ON THE COVERGay Men and Lesbians in Medicine: Has Discrimination Left the Room?Uneasy Partners: The Lesbian and Gay Health Care Community and the AMADomestic Partnership Benefits at Medical UniversitiesHarassment of Lesbians as Medical Students and PhysiciansSexual Orientation and Youth Suicide FREE

Firenze by Rowland P. Robinson, Columbia University College of Physicians and Surgeons

Homosexuality remains controversial in this country. Last year's brutal murder of Wyoming college student Matthew Shepard brought forth recognition that despite much greater acceptance of gay and lesbian people than in recent history, antihomosexual discrimination marked by occasional violence remains a divisive fixture in the American social landscape.¹

Certainly conditions for gay and lesbian people have improved, especially within the medical profession. A moderate number of gay and lesbian medical students and physicians are now formally protected by antidiscrimination clauses in their academic and working lives. Many gay and lesbian physicians achieve high degrees of professional success, even after having disclosed their orientation to select colleagues and employers.

In this issue, Jason Schneider and Saul Levin, MD, trace the American Medical Association's growing acceptance of gay and lesbian visibility within its own ranks over the past 2 decades. Kate O'Hanlan, MD, follows with a discussion of her experiences over the past 10 years in successfully advocating for domestic partner benefits at Albert Einstein College of Medicine and Stanford University.

Yet despite successes in professional representation and employment status, surveys have documented that many admissions officers and residency directors are less enthusiastic about gay and lesbian candidates than heterosexual candidates, and that physicians who disclose a nonheterosexual orientation to colleagues face the potential loss of referrals and privileges.^2- 3 A nationwide survey of women physicians by researchers at Emory University provides new evidence that lesbian physicians do in fact experience harassment at a greater prevalence than their heterosexual counterparts.

From the patient's perspective, sexual orientation may expose one to specific health risks. Gary Remafedi, MD, provides a sobering view of the most catastrophic result of antihomosexual discrimination in his review of studies linking adolescent suicide to gay or lesbian sexual orientation. Suicide is among the most dangerous health risks of all to gay and lesbian teenagers and the most unrelentingly tragic.

Finally, this month's MSJAMA online features an interview with and new poetry by Rafael Campo, MD, who has made a point of addressing gay issues and the patient-physician relationship in his writing. Campo inspires readers with his injunctions to remain human as a first priority—and thereby to experience empathy with others. One may hope, as Campo's writing and these articles suggest, that these special insights of the medical profession have the potential to better a society still struggling with its prejudices.

Organized medicine's increased acceptance of lesbian and gay physicians and patients indicates greater recognition that sexual orientation influences health care delivery. The American Medical Association's (AMA's) 20-year history relating to these issues reflects the evolution of organized medicine's response to sexual orientation concerns.

In December 1973, the American Psychiatric Association deleted homosexuality from its list of mental disorders. Though no longer a diagnosis requiring correction or treatment, homosexuality remained controversial in the medical establishment. At the AMA's 1980 annual meeting, the House of Delegates, via a Medical Student Section (MSS) resolution, requested a "study of the health care needs of homosexuals."¹ The resulting report by the Council on Scientific Affairs was adopted as policy in December 1981 and published in JAMA.^1- 2 The report "encourag[ed] the development of educational programs for homosexuals to acquaint them with . . . sex-preference reversal in selected cases," and acknowledged that "some physicians may be less than objective in dealing with a professed homosexual if they harbor traditional antihomosexual biases or disapprove of the politics of ‘gay liberation.'"² The report also suggested that such physicians might miss disease manifestations common in gay and lesbian people and concluded that "an open, accepting, non-judgmental attitude . . . can be difficult for some physicians to achieve when treating a homosexual patient . . . but a sick individual—heterosexual or homosexual—deserves the best care that the psychiatric or other medical condition demands."²

At this time, a California-based doctors' group approached the AMA's national leadership seeking to form a lesbian and gay caucus. The request was denied. The group organized the American Association of Physicians for Human Rights (AAPHR) in 1982. That summer, AAPHR made the first public appearance of an organized gay and lesbian physician group in San Francisco's annual gay pride march (L. Siegel, MD, oral communication, August 1999).

Efforts to establish policy on issues of sexual orientation continued within the AMA. At the 1988 interim meeting, the MSS adopted Resolution 26 urging the Liaison Committee on Medical Education (LCME) and the Accreditation Council for Graduate Medical Education (ACGME) to amend standards to prohibit discrimination in admissions decisions because of sexual orientation. This was adopted as organization-wide policy in 1989.³ However, the LCME and ACGME have yet to modify medical schools' and residency programs' accreditation standards to protect lesbian and gay applicants.^4- 5

More contentious was the adoption of internal policy. Prior to 1993, AMA bylaws made no reference to sexual orientation in the nondiscrimination clause. In December 1989, the District of Columbia's delegation proposed amending clause B-1.50 so that membership could "not be denied or abridged on account of sex, color, creed, race, religion, ethnic origin, national origin or sexual orientation."⁶ Although the reference committee saw "no good reason not to prohibit discrimination on the basis of sexual orientation," floor debate was heated. Delegates presumed that the resolution intended "endorsement [of] an alternative lifestyle" and was "too prescriptive."⁷ The resolution failed.

The Resident Physicians Section offered similar language in December 1991, citing failure to revise the bylaws as de facto discrimination. Several delegations expressed support, but the majority opposed. One dissenter did not "want to make or give minority rights to this group of people."⁸

Finally, in June 1993, citing 7 failed resolutions on the subject, the AMA's Board of Trustees, led by chair Raymond Scalettar, issued Report A, recommending that the bylaws' nondiscrimination clause be amended to include sexual orientation (A. Novick, MD, personal communication, August 1999). Debate was no less contentious 5 years later. Delegates worried that the AMA would be seen as accepting a "deviant lifestyle" or succumbing to "outside pressure groups."⁹ Then AMA President John Clowe spoke strongly in favor: "All these individuals want . . . is some recognition without fear to appear at this House. . . . We are not condoning it. We are merely saying it is time for this House of Delegates to vote for the approval of including sexual orientation in their Bylaws." To the sounds of cheers and applause throughout the chamber, the House of Delegates approved the change in bylaws.⁹

The end of the tumultuous 5-year battle to amend the AMA bylaws prompted other organizations to make more expeditious changes.¹⁰ In October 1993, a statement from the American Academy of Pediatrics charged pediatricians to care for gay and lesbian adolescents' health concerns.¹¹ The American College of Physicians–American Society of Internal Medicine linked effective patient care to the equal treatment of physician colleagues, regardless of "race, religion, ethnicity, nationality, sex, sexual orientation, age, or disability."¹² In 1996, the American Academy of Family Physicians adopted policy supporting equal treatment of lesbian, gay, bisexual, and transgender (LGBT) physicians, patients, and their families: "By encouraging diversity in their physician workforces, physician groups and health care systems can help ensure their ability to deliver culturally competent care to all segments of their patient populations."¹³ Meanwhile, on National Coming Out Day in 1994, to obtain greater visibility as the premier national organization advocating for gay and lesbian health care, AAPHR renamed itself the Gay and Lesbian Medical Association (GLMA).¹⁴ In August 1996, GLMA broadened its focus by incorporating bisexual and transgendered patients' health concerns into its mission statement.

In successive years, the AMA amended civil and human rights policies to be inclusive of sexual orientation.^15- 16 In addition, the Council on Scientific Affairs commissioned another report, "Health Care Needs of Gay Men and Lesbians in the United States." The report, published in 1996, cited 126 sources and provided a substantial summary of pertinent literature on gay and lesbian health care.¹⁷

A new force for change emerged in 1998, when the MSS convened the first-ever meeting of a lesbian, gay, bisexual, and transgender caucus to serve as a resource for LGBT members and as a center of policy development. At the interim meeting in Honolulu, caucus members secured passage of a medical student resolution calling for organizational support of the Hate Crimes Prevention Act of 1999. Upon transmittal at that same meeting to the AMA's House of Delegates, the resolution won passage.¹⁸ Similarly, in June 1999, the caucus organized intersectional support for a resolution petitioning the LCME to adopt a standard requiring nondiscrimination clauses including sexual orientation at all medical schools. At its fall meeting this month, the LCME will consider such language and potential adoption of the new standard. Also in June 1999, the MSS and the caucus sponsored an educational program on primary care of the lesbian, gay, and bisexual patient, with similar plans for future meetings.

A recent issue of American Medical News profiled GLMA, noting its growth to 2000 members and its impact on promoting quality health care.¹⁹ Recently, as part of the cultural competence initiative, the AMA published a compendium including references to sexual orientation and citing GLMA as a contact.²⁰

Internal policy commits the AMA to being a leader in lesbian and gay health concerns. Having articulated this commitment to diversity in medicine, the AMA could now provide standards of ethics, practice, and education that would break new ground in this field. Objectives could include acting as a clearing house on cultural competency and related lesbian and gay health care issues; collaborating with GLMA and other organizations in advocacy, initiatives, and legislative efforts; and improving the organization's internal environment by appointing gay and lesbian members to committees and task forces, welcoming GLMA representatives to the House of Delegates, and increasing membership recruitment efforts targeted at lesbian and gay physicians and medical students.

Acknowledgment: The authors wish to thank Lila Valinoti, AMA Medical Student Section policy analyst, and Robert Tenuta, AMA reference archivist, for their research assistance.

Proceedings of the American Medical Association are copyrighted by the AMA and reprinted courtesy of the AMA Archives.

At present, an employee's benefits package can represent as much as 30% to 40% of value added to a base salary.¹ So-called soft benefits, such as bereavement leave, facilities use, and employee assistance programs, cost little to provide. "Hard benefits," on the other hand, are usually of considerable value; these include medical, dental, vision, and mental health coverage, prescription drugs, tuition grants, and accidental death and dismemberment and dependent life insurance.

Over the last half-century workers have received medical insurance as an employment benefit because employers were able to negotiate volume-discounted costs for expensive policies and programs. Benefits have historically been nontaxable and limited to the employee, a legal spouse, and dependent children.

Gay men and lesbians cannot obtain a civil marriage license or access employee spousal benefits. As a result, the couple must purchase a separate individual, usually more expensive insurance policy for the nonemployed partner and the partner's biological children.

In December 1988, colleagues at Albert Einstein College of Medicine informed me that their spouses received medical insurance as an employment benefit. Upon inquiry at the benefits office, I was informed that state laws did not require coverage of my life partner, so we would not be offered medical insurance. My department chair suggested asking the faculty senate to mandate coverage.

On September 13, 1989, my resolution was presented, citing Yeshiva University's "long-standing commitment to equal opportunity . . . without regard to race, religion, creed, color, natural origin, sex, age, handicap, veteran or disabled veteran status, marital status, or sexual orientation."² The resolution noted that not all people could obtain marriage licenses entitling them to certain privileges of employment. The resolution asked that the senate mandate "insurance benefits, education benefits, and housing accommodations without regard to sexual orientation for all faculty and students who share domicile and mutual responsibility for each other's welfare and basic living expenses, and who have either a marriage license or mutual power of attorney."² It passed unanimously.

Coverage was not forthcoming, however, because the health plan "treat[ed] all unmarried individuals equally and cannot differentiate between groups of unmarried individuals who may happen to cohabitate" (K. Prince, Manager of Employee Benefits of Albert Einstein College of Medicine, written communication, November 22, 1989). Upon consultation, the American Civil Liberties Union's lawyers drafted a letter citing New York City's nondiscrimination policy inclusive of sexual orientation and a state court verdict recognizing domestic partnership in rental disputes (J.D. Marks, written communication to C. Margolin, Associate General Counsel to Montefiore Medical Center, December 13, 1989).³ Legal counsel drew up a confidential settlement contract for me in early 1990. In March 1991, after other staff and faculty sought similar contracts, Montefiore Medical Center became the largest private employer to provide domestic partner health coverage, announcing it was "the fair thing to do."⁴

When I began employment at Stanford University, I and my life partner again purchased separate medical insurance. This time, I joined other staff and faculty to establish an equal benefit policy. Three hard benefits and 4 soft benefits tied to legal marriage status formed the basis of the Benefit Parity Bill.⁵ Over 240 supportive faculty members signed an open letter to the faculty senate, and the undergraduate student union and the medical school faculty senate both passed the bill.

In May 1991, the senate discussion included comments comparing domestic partner insurance coverage with tuition grant reimbursement for children one did not have,⁶ neither being a deserved benefit. Another worry was that gay men and lesbians would flock disproportionately to Stanford seeking greater benefits.⁶ The bill was sent for subcommittee review, which recommended passing the bill and reported the following:

"One imagines, for example, that a decision by Stanford 40 years ago to take the lead in eradicating discrimination against blacks, women, or Jews in admissions, hiring, memberships in sororities and fraternities, etc, would have been politically unpopular with many alumni, as well as with the larger political community. One also imagines that had Stanford taken such a leadership role, few in the Stanford community would look back on that decision now with anything but pride."⁷

Over a year after introduction, the bill passed in September 1992. Stanford's trustees voted to implement it the following February.⁸

Currently, 141 colleges and universities, 87 cities and counties, and 570 companies provide domestic partner benefits. Utilization rates run from 0.5% to 2.5%, making the costs of equal treatment minimal.¹ Often employees request benefit parity, and employers respond to stay competitive.

Gay and lesbian employees have familial responsibilities like other people and are more productive when secure and financially stable. Medical institutions are more attractive to all potential employees if benefits packages are equally-accessible. If a medical center or university bars discrimination due to marital status, equal access to employment benefits is a reasonable corollary. The American Psychological Association, American Psychiatric Association, American Medical Association, and American Medical Women's Association endorse equal treatment for all regardless of sexual orientation.^9- 12

Despite evidence of discrimination against lesbian physicians, the prevalence of harassment due to sexual orientation is not accurately known. The Women Physicians' Health Study (WPHS) queried the prevalence of harassment among lesbian and heterosexual physicians during premedical school education, medical school, graduate medical education, and medical practice.

WPHS is a cross-sectional survey of women physicians aged 30 to 70 years who were not in residency training and resided in the United States in 1993.¹ A probability sample of 10,000 women physicians was extracted from the AMA Physician Masterfile, stratified by decade of graduation (1950 to 1989), with oversampling of earlier decades. Of this sample, an estimated 23% were ineligible to participate. Among eligible respondents, 59% (4501) completed a 716-item, self-administered questionnaire covering health status, history, and behavior, as well as aspects of medical practice and demographics. A more complete description of the survey methods can be found elsewhere.²

Sexual orientation was assessed by 2 items querying self-identification and sexual behavior: "1. Do you now self-identify as: heterosexual; bisexual; lesbian/gay/homosexual; other; 2. Are you now sexually active with: men/women/both/neither." A woman was defined as lesbian if she either self-identified as lesbian in question 1 or reported current sex with women in question 2 (115 respondents). Heterosexuals were defined as those who self-identified as heterosexual and who did not currently have sex with women or who responded "other" or "no response" but reported current sex with men (4177 respondents). "The remaining 209 respondents included bisexual women (excluding those who identify as bisexual but report current sex with women, n = 41), "other" (n = 24), and not classifiable (n = 144)."

Harassment was assessed by the question "Have you ever been harassed in a medical setting? (ie, received unwanted physical or verbal attention, propositions, hostilities or threats)." Types of harassment (queried as gender-based but nonsexual, sexual, lifestyle-based, and ethnically based) and time of harassment occurrence (before medical school, during medical school, in training, or in practice) were cross-classified. Chi-square tests were used to compare lesbians and heterosexuals in the prevalence of types of harassment; SUDAAN, a sample survey software package, was used for all analyses. Although the word lifestyle was used in the 1993-1994 survey, orientation is used in this report, because sexual orientation is no longer described as a lifestyle.³

Lesbians and heterosexuals did not differ in age, but lesbians were significantly less likely to have ever been married or pregnant, to have had a live birth, to have children, or to be part of a current couple or marriage, or to be politically conservative (P<.01). Lesbian respondents were more likely to be white/Caucasian and less likely to be Asian. Lesbians and heterosexuals were equally likely to be in a primary care specialty.

Lesbians were about 4 times more likely than heterosexual physicians to report ever having experienced sexual orientation–based harassment in a medical setting (41% for lesbians vs 10% for heterosexuals, P< .0001) using the identity definition; results were similar (36% vs 10%, P<.0001) using the identity/behavior definition. However, lesbian and heterosexual physicians reported similar (P>.01) lifetime prevalences of gender harassment (approximately 50%) and sexual harassment (approximately 40%), and reported similar prevalences of gender and sexual harassment in any of the 4 medical settings.⁷

Lesbians, compared with heterosexual physicians, are significantly more likely to report experiencing sexual orientation–based harassment during graduate medical education (18.2% vs 3.6%, P< .005) and during medical practice (18.5% vs 5.4%, P<.01). Lesbian physicians are significantly more likely than heterosexual female physicians to report sexual orientation harassment in any work setting after medical school (32.6% vs 7.7%, P<.0001).

Our findings indicate that lesbian physicians are about 4 times more likely than heterosexual women physicians to report harassment related to sexual orientation in any medical setting, but primarily during training and medical practice. Also, lesbian and heterosexual physicians did not significantly differ in prevalence of reported gender or sexual harassment, suggesting that higher sexual orientation–based harassment prevalence among lesbians cannot be explained by assuming that lesbians overreport harassment.

This study, based on a nationwide probability sample of women physicians, confirms findings from less scientific reports based on case studies, volunteer opinions, and nonprobability surveys that orientation harassment is experienced at a fairly high rate by lesbian physicians. A survey of San Diego County physicians, in 1982, revealed that 23% of respondents scored in the homophobic range on a scale of heterosexual attitudes toward homosexuality.⁴ Specifically, 30% were opposed to admitting highly qualified gay or lesbian applicants to medical school, and 45% and 39% opposed admittance to residency training in pediatrics and psychiatry, respectively. The most homophobic specialties were general surgery, obstetrics/gynecology, and general/family practice, in which 30% of physicians scored in the homophobic range. Similarly, a 1994 study found that 25% of family practice program directors "might" or "most certainly would" rank gay applicants lower than heterosexuals in the residency match.⁵ Also in 1994, the Gay and Lesbian Medical Association published findings that 17% of its physician members felt they had been refused privileges or denied employment or promotion based on sexual orientation, with 34% reporting orientation-related verbal harassment by medical colleagues.⁶

We have previously established that harassed women physicians are less satisfied with their careers, feel less in control of their work environments, and are more likely to have histories of depression and suicide attempts.⁷ At present, many universities, medical schools, and medical centers have nondiscrimination policies that include sexual orientation. Some also provide training to familiarize heterosexuals with gay and lesbian health issues and problems associated with homophobia. Such protection and training may reduce the high prevalence of harassment experienced by lesbian physicians.

Funding/Support: We acknowledge the support of the Lesbian Health Fund and the American Medical Association Foundation.

The US government's Report of the Secretary's Task Force on Youth Suicide, which appeared in 1989, sparked a controversy that continues to the present day. In his chapter on gay and lesbian youth suicide, Gibson projected that "gay youth are 2 to 3 times more likely to attempt suicide than other young people. They may comprise up to 30% of completed youth suicides annually."¹ Some experts rejected the conclusions as being drawn from biased samples.² Considerable work since then has addressed the putative association between sexual orientation and suicide.

The problem of suicide first surfaced as an incidental finding in pioneering research on homosexuality that identified a high prevalence of such attempts among young men.³ Two of the earliest studies of gay youths revealed that as many as 1 in 3 had attempted suicide.^4- 5 The next generation of research specifically studied suicidality and sexual orientation in convenience samples.^3,6- 14 Ten such studies found consistently high rates of attempts among homosexual youths—in the range of 20% to 42%. Six of the studies involved both women and men.^{7,9- 11,13- 14} Three found women's attempt rates to be as high as or higher than those of boys.^9,13- 14

Of the 10 studies, 6 explored risks for suicide by comparing attemptors and nonattemptors. They found that suicide attempts were neither universal nor attributable to homosexuality per se, but they were significantly associated with gender nonconformity,³ early awareness of homosexuality,^3,8,11,14 stress,¹² violence,¹⁴ lack of support,^8,11,14 school drop-out,¹² family problems,⁸ acquaintances' suicide attempts,¹² homelessness,¹² and substance abuse³ or other psychiatric symptoms.^6,11,14

While providing valuable descriptive information, the prior studies were limited by potential sample biases, the absence of comparison groups, or both—problems recently surmounted by controlled, population-based surveys. Five of 6 such studies involved representative samples of US secondary school students^15- 19 and one, a community sample of young men from Calgary, Canada.²⁰ All found higher rates of attempted suicide among homosexual youths compared to their heterosexual peers. Surveys large enough to examine sex differences among Minnesota¹⁵ and Massachusetts students¹⁹ found a significant association between homosexuality and suicidality in males only. Suicide risk factors such as gender nonconformity may be particularly detrimental to boys.^15,19

A unifying explanation for the prevalence of suicidality among homosexual youths remains to be determined, as does the extent to which attempts end in death. Two psychological autopsy studies^21- 22 have tried to unearth the sexual orientation of suicide victims. One found that 11% (13/119) of the young men who died from suicide in San Diego from 1981 to 1983 were known to be gay, but of the women, none were known to be lesbians.²¹ A second New York suicide study involved adolescent suicides from 1984 to 1986 and found that 3.2% (3/95) of male suicides and none of the living controls were found to have had homosexual experiences.²² Studies of this type can be limited in their ability to ascertain sexual behavior and orientation posthumously.

Although the understanding of gay, lesbian, and bisexual youth suicide is increasing, many questions remain regarding sex and ethnic differences, predisposing social and psychiatric conditions, protective factors, and constructive interventions. Future population-based surveys should routinely inquire about sexual orientation to retest prior findings in diverse settings. Prospective, longitudinal studies are needed to examine the evolving risk of suicide across the lifespan of homosexual persons. As we continue to assess the problem, existing data are sufficiently compelling to teach clinicians about the association between suicidality and sexual orientation and to plan preventive interventions.

Funding/Support: This work was supported in part by the University of Minnesota Adolescent Health Training Program, grant 5-T71-MC-00006-22, Maternal and Child Health Bureau, Health Resources and Services Administration.

Visit the MSJAMA Web site at http://www.ama-assn.org/msjama and the new Virtual Mentor.