percent from where they would have stood in its absence.¹ Despite its inroads of managed care in 1996, the income of the average physician in America was still six times the national per capita gross domestic product (compared, for example, with 2.4 times in Japan and Denmark). Nevertheless, physicians with fond memories of the 1975 boycott, particularly those already in unions, have once again begun touting the virtues of physician “unionization.” The (frequently) stated goal expressed to a newly receptive physician audience has been preservation of the “quality of care.” The actual goal, albeit less frequently or publicly stated, may in fact be the preservation of professional incomes and clinical discretion. This article presents an overview of the legal landscape surrounding this recent phenomenon—popularly referred to (with varying degrees of accuracy) as the physician “unionization” movement.

The Primer
Any discussion of true physician unionization ought to begin with the federal law that governs the right of individuals in the private sector to “unionize”—the National Labor Relations Act (NLRA), the centerpiece of which, §7, provides that “employees shall have the right . . . to form, join, or assist labor organizations . . . for the purposes of collective bargaining.”² Given the NLRA’s limitation to “employees” as defined in the statute, however, a brief digression into events predating the 1935 passage of the NLRA by 100 or more years is necessary background for the discussion.

Beginning in the late 1700s, medical societies and elected bodies in several cities and states initiated efforts to transform what had until that time been a loose amalgam of “healing arts” into a “profession,” and by 1830, five states—Massachusetts, South Carolina, New York, Maryland, and Ohio—had laid the foundation for the professional regulation of physicians.³ In 1847, members of the New York Medical Society formed the American Medical Association, an entity which promptly promulgated academic standards for medical education. Eventually, with the active encouragement of the AMA, state courts began to find as a corollary to state professional licensing laws limiting the practice of medicine to natural persons, that corporations could not employ licensed physicians without engaging in the unlicensed “practice” of medicine.⁴

This legal doctrine, which came to be known as the “corporate practice doctrine,” has eroded somewhat in recent years, particularly as the proliferation of innovative health care delivery models have made it increasingly difficult for state legislatures (and attorneys general) to maintain outright bans on the employment of physicians.⁵ Nonetheless, the doctrine remains viable today in one form or another in more than 30 states,⁶ with some states, such as California, enforcing their ban quite aggressively,⁷ while others, like Florida, have more or less consigned the doctrine to the dustbin of history.⁸ The perdurable impact of this “professional” history, however, is that, even today, probably fewer than 30 percent of the more than 600,000 physicians practicing in the United States are “employed” by entities covered by the NLRA, and are thus even eligible to “unionize” in the traditional sense.⁹

Indeed, even in states where physicians can be “employed” within the meaning of the NLRA, at least two other significant exceptions to the statute’s coverage, one found in the statute itself, the other, the product of administrative policymaking, further limit the current ability of many physicians to “unionize” in the traditional sense.¹⁰

First, the text of the NLRA itself provides that “the term ‘employee’ . . . shall not include . . . any individual employed as a supervisor,” and the statute separately defines a “supervisor” as “any individual having authority in the interest of the employer, to hire, transfer, suspend, lay off, recall, promote, discharge, assign, reward, or discipline other employees, or responsibly to direct them, or to adjust their grievances, or effectively to recommend such action, if in connection with the foregoing the exercise of such authority is not of a merely routine or clerical nature, but requires the use of independent judgment.”¹¹ A large body of cases, much of which has arisen in the health care context, interprets this statutory definition, though their conclusions on similar sets of facts are often virtually impossible to reconcile with each other. As a practical matter, however, if a physician actually performs (or can effectively recommend) any of the 12 enumerated supervisory functions other than “responsible direction,” there is a good chance he or she will be a “supervisor,” and thus, will be ineligible to “unionize” in the traditional sense.

In contrast, the gloss that the National Labor Relations Board (NLRB) has given the “supervisory” duty perhaps most widely practiced by health care professionals such as physicians—“responsible direction”—has been a consistent source of controversy, and for the second time in seven years, is before the U.S. Supreme Court, this time on review of a Sixth Circuit decision rejecting the NLRB’s most recent interpretation.¹²

In particular, prior to a 1994 Supreme Court decision,¹³ the NLRB had limited the scope of the “responsible direction” “supervisory” exclusion by reasoning that when medical professionals direct subordinate caregivers, they do so not “in the interest of the employer” (as required by the statute), but rather in the interest of the patient. The Supreme Court rejected this reasoning, however, characterizing it as a “false dichotomy.” Undeterred, shortly after the Supreme Court’s decision, the NLRB found another rationale to reach the same result, this time reasoning that when medical professionals direct subordinate caregivers, they do so not by using their “independent” judgment, but rather by using their “routine professional” judgment.¹⁴ Whether the Supreme Court views this reasoning as involving another “false dichotomy” will soon be known.

Second, even if a physician is not an independent contractor or a supervisor, he or she may still be separately excluded from the coverage of the NLRA if found to be a “manager.” Unlike the exclusions for independent contractors or supervisors found in the text of the statute, the NLRA’s “managerial” exclusion is a pure product of NLRB policymaking in the 1940s (but approved by the Supreme Court¹⁵) based on a common-sense recognition that to allow managers to unionize would inherently create an unworkable conflict of interest. Under NLRB precedent, a “manager” under the NLRA is an employee “in a position to formulate, determine and effectuate management policies,”¹⁶ who has “discretion in the performance” of his or her job “independent of the employer’s established policies.”¹⁷ Again, the many decisions applying these standards are often difficult to reconcile, but generally speaking, the more often a physician makes and implements final decisions regarding important corporate policies, and the less that decisionmaking derives from his or her professional (i.e., medical) expertise, the more likely the physician will be a “manager” unable to unionize under the NLRA.

In a landmark 1980 case, the Supreme Court significantly expanded the potential application of this exclusion to physicians when it found that the entire faculty of Yeshiva University in New York was “managerial,” and thus ineligible to unionize under the NLRA, since it pervasively governed itself through a structure of democratic decision-
making bodies.¹⁸ Five years later, in 1985, the Reagan-era NLRB applied this precedent to a health maintenance organization (HMO) operating in Southern California and found that since that organization’s employed physicians all either actually (or potentially) participated in “medical staff” committees with significant policymaking authority, they were, like the faculty of Yeshiva University, all managers, and thus ineligible to organize under the NLRA.¹⁹

The Progress of the “Movement”
Given these many impediments to traditional “unionization” faced by physicians, not long after the first rumblings in favor of such action began in the mid-1990s, physicians also began exploring other avenues to achieve their ends, many which were also characterized (sometimes erroneously) as physician “unionization.”

In particular, in a landmark test case filed in November 1997, a group of physicians who had contracted with AmeriHealth, a New Jersey HMO, filed a petition with the NLRB, asserting that they were AmeriHealth “employees” and asking for an election to determine whether the United Food and Commercial Workers’ Union should represent them in collective bargaining with AmeriHealth.

Prior to the hearing on this petition, the NLRB regional director with jurisdiction over the matter ordered the parties to brief the issue of whether the physicians were in fact “employees” covered under the act. AmeriHealth subsequently presented evidence that its average participating physician was on more than four other HMO panels, and in 1996, had received an average of $2,657.18 from AmeriHealth. Though the evidence also showed that AmeriHealth had promulgated substantial requirements governing the credentials and operations of its “panel” physicians, the NLRB regional director found that many of these requirements derived from the New Jersey HMO act and thus discounted them as evidence of the control of an “employer.” Based on this evidence, on January 8, 1998, the regional director dismissed the physicians’ petition without a hearing. Subsequently, the union successfully appealed this dismissal, however, and the matter was remanded for a full hearing. On May 24, 1999, the same regional director dismissed the petition a second time, having satisfied herself following the hearing that the physicians at issue were in fact “independent contractors.” This time, the NLRB upheld the regional director’s decision on review.²⁰

With the adverse decision in the AmeriHealth case, the focus of a significant number of independent contractor physicians attracted to the idea of “unionization” increasingly turned to affiliation with so-called “messenger model” “independent practice associations” (IPAs) organized by, among others, a number of physician “unions” such as the Tallahassee-based Federation of Physicians and Dentists. If operated in conformity with guidelines promulgated by the Federal Trade Commission, such IPAs can aggregate offers to deliver health care services on behalf of groups of “independent contractor” physicians, thus engaging in a kind of “collective” bargaining without violating federal antitrust laws. In addition, independent contractor physicians increasingly began turning to self-styled physician “unions,” such as the Florida Physicians’ Union, an affiliate of the Jacksonville-based Florida Physician Association, to file class action claims against insurers for breach of contract.

Perhaps most significantly, however, following the AmeriHealth decision, independent contractor physicians in many states began lobbying their state legislators and members of Congress for outright relief from state and federal antitrust restraints. Indeed, to date, legislation had been introduced in 18 states (including H.B. 803, introduced March 2, 2001, by Florida State Representative Negron) as well as in the District of Columbia, whereby groups of physicians would be allowed to bargain collectively under the supervision of state agencies in order to avail themselves of the “state action” exemption to the federal antitrust laws.²¹ With the exception of Texas, however, where then Gov. George W. Bush signed such legislation on June 20, 1999, none of these efforts have been successful.²²

On the federal front, in 1998, Rep. Tom Campbell (R.Ca.), the former director of the Federal Trade Commission’s Bureau of Competition, introduced legislation in the 105th Congress that would have exempted NLRA-style collective bargaining by independent contractor physicians from antitrust prosecution. On March 25, 1999, Campbell reintroduced the legislation in the 106th Congress, and on June 30, 2000, the bill passed the House on a 276-136 vote.²³ A companion bill remains to be introduced in the Senate, however, and Campbell, having failed to oust Diane Feinstein from her Senate seat in the November, 2000, election, has returned to teaching at Stanford Law School.

At the same time, unions were also undertaking cooperative efforts to pursue the unionization of “employee” physicians more efficiently. In particular, in March 1999, the AFL-CIO’s Service Employees International Union announced the merger of three previously independent unions, The Doctors Council, the United Salaried Physicians and Dentists, and the Committee of Interns and Residents (CIR), thereby creating the 15,000-member “National Doctors Alliance,” an entity with a $1 million annual organizing budget. Not to be left behind, in June 1999, the AMA joined the parade as well when it voted to form an independent union of its own, “Physicians for Responsible Negotiations” (PRN). Five months later, in November 1999, the physician unionization movement would receive yet another shot in the arm when the NLRB would decide a case involving interns and residents at Boston Medical Center.

The Boston Medical Center Case
Boston Medical Center (BMC) was the product of the 1996 merger of Boston City Hospital (BCH), a public hospital, and Boston University Medical Center Hospital (“University Hospital”), a private entity. Since 1969, BCH had bargained with its interns and residents (or “house staff”) collectively under state law. University Hospital, in contrast, was subject to the NLRA, but under longstanding NLRB policy its residents and interns could not unionize.²⁴

Under the merger agreement between BCH and University Hospital, BMC was committed to continue recognizing the union at BCH and had agreed to a “private” election among the entire house staff at the merged entity, which the house staff “won,” leading BMC to engage in collective bargaining at both hospitals. Subsequently, however, the AFL-CIO’s Committee of Interns and Residents filed a petition for an election with the NLRB. On October 17, 1997, the NLRB’s regional director in Boston dismissed the CIR’s petition, relying on 20-year-old NLRB precedent finding “house staff” to be primarily “students” rather than “employees,” and thus are not covered by the NLRA. The CIR appealed.

On appeal, in urging the NLRB to adhere to its historical policy, BMC argued that allowing house staff to bargain collectively would interfere with academic tradition and freedom by making educational standards the subject of bargaining. BMC further argued that Congress had evinced a clear intent that residents be excluded from the definition of “employee” when it rejected a 1979 proposal to amend the NLRA that could have effectively included residents in that definition. The CIR, in contrast, argued that the board’s precedent on the issue was no longer consistent with several subsequent Supreme Court cases that had interpreted the term “employee” under the NLRA quite broadly,²⁵ and that the board had misconstrued the motivations of house staff when it had concluded in earlier decisions that their financial interests were of secondary importance to them, pointing to the fact that house staff in the public sector had been organizing for years (though not under federal law) in support of this position. Ultimately, on November 26, 1999, the NLRB announced that it newly agreed with the CIR’s position, and that it had decided to overturn its precedent by finding that BMC’s residents, interns, and fellows were “employees” under the NLRA. Following this decision, on December 21, 1999, an election was conducted in a bargaining unit composed of “all physicians” employed by BMC—one that the union won by a lopsided count of 177 to 1, and BMC subsequently determined that it would bargain with the union, effectively precluding an appeal of the NLRB’s decision. Any challenge to the board’s new view of residents under the NLRA will thus have to wait until another day: when a union succeeds in organizing another group of residents and another “employer” hospital decides—in contrast with BMC—to challenge the NLRB’s new interpretation of the statute.

Where Things Stand
The CIR estimates that there are some 90,000 resident physicians in the United States who can newly organize under the NLRA and has indicated that it intends to use its $1 million annual organizing “war chest” to organize at least 50,000 residents. For its part, the AMA has budgeted $1.2 million for PRN to use with the goal of successfully organizing physicians.
PRN had its first such success in March 2000, when physicians employed at several clinics owned by The Wellness Plan, a Detroit HMO, voted 27 to 8 in favor of union representation. More recently, on August 16, 2000, PRN filed a first petition on behalf of a house staff at Lutheran General Hospital in Park Ridge, Illinois, a Chicago suburb. On November 7, 2000, the Regional Director ordered an election be held on December 6 and 7, 2000. Lutheran General appealed the regional director’s decision, however, claiming that it improperly excluded certain rotating residents from voting, and thus, though the voting went ahead as ordered, the ballots have been impounded pending the outcome of that appeal. In early January 2001, the PRN filed an unfair labor practice charge accusing Lutheran General of spying on residents and a variety of other unfair labor practices, which Lutheran General and the PRN subsequently settled.

A Prescription
Many hospital and residency program administrators may be wondering whether there is anything they should be doing in the face of this newly energized “unionization” movement.

From a management labor lawyer’s perspective, as in any industry where unions are involved, an “ounce of prevention” will likely be worth a pound of cure. In particular, perhaps the best method of minimizing the perceived need for a union is to improve communication between “management” and the “rank and file.” Historically, some of the most effective methods of encouraging such communication are through mechanisms such as suggestion boxes or periodic employee polls. Even more effective—although at times legally problematic—are committees in which employees regularly share information and concerns with management. Although an employer that establishes such committees purely in order to “dominate” them may be found in violation of the NLRA, hospitals are already independently required to maintain many “medical staff” committees as a condition of participation in the Medicare program, and, if they have a medical residency program, to maintain committees including resident representatives as a condition of accreditation with the Accreditation Council for Graduate Medical Education (ACGME). Hence, the risk that such committees might be deemed management-dominated labor organizations is probably inherently less than in other industries, and employers should consult with legal counsel in order to make full use of them as communication devices.²⁶ Indeed, if such committees have sufficient policymaking authority, and physician participation in those committees becomes sufficiently pervasive, not only may any perceived need for “unionization” dissipate, a hospital may have a good argument that, like the faculty of Yeshiva University, all its employed physicians are managers.


¹ Krishna Kundu, “A Union Is What The Doctor Ordered,” Knight Ridder/Tribune Information Services (May 24, 2000).

² 29 U.S.C. §157.

³ See Joseph F. Kett, The Formation of the American Medical Profession: The Role of Institutions, 1780-1860, Appendix B (1968).

⁴ See, e.g., People v. Woodbury Dermatological Institute, 192 N.Y. 454, 854 N.E. 697 (N.Y.1908).

⁵ See generally Adam M. Freiman, The Abandonment of the Antiquated Corporate Practice of Medicine Doctrine: Injecting a Dose of Efficiency Into the Modern Health Care Environment, 47 Emory L.J. 697 (1998).

⁶ See D. Cameron Dobbins, Survey of State Law Relating to the Corporate Practice of Medicine, 9 The Health Lawyer, no. 5 (1997).

⁷ See, e.g., California Medical Association, Inc. v. Regents of the University of California, 79 Cal.App.4th 542 (2000).

⁸ See Fla. Rev. Statutes Ch. 458; Rush v. City of St. Petersburg, 205 So. 2d 11 (Fla. Dist. Ct. App. 1967) (finding that hospital employment of a physician does not constitute unauthorized practice of medicine to the extent that the physician maintains independent judgment in making medical decisions).

⁹ See generally David W. Emmons & Phillip R. Kletke, The Practice Arrangements of Patient Care Physicians 1998 (1999). Employees of state and local governments are not covered by the NLRA, See 29 U.S.C. §152(2).

¹⁰ In addition, as discussed infra, at least until late 1999, an estimated 90,000 interns and resident physicians were deemed to be primarily “students” (rather than “employees” under the NLRA) and for this reason as well, were unprotected under federal labor law.

¹¹ 29 U.S.C. §§152(3)(11).

¹² See Kentucky River Community Care v. NLRB, 193 F.3d 444 (6th Cir. 1999).

¹³ See NLRB v. Health Care & Retirement Corp. of Am., 511 U.S. 571 (1994).

¹⁴ See Providence Hospital, 320 NLRB 717 (1996), enf’d, 121 F.3d 548 (9th Cir. 1997).

¹⁵ See NLRB v. Bell Aerospace, 416 U.S. 267 (1974).

¹⁶ Ford Motor Company (Chicago Branch), 66 NLRB 1317 (1946).

¹⁷ Palace Laundry Dry Cleaning Corp., 75 NLRB 320 (1947).

¹⁸ See NLRB v. Yeshiva University, 444 U.S. 672 (1980).

¹⁹ FHP, Inc., 275 NLRB 168 (1985).

²⁰ See AmeriHealth, Inc., 329 NLRB No. 76 (1999).

²¹ See Parker v. Brown, 317 U.S. 341 (1943); see also generally William G. Kopit, White Coats and Blue Collars: Physician Collective Bargaining Legislation on the National and State Levels, ALI-ABA (2000).

²² Texas S.B. 1486 went into effect on September 1, 1999, and regulations were promulgated effective June 6, 2000, whereby physicians in Texas can apply to the state attorney general for permission to engage in supervised collective negotiations with insurers.

²³ See H.R. 1304, 106th Cong. (1999).

²⁴ See Cedars-Sinai Medical Center, 223 NLRB 251 (1976); St. Clare’s Hospital, 229 NLRB 1000 (1977) (unit comprised of students only not conducive to bargaining since interests are more academic than economic).

²⁵ See, e.g., Sure-Tan, Inc. v. NLRB, 467 U.S. 883 (1984) (finding individuals to be “employees” within the meaning of NLRA despite their undocumented alien status).

²⁶ See, e.g., NLRB v. Peninsula General Hospital Center, 36 F.3d 1262 (6th Cir. 1994) (finding nursing committee not a dominated “labor organization” under NLRA).


Guy O. Farmer is a partner in the Jacksonville office of Foley & Lardner. he represents employers in the full range of employment-related problems. Mr. Farmer is a graduate of the University of Virginia Law School (LL.B., 1966) and West Virginia University (B.A., 1963).

John H. Douglas is an associate in the San Franciso office of Foley & Lardner. A member of the firm’s labor and employment practice group, his practice focuses largely on health cae employment related matters. Mr. Douglas is a graduate of Boalt Hall School of Law, University of California, Berkeley (J.D., 1995) and Harvard University (B.A. 1983).

This column is submitted on behalf of the Labor and Employment Law Section, Stuart Alan Rosenfeldt, chair, and F. Damon Kitchen, editor.