I. A. Definitions and Scope *

I. B. Stigmatization of LGBT Populations in the U. S. *

I. C. Health of Lesbian, Gay, Bisexual, and Transgender Populations *

II. Threshold Issues *

II. A. Public Health Infrastructure *

II. B. Access to Quality Health Services *

II. C. Health Communication *

II. D. Educational and Community-Based Programs *

III. Lesbian, Gay and Bisexual Health Concerns *

III. A. Cancer *

III. B. Family Planning *

III. C. HIV/AIDS *

III. D. Immunization and Infectious Diseases *

III. E. Mental Health and Mental Disorders *

III. F. Sexually Transmitted Diseases *

III. G. Substance Abuse *

III. H. Tobacco Use *

III. I. Violence and Sexual Assault *

IV. Transgender Health Concerns *

IV. A. Overview and Definitions *

IV. B. Barriers to Care *

IV. C. Special Populations *

IV. D. Medical and Research Institutions *

IV. E. Standards of Care *

IV. F. Mental Health Needs *

IV. G. Genital Reassignment Surgery *

IV H. HIV/AIDS *

V. Methodological Challenges to Studying Lesbian, Gay, Bisexual and Transgender Health *

V. A. Overview *

V. B. Defining the Populations *

V. C. Measuring the Populations *

V. D. Sampling Rare Populations *

V. E. Sampling Sensitive Topics *

V. F. Summary *

VI. References *

Because of negative attitudes prevalent in the U.S. public as well as among physicians and other medical staff, LGBT individuals are subject to discrimination and bias in medical encounters. Moreover, they are likely to receive substandard care, or remain silent about important health issues they fear may lead to stigmatization (Schatz & O’Hanlan, 1994; Bradford & Ryan, 1988; Bocktinet, et. al., 1998). Bias from health care professionals ¾ and perception of such bias ¾ have been identified as personal and cultural barriers to care, leading to reduction in help-seeking and quality of care (Millman, 1993). In addition, stereotyping and lack of education may lead health care providers to ignore known special preventive care and treatment needs of LGBT people (e.g., provision of pap smears to lesbians, pain management after genital reassignment surgery, examination for infections of the anal canal, and others). Medical forms and the format of medical intake and history are often insensitive to the experience of LGBT patients, and likely to discourage disclosure of sexual orientation and behavior.

In mental health care, stigma, lack of cultural sensitivity, and unconscious and conscious reluctance to address sexuality may all hamper effectiveness of care. An extreme example is found in "reparative" therapy that seeks to reverse sexual orientation or gender identification, an approach that may lead to increased self hatred and mental health problems (Ryan, Bradford, & Honnold, 1999). Treatment modalities that rely on group therapies and support groups are also vulnerable to the effects of discrimination, with participants often forming a justifiable fear that full disclosure of personal details may adversely affect their standing in the group or health care setting. LGBT people report discriminatory treatment following disclosure of sexual orientation in paramedical and auxiliary care settings, including nursing homes, domestic violence centers, senior centers and others (Wolfe, in press).

Barriers to care for LGBT people include systemic bias in health insurance and public entitlements, which routinely fail to cover gay and lesbian partners or to provide reimbursement for procedures of particular relevance to LGBT populations (e.g., fertility services to lesbians, surgical procedures required by transsexuals). Obstacles to LGBT care are likely to increase as greater numbers of employers move toward self-insurance and as health maintenance organizations (HMOs) require more detailed reports for ongoing mental health care. These and other trends will increase LGBT individuals’ fears of breaches in confidentiality and consequent stigmatization.

 

This report covers a wide variety of health concerns of particular relevance for LGBT populations, organized by health topic. It is also important to consider these health outcomes in the context of their social origins. The following table conceptualizes health outcomes and their putative relationship to social/behavioral factors specific to LGBT populations, listing social/behavioral factors (across the top) and health outcomes specific to each (vertically below each factor). While not an exhaustive list of either, the table may serve as a useful heuristic in considering the relationship of social/behavioral factors and LGBT health concerns.

While knowledge of and standards of care for particular conditions prevalent among LGBT populations vary widely, several threshold areas impact LGBT health delivery generally and provide important opportunities to enhance delivery of care to LGBT populations. These topics are summarized in Section II, and include:

• public health infrastructure;
• access to quality health services;
• health communication; and
• educational and community-based programs.

Although there are many common issues affecting LGBT communities, transgender concerns present unique health challenges that merit separate discussion. Subject-specific summaries of leading lesbian, gay, and bisexual health concerns (e.g., cancer, substance abuse, etc.) are found in section III of this document. Section IV provides an overview of some of the complex issues related to transgender individuals.

Central to study of LGBT health concerns are the methodological challenges posed by studies of these populations. Relevant issues include definition and measurement of critically unexamined and socially constructed categories, as well as challenges of sampling rare and hidden populations concerning sensitive topics. These methodological challenges are examined in Section V.

Many topics of concern to the health of LGBT individuals are not addressed here. While this report summarizes existing research findings, clinical and public health research for these populations have been scarce (Bradford & White, 2000). In addition, there is currently no public health infrastructure for funding and supporting research on the health of LGBT communities. Coordination of research into LGBT health, and systemic reform and education toward elimination of anti-gay bias and insensitivities, remains imperative.

While this report is brief, its findings will prove useful if it engenders increased dialogue, understanding, concern, and support for research, education, and training with respect to LGBT health concerns. These are among the steps necessary for increasing knowledge about LGBT health and LGBT access to health services; for improving the health of members of lesbian, gay, bisexual, and transgender communities; for addressing health disparities in the U.S.; and for ensuring the health of all U.S. residents.

Efforts to research and address the health needs of LGBT populations are hampered by a lack of infrastructure to support and direct funded initiatives. Currently there is no identified agency within the Public Health Service with responsibility to oversee and/or coordinate such initiatives. Various agencies have funded competitive research and policy studies, but these efforts have been scattered and without central, overarching guidance relevant to population health. For gay and bisexual men, competitive research grants have been funded primarily in areas related to HIV/AIDS. For lesbians, several grants have been awarded in the areas of substance abuse, health care, and mental health (Solarz, 1999). While useful, these uncoordinated studies do not comprise an organized program of population-based research, nor is it easy to understand how they may mature into such a program without the intentional development of support structures. Research and public health interventions targeted to transgender individuals have lagged even more seriously.

Recent reports on LGBT health, most notably the Institute of Medicine’s landmark report on lesbian health, contain a number of recommendations that will be difficult to implement without significant infrastructure development (Solarz, 1999). Political constraints may limit the feasibility of creating an LGBT-dedicated research organization within the federal bureaucracy. However, alternatives such as cooperative planning and funding of critical population-based research and initiatives may be feasible. The IOM study on lesbian health, supported by funding from two agencies in the federal public health infrastructure – the Office of Research on Women’s Health in the Department of Health and Human Services (DHHS) and the Office of Women’s Health at the Centers for Disease Control and Prevention (CDC) – represents one example of such a collaboration.

Other promising cooperative efforts have followed publication of the IOM report. In September 1999, the National Institute of Mental Health (NIMH) collaborated with the American Psychological Association to present and host a two-day workshop on current research regarding LGBT populations. A current collaboration of DHHS with the Gay and Lesbian Medical Association/Lesbian Health Fund (GLMA/LHF) to implement a March 2000 scientific meeting as follow-up to the IOM report is another related endeavor. Cooperative efforts such as these can help fill gaps in the infrastructure, and can provide a foundation on which to create an enduring structure with appropriate sharing of responsibility among public and private organizations.

II. B. Access to Quality Health Services9;

LGBT individuals face financial, structural, personal, and cultural barriers as they attempt to access competent, sensitive health care services (Millman, 1993). These barriers and anxiety about facing them prevent LGBT individuals from receiving the screening and prevention services they need, and cause delays in receiving care for acute conditions.

Provider attitudes, communication difficulties, and systemic obstacles pose major impediments for sexual minorities to access care. LGBT persons, despite generally higher educational levels, have a lower socioeconomic status than their heterosexual counterparts. Many are self-employed, work as artists or craftspersons, or work part-time; consequently, many have few or no health insurance benefits (Badgett, 1998; Bradford, 1988).

An analysis of data from the National Lesbian Health Care Survey (NLHCS) (Bradford, in press) suggests that subgroups of lesbians may be at particularly high risk of negative health consequences due to lack of insurance coverage. Within the NLHCS sample, lack of health insurance was significantly correlated with being younger, unemployed, in school, of lower income, and African-American. Mental health issues were more prominent among uninsured respondents, many of whom reported significantly higher levels of anxiety and suicide ideation. Uninsured respondents are also more likely to have experienced physical and/or sexual abuse and reported much greater concern about sometimes feeling unable to meet their routine responsibilities. Certain physical health conditions were also more prominent, including ulcers and other intestinal disorders, substance abuse, and eating disorders. There was a statistically significant correlation for this sample between not having health insurance and believing that being lesbian affected their access to health care.

Both gay men and lesbians in committed relationships are at a disadvantage compared to married heterosexuals because many insurance companies and employers deny spousal benefits to unmarried partners (Denenberg, 1995; Stevens, 1995). Systemic bias in favor of heterosexuals is also found in regulations allowing one member of a married heterosexual couple to retain a jointly owned house without jeopardizing the other’s right to Medicaid coverage. Moreover, married heterosexuals receive a spouse’s Social Security payments following his or her death. While some 30 states have passed constitutional amendments barring recognition of gay marriage even if found legal by their courts, a recent ruling in Vermont provides a promising model. The State’s highest court has instructed legislators to recognize gay marriage or to structure domestic partnership regulations affording committed gay and lesbian couples benefits equivalent to those of heterosexual spouses. A number of local municipalities and private corporations have also widened insurance coverage for their employees to include unmarrieddomestic partners, both homosexual and heterosexual

LGBT populations may also find it difficult to access other publicly-supported programs. While many localities offer critical low cost or free health care and screening to women who are seeking birth control, lesbians who do not need birth control find it hard to locate affordable health care services. Men, already less likely to seek out such services than women, may be further impeded by reluctance to reveal their sexual practices or fear of homophobia in medical settings.

Additionally, same-sex partners are denied rights granted to married heterosexuals in hospitals and clinics. Unless a gay or lesbian couple has signed legal papers (e.g., durable power of attorney, health proxy, etc.) authorizing mutual medical decision making, blood relatives – including those who know less about the patient’s ethical, medical or religious preferences – can override decisions by a homosexual partner. Some health care settings also limit visits or participation in medical consultations to legally recognized spouses or blood relatives, a practice particularly damaging to the many LGBT people who prioritize families of choice over families of origin (O’Hanlan, Cabaj, et. al., 1997).

Even those LGBT individuals who have individual private insurance may be reluctant to use it to access care. The vast majority of U.S. employers who have more than 1,000 employees now self-insure, which means they have access to employee health care records and claims (Wolfe, in press). For some LGBT persons, disclosure of information about sexual orientation or gender identity would be an unacceptable consequence of seeking care. Similar confidentiality concerns may impede LGBT individuals from seeking counseling or support, or care for stigmatized conditions such as HIV, through HMOs which often require detailed justifications to ensure continuation of benefits.

In response to poor access to the health care system, during the past two decades LGBT communities have begun building elements of their own system. Systemic reform and the elimination of anti-gay bias in health care and social service settings remain essential goals for the health of LGBT communities. These goals are critical both to address medical concerns with unique effects for LGBT populations ¾ detailed in the pages that follow ¾ and to address general medical concerns shared with Americans at large.

II. C. Health Communication

In addition to difficulties in accessing health care, lesbian, gay, bisexual, and transgender individuals face significant obstacles in communication with health care providers. First and most challenging are negative attitudes toward homosexual, bisexual, and transgender persons held by many providers. Forty percent of physicians in one study were sometimes or often uncomfortable providing care to lesbian or gay patients (Matthews, Booth, and Turner, 1986). In a nonrandom survey of members of the Gay and Lesbian Medical Association (GLMA), 67% of respondents believed they had seen gay or lesbian patients receiving "substandard" care because of their sexual orientation (Schatz & O’Hanlan, 1994). Many lesbians and gay men have reported that their doctors are not sensitive to or knowledgeable about their particular health risks and needs, and do not disclose pertinent information about treatments or prevention (Smith, Johnson, & Guenther, 1985; Trippet & Bain, 1992; Schatz & O’Hanlan, 1994).

Whether patients disclose their sexuality and sexual practices to providers may depend in part on where patients are in their coming out process. For example, gay and lesbian adolescents ¾ who often lack structural supports such as financial independence and social networks that can sustain older gay men and lesbians (Allen, Glicken, Beach, et. al., 1998; D’Augelli & Hershberger, 1993; Newman & Muzzonigro, 1993) ¾ are likely to delay disclosure of sexuality to clinicians. Homosexually active men and women who identify as heterosexual, or those who are at the early stages of the distinct developmental processes theorists have associated with coming out (Brady & Busse, 1994) may also choose not to disclose their sexual histories to a health care provider.

Still, even individuals who enjoy significant social support as openly gay men and lesbians may find it difficult or imprudent to reveal their sexuality in a doctor’s office. Intake forms covering sexual history rarely include the option for providing information on same-gender sexual partners. Physicians and researchers routinely ask heterosexually-biased questions, such as: "Are you married, single, widowed or divorced?" or "What kind of birth control do you use?" Consequently, disclosure of sexual orientation in a health care setting remains infrequent for the majority of gay men and lesbians (Robertson, 1998; White & Dull, 1997; Allen, et. al., 1998; Cochoran & Mays, 1988; Siegel, Krauss & Karus, 1994). In a survey of lesbians in Michigan, 61% felt unable to disclose their sexual orientation to their providers (Bybee, 1990). Some 9% of respondents reported that their health providers had not allowed their female partners to stay with them during treatment or see them in a treatment facility; 9% also said that providers had not included their partners in discussion about treatment (Bybee, 1990).

The medical educational system has failed to educate providers and researchers regarding the unique aspects of lesbian and gay health (Wallik, 1992), including examination techniques, taking of patient histories, and preventive recommendations. Although homosexuality has been removed from the list of diagnoses in the diagnostic manual of the American Psychiatric Association, the relationship between homosexuality and sickness has proved more enduring in the minds of many providers. A variety of studies describe provider hostility (Hayward & Weissfeld, 1993; Gerber, Maguire, Bleeker, et. al., 1991) or instances of gay men or lesbians being described as "deserving" of illness or unworthy of treatment (Schatz & O’Hanlan, 1994). Tellingly, stigma can be most pronounced in those instances where sensitive treatment is needed. A number of respondents to the GLMA survey, for example, said they had seen doctors performing "rough" or "violent" digital rectal exams on patients after discovering that they were gay (Schatz & O’Hanlan, 1994).

Many clinicians sympathetic to gay men or lesbians may lack a repertoire of questions about social and sexual history appropriate to homosexuals, or be unaware of why they might be necessary. Among physicians interviewed for a cancer screening project conducted by the Mautner Project for Lesbians with Cancer, approximately half stated that they assumed lesbians were in their practices, but did not see any reason to address this in a direct way (Bradford & Dye, unpublished manuscript). These providers expressed an eagerness to learn more about the needs of lesbians and stated emphatically that they would make changes when they had information about what steps would be appropriate (Bradford & Dye).

Physicians are not alone among health providers in facing difficulties in communication with gay and lesbian patients. In a random sample survey of Virginia mental health providers, respondents acknowledged having lesbians in their practices, yet had little or no training about the special needs of these clients (Ryan, Bradford, & Honnold, 1999). Some mental health practitioners continue to practice "reparative" therapy for homosexuality (Berger, 1994), to use inadequate or inappropriate definitions for lesbians and gay men, and/or state that they do not think the concerns of gay or lesbian clients are different from those of heterosexuals (Robertson, 1998; Ryan, Bradford, & Honnold, 1999). Findings were similar in a California study of physicians dealing with lesbians and gay men (Mathews, et. al., 1986).

Barriers to communication about the needs and realities of gay and lesbian life are manifested at the systemic as well as individual level. In hospitals, emergency room

Transfer interrupted!

Finally, gay men and lesbians themselves, when faced with an uncomfortable interaction with a clinician, may lack the skills or self-efficacy to defend against negative experiences. They may feel unable to change physicians’ conduct, to resolve an uncomfortable situation, or to speak openly with a clinician about their discomfort. This may stem from a history of discrimination and the power imbalance traditionally experienced between clinician and patient, and/or from past memories of difficult experiences.

Disclosure of sexual orientation in the health care setting is crucial to the provision of appropriate, sensitive, and individualized care. Failure to establish rapport and communication between physicians and patients is associated with decreased levels of adherence to physician advice and treatment plans, and decreased rates of satisfaction (Inui & Carter, 1989). Additionally, clinicians unaware of their patients’ sexual orientation may fail to accurately diagnose, treat, or recommend appropriate preventive measures for a range of conditions.

While more research is needed on the effects of communication related to sexual orientation and medical care, small surveys suggest that successful communication and ease of sexual orientation disclosure may positively affect health risks and screenings. In a study of lesbians in Oregon (White & Dull, 1997), 90% disclosed their sexual orientation to providers, and of these, 92% raised the issue themselves (White & Dull, 1998). Communication style of the provider was rated by respondents as the most important characteristic in determining ease of discussion about difficult issues. The lesbians who disclosed their sexual orientation were more likely to seek health and preventive care, to have a Pap test, to be non-smokers, and to report comfort in communication with providers. By contrast, difficulty communicating with the primary care provider was associated with delay in seeking health care (White & Dull, 1998).

Lesbian, gay, bisexual, and transgender communities and their allies began to promote LGBT health concerns in the U.S. in the 1950s when educational, health-care, and other government and private systems proved inadequate. Organizing began with efforts to have homosexuality declassified as a medical illness. These efforts extended through the 1960s and intensified in the 1970s, with successful advocacy to remove homosexuality from the Diagnostic and Statistical Manual of Mental Disorders of the American Psychiatric Association (DSM) (Bayer, 1981). In related early efforts, LGBT communities created a variety of professional and volunteer health care initiatives to offer non-judgmental treatment and education about sexually transmitted diseases (STDs) and mental health issues related to coming out and stigmatization. Lesbian health initiatives included the creation of "lesbian health nights" at women’s health clinics (Plumb, 1998). Gay counseling sessions, peer education groups, and STD screenings at ordinarily non-gay community health facilities were also organized. Then, as the gay community faced the HIV epidemic, work ensued to expand the focus of mainstream organizations, and to create new and powerful lesbian and gay-focused health education, treatment, and advocacy organizations (Deyton & Lear, 1988; Andriote, 1999).

Today, a variety of LGBT organizations – some of which are now celebrating their 25^th anniversary – deal in a comprehensive manner with HIV/AIDS, mental health, cancer, violence, and other issues. Clinics and volunteer initiatives providing community-based primary health care services directly to LGBT communities are found throughout the U.S. Among the largest and most notable of these are the Callen-Lorde Community Health Center in New York, the Whitman-Walker Clinic in Washington D.C., and the Howard Brown Memorial Clinic in Chicago, whose services include: screening, testing and care for HIV, breast cancer, and STDs; mental health services; family planning services, including artificial insemination and adoption; and support services related to violence and coming out.

Lesbian and gay community centers, of which there are over 100 in 33 states, may provide similar or additional services to those offered by health care clinics. These community centers include the modest Up the Stairs Community Center in Fort Wayne, Indiana and the Panhandle Gay & Lesbian Support Services in Scottsbluff, Nebraska, to the much larger centers serving the gay and lesbian communities of Los Angeles, Chicago, and New York. Many of these offer counseling and support for people in crisis, youth, the elderly, people living with HIV and AIDS, people struggling with substance abuse, and survivors of anti-gay violence. For example, the Los Angeles Gay and Lesbian Center ¾ founded in 1971 as the nation’s first lesbian and gay community services center and now the nation’s largest – offers primary care counseling; addiction recovery groups; testing and treatment for sexually transmitted diseases; and shelter for homeless youth (Burns & Rofes, 1988). Others, such as the New York Lesbian and Gay Community Services Center, serve as focal points for hundreds of support groups, and host cultural events and political advocacy efforts centered around LGBT health issues. New York’s community services center is home to the nationally recognized Gender Identity Project, providing counseling, advocacy, social support and training for transgender individuals and their providers. Many community centers provide a "safe space" to diminish the sense of isolation and self-judgement that are among the particular stresses of minority sexual orientation.

Not surprisingly, in the 1980s and 1990s, the HIV/AIDS epidemic forced a restructuring of existing educational and community-based programs in LGBT communities, and the creation of new LGBT health-focused organizations (Altman, 1994; Drucker, 1994; Jonsen & Stryker, 1993; Van Vugt, 1994). The first of these new organizations and prototype for many others was Gay Men’s Health Crisis (GMHC), founded in New York in 1982. Formed by a group of gay men before there was a name for AIDS, GMHC was a community effort to educate and care for itself as a growing number of its members became ill and died (Reinfeld, 1994). GMHC and many other AIDS organizations have grown from all-volunteer efforts to professionally-staffed, non-profit organizations that provide services such as support and advocacy, primary and secondary risk reduction, and community education. Other leading AIDS organizations include AIDS Project Los Angles, San Francisco AIDS Foundation, and AIDS Action Committee in Boston. Gay men and lesbians have also played prominent roles in the organizations formed to advocate for particular needs and concerns of people of color with HIV, including for example the National Minority AIDS Council, the National Latino/a Lesbian and Gay Organization, the National Native American AIDS Prevention Center, the Latino Commission on AIDS, the Black Leadership Commission on AIDS, Us Helping Us, African American AIDS Support Services and Survival Institute, the Minority AIDS Project, Bienestar, the Asian and Pacific Islander Wellness Center, the Asian Pacific Islander AIDS Intervention Team, and the Asian and Pacific Islander Coalition on HIV/AIDS.

HIV/AIDS strengthened the gay health movement’s emphasis on patient self-determination, and broadened its focus to include the previously ignored arena of clinical trials and medical research. The Community Consortium (CC) in San Francisco and the Community Research Initiative (CRI) of the People with AIDS Coalition in New York were created in the 1980s to increase community involvement in the testing of HIV drugs (Andriote, 1999). The AIDS Coalition to Unleash Power (ACT-UP), with chapters across the country, focused its early efforts on increased scrutiny of governmental agencies and privately-held companies perceived to be impeding the development of or access to potentially lifesaving AIDS medications. While tactics differed, CRI and CC helped establish a network of community clinical trials sites, and ACT-UP used public education and civil disobedience to achieve greater scientific accountability and patient involvement in research. Advocates for a range of illnesses have since adopted similar approaches and goals.

Efforts to foster health as well as combat illness have expanded to address the needs of LGBT individuals across the lifespan. For example, the Gay, Lesbian, and Straight Education Network (GLSEN) works to create a safe learning environment for lesbian and gay students in K-12 schools. GLSEN currently has a network of 85 chapters whose members work to educate school officials about non-discrimination policies. They also train teachers to prevent anti-gay attacks and create non-hostile environments. Other organizations working to protect and advocate for LGBT youth are the Hetrick-Martin Institute in New York City and Horizons Community Services Center in Chicago (Ryan & Futterman, 1998). For older gay men and lesbians, a number of organizations – many of which participate in a network organized by Senior Action in a Gay Environment (SAGE) – now provide support groups, social support, provider education, and advocacy.

LGBT professional organizations, or committees and working groups within larger professional organizations, serve as important resources for health care provision and social support efforts. Both the American Psychological Association and the American Society on Aging have committees or working groups focused on gay and lesbian concerns. LGBT professional organizations include the Gay and Lesbian Medical Association (GLMA), the Association for Gay, Lesbian, and Bisexual Issues in Counseling, and the National Association of Lesbian and Gay Addiction Professionals. The mission of GLMA, for example, is: "to combat homophobia within the medical profession and in society at large, to promote quality health care for LGBT and HIV-positive people, to foster a professional climate in which our diverse members can achieve their full potential, and to support members challenged by discrimination on the basis of sexual orientation (GLMA, 1999)."

In some locations, governmental entities have been created to promote the health of LGBT people beyond concerns of HIV/AIDS. For example, New York City established the Office of Gay and Lesbian Health Concerns (OGLHC) in 1983 in response to the AIDS crisis and redefined it in 1985 to address all lesbian and gay health concerns. The OGLHC has addressed the lack of health-care services for lesbian and bisexual women, anti-gay and lesbian violence, gay and lesbian suicide, and chemical dependency and substance abuse among gays and lesbians. Similarly, in San Francisco a Coordinator of Lesbian and Gay Health Services was created in the early 1980s (Vachon, 1988).

Other governmental agencies extended their reach to LGBT people by collaborating with existing community organizations. In 1996 the Massachusetts Department of Public Health, in conjunction with the Justice Resource Institute, the Medical Foundation, and the JSI Research and Training Institute, created the Gay, Lesbian, Bisexual and Transgender Health Access Project (GLBTHAP). The project works "to strengthen the Massachusetts Department of Public Health’s ability to foster the development of comprehensive, culturally appropriate health promotion policies and health care services for gay, lesbian, bisexual and transgender people through a variety of venues including community awareness, policy development, advocacy, direct service and prevention strategies (GLBTHAP, 1999)."

In conclusion, while some governmental agencies, professional organizations and mainstream health care entities have begun to address and respond to the concerns of lesbian, gay male, bisexual, and transgender individuals, LGBT people still largely depend upon self-created community-based and professional organizations to address their special health care needs. In addition, no federal infrastructure exists to facilitate the creation of LGBT health initiatives or research.

While definitive studies are lacking, preliminary data lends credence to the suggestion that gay men and lesbians are at increased risk for certain cancers, and that more research is needed to understand the natural history of disease in LGB populations. Innovative studies in psychoneuroimmunology further suggest a direct association between psychological phenomena, reduced immunity, and tumor growth; thus, gay men and lesbians who do not disclose their sexual orientation may be at increased risk for melanoma or other cancers due to psychogenic suppression of the immune response (Cole, Kemeny, Taylor, et. al., 1996).

Breast cancer is probably the most researched topic in lesbian health. Several investigators have hypothesized that lesbians are at higher risk for breast cancer than are heterosexual women due to higher rates of risk factors such as obesity, alcohol consumption, nulliparity, and lower rates of breast cancer screening (Haynes, 1994; Denenberg, 1994). While definitive studies in this area have yet to be completed, data on prevalence of each of the risk factors confirms the plausibility of this hypothesis (Dibble, et. al., 1997; Valanis, Bowen, et. al., in press). Lesbians also receive less frequent gynecologic care than do heterosexual women (Robertson, et. al., 1981) and therefore might also be at greater risk for mortality and morbidity from gynecological cancers. Both of these risks are likely compounded by the difficulties many lesbians experience in communicating with or receiving standard clinical care from physicians and health care systems (Trippet, et. al., 1992). Little is known about prevalence and incidence of other cancers among lesbians.

Among gay men, high rates of Kaposi’s sarcoma (KS) were some of the first described indicators of AIDS (CDC, 1981). Estimated risk for KS among gay and bisexual men was thousands of times higher than in the general population prior to approval of highly active antiretroviral therapies (Koblin, et. al., 1996). This was a by-product both of HIV’s weakening of the immune system and of the sexually transmitted nature of the herpes virus (HHV-8) now thought to cause the cancer (Martin et. al., 1998). The risk for AIDS-related non-Hodgkin’s lymphoma is also elevated among gay men, although at lower magnitudes (Koblin, et. al., 1996). Between 1973 and 1987 non-Hodgkin’s lymphoma incidence increased ten-fold and Kaposi’s sarcoma incidence increased over 5,000-fold in single San Francisco men ages 20 to 49 (Rabkin, Biggar, & Horn, 1991). While highly active antiretroviral therapies have significantly reduced KS rates (Buchbinder et. al., 1999), non-Hodgkins lymphoma remains among the AIDS-related malignancies that continue to occur at sharply higher rates among HIV-positive individuals. An increase in the incidence of Hodgkin’s disease has also been observed, but its association with HIV/AIDS is equivocal and the disease is not included in the AIDS case definition (Koblin, et. al., 1996).

In a cohort design, using studies of records in New York and California cancer registries and the National Death Index, Koblin and colleagues (1996) found gay and bisexual men to be at excess risk for anal cancer (standardized incidence ratio [SIR] = 24.2), non-Hodgkin’s lymphoma (SIR = 12.7) and Hodgkin’s disease (SIR = 2.5). These accounted for an increased risk for all cancers in this population (SIR = 1.6). The authors found no difference in the incidence of cancers in any other site, including lip, oral cavity and pharynx; digestive system and peritoneum; respiratory; bone and connective tissues; skin; genitourinary; multiple myeloma; leukemia; or other and unspecified sites. While Koblin and colleagues determined that the increase in risk for both non-Hodgkin’s lymphoma and Hodgkin’s disease was related to increased incidence of HIV/AIDS among homosexual men, they found the increased risk for anal cancer was unrelated to HIV/AIDS.

Evidence increasingly suggests that gay men are at increased risk for anal cancer. Higher risk for anal cancer among gay and bisexual men was demonstrated by Daling et. al. (1987) in a case-control study of anal cancer. Daling et. al. (1987) found that history of anal intercourse was associated with anal cancer (RR = 33.1) in men. History of STDs was also a risk factor for the disease.

Research suggests that risk factors for the excess incidence of anal cancer among homosexual men included increased rates of Human Papillomavirus (HPV) and anal squamous intraepithelial lesions (ASIL), both putative anal cancer precursors (Beckman, et. al., 1989; Breese, Judson, Penley & Douglas, 1995; Daling, et. al., 1987; Melbye, Palefsky, Gonzales, et. al., 1990; Palefsky, Holly, Ralston, & Jay, 1998). HPV has been found in both HIV-positive and HIV-negative men, but prevalence of infection is higher among HIV-positive men, with highest prevalence associated with latest stages of HIV disease (Breese, Judson, Penley, & Douglas, 1995; Melbye, Palefsky, Gonzales, et. al., 1990). In a study of HIV-positive and HIV-negative gay and bisexual men, Palefsky et. al. found high prevalence of HPV (93% and 61%, respectively), most of subtype HPV-16. Infection with multiple HPV types was found in 73% of HIV-positive and 23% of HIV-negative men. Higher risk for HPV has been related to a history of receptive anal intercourse, rectal administration of recreational drugs (Palefsky, Holly, Ralston, & Jay, 1998), and a higher number of lifetime sexual partners (Breese, Judson, Penley, & Douglas, 1995). In addition to sexual behavior, smoking – prevalence of which is sharply higher among gay men – increases risk of anal cancer (Daling, et. al., 1987).

Survival time among gay men with cancer may also be lower than the general population. Koblin (et. al., 1996) found five-year survival rates for non-Hodgkin’s lymphoma (9.8%) and Hodgkin’s disease (32.8%), much shorter then the expected national rates (50.2% and 75.7%, respectively). The shorter-than-expected survival rate is probably related to HIV/AIDS comorbidity. Additional reasons for the lower survival rate may include delay in detection and treatment, possibly related to barriers in accessing care or communication with health care providers. The disparity in survival rate "highlights the need for rapid identification of such patients, and more effective approaches to detection and treatment of malignancies (Koblin, et. al., 1996)." The use of rectal pap smear for detection of HPV infections has been suggested (e.g., Palefsky, et. al., 1998) as one such intervention, although further research needs to assess the association of HPV subtypes and anal cancer and the efficacy of pap smear in detection.

Finally, difficulty in assessing sexual orientation in the general population may lead to bias in studies of gay/bisexual men and cancers. Available studies typically use urban, highly sexually active men, usually of midlife age. It is not known to what extent the results are applicable to less sexually active, older, or younger populations of gay and bisexual men.

Fear of discrimination – particularly among lesbian and gay parents, who may face loss of child custody or visitation rights as a result of their sexual orientation – has made it difficult for researchers to produce reliable figures on how many lesbians or gay men are currently raising children (Patterson, 1996; Havemann, 1997). Estimates of the number of children being raised by lesbian or gay parents in the U.S. range from six to 14 million (Patterson, 1996; Havemann, 1997).

At one time most children parented by lesbians or gay men were conceived during a previous heterosexual relationship. That is now changing. The newly-coined phrase "lesbian baby boom" describes the increasing numbers of lesbians who are now choosing to have children after coming out, a pattern also reported to a lesser extent among gay men (Patterson, 1996). In 1990, it was estimated that five to ten thousand lesbians had chosen to have children after coming out (Patterson, 1996) and the phenomena has only continued to grow. In addition to the donor inseminations that make up the majority of such pregnancies among lesbians, increasing numbers of gay men and lesbians are exploring other routes to parenthood, including foster care, adoption, and co-parenting (Patterson, 1999; Cowan & Cowan, 1999).

Gay and lesbian parents have been routinely threatened by courts as being unfit per se, simply by virtue of their sexual orientation. Perhaps this explains why much of the limited research on lesbian and gay parenting is focused on the question of whether there are adverse effects to children of gay and lesbian parents. The literature does not show any negative outcomes for children raised by lesbian or gay parents (Patterson, 1995; Allen & Burrell, 1996), nor does it demonstrate that those children are more likely to become gay or lesbian themselves (Patterson, 1996). Golombok’s research showed that lesbian families had a greater mother-child interaction than heterosexual families in the sample, and greater psychological wellbeing in families where there was no genetic link to the child (Golombok, Tasker & Murray, 1997). National organizations that have policy statements supporting gay and lesbian parents include the American Academy of Matrimonial Lawyers, American Psychological Association, American Academy of Child and Adolescent Psychiatrists, and the National Association of Social Workers (Patterson, 1996).

Gay and lesbian families have few if any legal protections. The courts rule frequently in favor of biological parents and against the interests of the non-biological parent in custody cases (Editors of the Harvard Law Review, 1990; Polikoff, 1990). Agreements between known sperm donors and lesbian mothers, too, are rarely recognized as legally binding; courts fail to recognize any role for known donors except that of parent, an interpretation which is often unsatisfactory to both the donor and the mothers raising the child (Bernstein, 1998). Gay and lesbian families, particularly non-biological parents, face a range of systemic impediments to care and custody of children, including exclusion from a spouse’s health insurance coverage and hostility in school systems and health care settings. Some states expressly prohibit gay men and lesbians from adopting or serving as foster parents, and other agencies advise gay men and lesbians to pursue those options as single parents rather than introducing the subject of sexual orientation (Patterson, 1996; Martin, 1993).

Little research has addressed the effects of stress produced by such institutionalized discrimination, or the ways in which gay and lesbian families must struggle with many psychosocial factors related to their alternative status. Factors include whether the non-biological parent will be recognized by others as a parent, how the extended families will react to the new family structure, how to deal with a surrogate mother or a known-donor father, whether to choose a sperm donor who allows himself to be known later in the child’s life, how to provide the children with peers who have similar families, and what and when to tell children about donors.

In addition, evidence suggests that communication difficulties between lesbians and health care providers may impede screening and care commonly delivered through family planning clinics or by ob-gyn physicians. In the National Lesbian Health Care Survey, 27% of respondents said their current providers had assumed they were heterosexual, 16% felt they could not come out to their providers, and 11% said providers had "forced" birth control on them. One in seven of these lesbians (14%) said they had difficulty talking to their ob-gyn providers (Bradford, 1988).

In 1981, reports of unusual cancers and pneumonias among gay males were the first hint of the HIV pandemic to come (CDC, 1981). Researchers looked to the number of sexual partners, recreational drug use, and other factors to identify the cause of what was originally known as GRID, or gay-related immunodeficiency. However, it was anal sex, a common sexual practice among men who have sex with men, that proved to be the most common means of transmitting the virus that caused the symptoms now known as AIDS. The history and future of LGBT communities will forever be shaped by the decimation experienced as a result of HIV.

Since 1981, it is estimated that more than 702,000 Americans have been diagnosed with AIDS (CDC (2), 1999). Of those, 54% are reported to be men who have sex with men (MSM) (CDC (2), 1999). While the popular image of a person with AIDS remains that of the white gay man, African-American and Latino men now constitute the majority of AIDS cases among MSM since 1998 (CDC, 2000). In the states that report HIV infection, men who have sex with men (including those cross-listed as MSM and injecting drug users) constitute 38% of all newly reported cases (CDC (2), 1999), with African American and Latino MSM becoming infected in greater numbers, and at a younger age, than white men (CDC, 2000).

Growing understanding of HIV transmission has underscored the importance of distinguishing between sexual identity and sexual behavior. As indicated by the category "men who have sex with men," those at risk for HIV infection through homosexual sex include men who describe themselves as gay, bisexual, and heterosexual, and those who reject such categories altogether. Cultural differences in assessing sexuality, differences in homophobia levels within cultural and ethnic groups, and conflicts between racial and sexual identity may also contribute to reluctance to identify as gay, particularly for African American and Latino men (Diaz, Stall, Hoff et. al, 1996; Stokes & Peterson, 1998; Johnsen & Stryker, 1993; Icard, 1986). One recent study of more than 8,000 MSM of color with HIV found that as many as 24% of homosexually active African American men with HIV identified themselves as heterosexual (CDC, 2000). Some 15% of Latino men who had contracted HIV through homosexual sex identified themselves as heterosexual (CDC, 2000). The threat of HIV, or actual infection with the virus, may itself complicate the process of disclosure of sexual identity and identification with a gay community. This situation can increase an individual’s belief in negative stereotypes about gay men, binding attachment and intimacy to fears of illness, death, and loss of self and others (Saddul, 1996).

Many HIV research and prevention materials for homosexually active men have focused on men who identify as gay or bisexual, or have grouped bisexual and homosexual men together as a single group (Heckman, et. al., 1995). In fact, this elision may have most serious consequences precisely in those men of color at highest sexual risk for HIV. Studies of bisexual behavior in men have found it to be correlated with lower intention to use condoms, knowing fewer HIV-positive people, weaker perceived norms for safer sex (Heckman, et. al., 1995), and higher rates of sex that is high-risk for HIV (Doll & Beeker, 1996).

In addition, several studies have found that bisexually active men are often unlikely to disclose their bisexuality to female partners (Kalichman, Roffman, Picciano, & Bolan, 1998; Weatherburn, et. al., 1998). It is difficult to tell how many heterosexual female exposures are a result of sex with undisclosed bisexual males. Community planners have identified the role that "survival" behaviors – including convenience sex, sex for drugs, or sex for food – all play in the transmission patterns of the pandemic. These factors underscore the need for greater research and educational materials aimed at homosexually active men who do not identify as gay. In addition, some bisexual men, if infected with HIV, may encounter difficulties in accessing care, because the ability to identify with the gay community has been a proven predictor of supportive social relationships and subsequent health benefits in coping with HIV (Chapple, Kippax, & Smith, 1998; Jonsen & Stryker, 1993).

Behavioral interventions to reduce risk for HIV/AIDS are currently the only effective way of slowing the spread of HIV infection (NIH Consensus Statement, 1997). Gay communities have pioneered strategies to successfully lower risk of HIV exposure through community-based education (Coates & Collins, 1998; Kalichman et. al., 1997; Jonsen & Stryker, 1993; Saddul, 1996). In the aggregate, research has shown that while most gay men have protected sex all or most of the time, a significant percentage of MSMs – as many as one in three – have some incidence of unprotected anal sex (Hickson, et. al., 1996; Meyer & Dean, 1995). Reasons for and context of this unprotected sex, and perceived risk for HIV involved in it, vary widely. Studies show some gay men to engage in selective risk reduction strategies such as unprotected sex only or primarily with partners they believe to be of the same serostatus (both HIV positive or both HIV negative) (Kippax, et. al., 1997; Elford, Bolding, Maguire, & Sherr, 1999). In addition, a wide and sometimes contradictory range of other pyschosocial factors have been shown to influence sexual risk-taking, among them: self-esteem, social supports or lack thereof, mood prior to sexual encounter, optimism, fatalism, age, education, and alcohol or drug use (Hospers & Kok, 1995). A growing body of literature suggests the importance of moving beyond informational prevention education to consideration of mental health models and support in the service of effective HIV prevention.

Health implications of the HIV epidemic for gay men are myriad, ranging from life-threatening opportunistic infections and malignancies to mental health challenges noted above and elsewhere in this document. While the advent of highly active anti-retroviral treatments (HAART) has sharply reduced AIDS deaths and opportunistic infections (CDC (3), 1999), long-term effects of the combination anti-HIV drug regimens are unknown. Recent years have seen increasing reports of high cholesterol, diabetes, and redistribution of body fat known as lipodystrophy (Kaul, et. al., 1999), raising questions about the treatment’s long-term viability.

Gay men may also face particular obstacles in accessing health care, or in achieving the communication with health care providers that is critical to adherence to treatment regimens. The implications of failure to adhere, as well as of saturation in some gay communities of earlier, less effective treatment regimens, are only now beginning to be gauged by research. Studies have already documented passage of a drug-resistant virus from one gay man to another, and one recent study among newly infected gay men found that as many as 16% had HIV that was somewhat resistant to one or more AIDS drugs (Boden, et. al., 1999).

HAART’s efficacy may also impact HIV prevention efforts and other health supports for homosexually active men. Fueling the perception that AIDS has become a manageable, chronic infection, advent of HAART has been accompanied by reports of sharp drops in funding for community based AIDS prevention and service organizations serving gay men and increases in high-risk behavior. One study, for example, found that 18% of HIV-positive gay men were now practicing safe sex less often because of treatment advances (Kelly et. al., 1998).

While the long-term treatment effects of HAART are not fully understood, neither are its effects on HIV transmissibility. Some research points to a reduction of HIV in seminal fluid associated with a reduction in viral load (Gupta, et. al., 1997). Based on the CDC's retrospective case-control study of HIV-exposed health workers, and extrapolation from the use of antiretrovirals to interrupt perinatal transmission, a number of urban centers have begun offering gay men and others post-exposure prophylaxis in an effort to contain viral replication for those thought to have experienced recent exposure (Kowng, et. al, 1999). While these findings and efforts are preliminary and inconclusive, they suggest possibilities for future HIV prevention strategies.

Youth represent a subgroup of all MSM who are particularly likely to engage in high-risk behavior, and so are particularly at risk for HIV. Adolescents and young adults (ages 13-24) are the single most likely group to contract an STD (CDC, 1998). Among this same group, the rate of HIV infections is growing particularly rapidly, with 31% of all new infections being reported as MSM. In a sample of young MSMs (ages 15-22) in six urban counties, between 5 and 9 % were found to be infected with HIV – with a significantly higher percentage of African American youth and Latino youth being infected than white youth (CDC (1), 1999). Other studies have shown that adolescent MSMs show markedly higher rates of unprotected receptive anal sex than do older MSMs and that sex with older people is highly correlated with exposure to HIV (Morris, Zavisca, & Dean, 1995). One study of inner-city youth attending a gay community center program reported consistent condom use in only 13% of adolescent MSMs, while one quarter engaged in prostitution, and all reported sexual activity (Rotheram-Borus, et. al., 1992).

The impact of HIV on the lesbian community, and the risks of female-female HIV transmission, remain underresearched. Since people with HIV are categorized by the highest risk group to which they belong, lesbians with HIV who have had any heterosexual contact or injection drug use history are usually classed as heterosexual or injectors, regardless of their personal sexual identification or history (Cohen et. al., 1993). While small studies among serodiscordant lesbians have found no evidence of female-to-female HIV transmission (Raiteri et. al., 1994), numerous health advocates have urged for greater research, as well as consideration of broader assessment of HIV risk for lesbians; in 2000, the Centers for Disease Control will fund a 200-participant study of lesbians and HIV with cohorts in Washington, D.C., San Francisco and New York (Smith, 2000). Fears of elevated risk for HIV among some lesbians is not restricted to their sexual contact with women: HIV-positive women reporting female sexual partners in one San Francisco study, for example, were twice as likely to have used drugs and significantly more likely to have engaged in anal intercourse with a man than those not reporting female sexual partners (Young, Weissman, & Cohen, 1992).

Vaccination recommendations for gay and lesbian people do not differ significantly from those for the general population, with two notable exceptions. Both exceptions are for the prevention of viral hepatitis, for which gay and bisexual men are at increased risk of transmission. . Immunization against hepatitis B virus (HBV) for all homosexually active men has been recommended since the 1980s, and vaccination against hepatitis A virus (HAV) was recommended in 1996 by the Advisory Committee on Immunization Practices (ACIP) for gay and bisexual men, as well as for certain other high-risk groups (CDC, 1997).

Although HAV and HBV vaccination is recommended for gay and bisexual men, national prevalence rates for these vaccinations are currently unknown, hampering efforts to assess the successful implementation of vaccination campaigns for these men. However, available data suggest that rates of vaccination for men who have sex with men are low. In 1996, the Centers for Disease Control found that only 3% of the sample of young MSM were vaccinated against HBV (CDC, 1996). The CDC analyzed serologic data from the 1992-1993 Young Men's Survey (YMS) conducted by the San Francisco Department of Public Health to estimate hepatitis B vaccination coverage among young MSM. The survey used a targeted sampling method to enroll MSM aged 17-22 years at selected public venues in San Francisco and Berkeley, California. Of the 385 young MSM eligible for the study, 20% had evidence of previous or current HBV infection, while an additional 3% were positive for anti-HBs alone, suggestive of hepatitis B vaccination. Of the remaining 77% who lacked evidence of vaccination or infection, 80% reported having had anal sex or having injected drugs during the preceding 6 months. Of these, 86% reported receiving one or more types of health-care (CDC, 1996).

See also III. C., HIV, III. G., Substance Use, and III. I., Violence and Sexual Assault

Most studies of mental disorders among gay, lesbian, or bisexual people in the 1960s and 70s addressed issues related to the status of homosexuality as a mental disorder (Bayer, 1981; Gonsiorek, 1991; Morin, 1977). Such studies helped lead to the declassification of homosexuality as a mental disorder in 1973 and the removal of "ego-dystonic homosexuality" from the 1986 Diagnostic and Statistical Manual of Mental Disorders (DSM-III). Epidemiological advances of the past two decades, including the recognition of the importance of population surveys (rather than clinical studies) of mental disorders, the introduction of an improved psychiatric classification system, and the development of measurement tools and techniques for epidemiological research, have made earlier research on mental health among LGBT populations virtually obsolete.

Few recent studies have used these advances to focus needed research on gay men, lesbians, or bisexuals. For example, no information on LGB individuals has been provided by either of two recent large-scale, national psychiatric epidemiological surveys conducted in the U.S.–the Epidemiological Catchment Area study (Robins & Regier, 1991) and the National Comorbidity Survey (Kessler, McGonagle, Zhao, et. al., 1994). LGB populations may be at increased risk for mental distress, mental disorders, substance use, and suicide because of exposure to stressors related to societal antigay attitudes. Known social stressors include prejudice, stigmatization, and antigay violence (Meyer, 1995; Rosario, Rotheram-Borus, & Reid, 1996; Gonsiorek & Rudolph, 1991). Internalization of negative social attitudes (internalized homophobia) in particular has been related to intimacy and sexual problems and other adjustment difficulties (Meyer & Dean, 1998), as well as high HIV risk-taking behaviors among young gay and bisexual men (Meyer & Dean, 1995). Understanding patterns of and risk factors for mental disorders in LGB individuals is important for tailoring proper mental health treatment and for designing effective public health intervention and prevention programs (Council on Scientific Affairs, American Medical Association, 1996).

As noted above, methodologically sound studies of mental disorders in LGB populations are lacking. Many studies have used symptom scales in which the relationship to psychopathology is unclear (Bradford, Ryan, & Rothblum, 1994; Cochran & Mays, 1994; Gonsiorek, 1991); have included no comparison group (Bradford, et. al., 1994; Tross, et. al., 1987; Williams, et. al., 1987; Martin, 1990); have employed convenience samples unlikely to be representative of LGBT populations (Saghir & Robins, 1973; Pillard, 1988; Atkinson, 1988); or have expressed potential bias in their definition or detection of homosexuality (Fergusson, et. al., 1999; Cochran, et. al., in press). Most studies fail to distinguish between findings on gay or lesbian individuals and those on bisexuals, making comparisons of gay men and lesbians with bisexual men and women difficult.

A number of studies have used criteria-based case diagnoses to study mental illness among gay men, and used a comparison group to assess differences between gay and heterosexual men. Pillard (1988) found elevated rates of bipolar disorders among gay men; Atkinson (1988) found elevated rates for most mental disorders among gay men; and Cochran et. al. (in press) found homosexual men to have higher rates of major depression disorder than heterosexuals. Fergusson et. al. (1999) found lesbians, bisexuals and gay men to have higher rates of major depression disorder, generalized anxiety disorder, and conduct disorders than a heterosexual sample, though the author did not disaggregate data by gender. Other studies using case diagnoses included no explicit comparison group. Compared with estimates of U.S. rates of disorder reported by the Epidemiologic Catchment Area study (ECA, Robins & Regier, 1991), Williams et. al. (1991) found high lifetime prevalence of affective disorders, but no elevated prevalence of current disorders; Tross et. al. (1987) found a slightly elevated prevalence for current major depressive disorder; and Martin (1990) found no increased prevalence of alcohol abuse/dependence among gay and bisexual men.

Mental health among lesbians, too, has gone largely unstudied or been inadequately researched. As recently as 1990, the Women and Depression Task Force Report of the American Psychological Association (McGrath, Keita, Strickland, & Russo) reviewed several hundred studies on women and depression, none of which focused on lesbians. The largest and most comprehensive survey of lesbian mental health to date is the National Lesbian Health Care Survey (NLHCS) (Bradford, Ryan, & Rothblum, 1994) of a sample of 1,925 lesbians. Symptoms of depression in the study were roughly equivalent to those in studies among heterosexual women (McGrath, Keita, Strickland, & Russo). Similarly, Cochran and Mays (1994), reported similar levels of depressive symptoms among African American lesbians and gay men as those found in studies of African Americans in the general population.

While homosexuality has been removed from the American Psychiatric Association’s diagnostic manual, GID remains. As defined in the DSM, GID is "strong and persistent cross-gender identification, which is the desire to be, or the insistence that one is, of the other sex," and "persistent discomfort about one’s assigned sex or a sense of inappropriateness in the gender role of that sex (American Psychiatric Association, 1994)."

While not necessarily involving sexual attraction to the same gender, a GID diagnosis is seen by some gay and lesbian health professionals as reinforcement for normative gender standards that are hostile to feminist or gay-affirmative values (Corbett, 1999). A child or adult who perceives that he or she is attracted to a person of the same sex may experience some gender identity "confusion" because by definition their sexuality deviates from assigned gender roles. For example, a boy who feels same-sex attraction, having internalized societal values, may feel that he is girl-like in that attraction. This may indicate nothing but the beginning of a process of questioning social norms, but can be labeled a disorder according to the DSM’s categories. Similarly, the diagnosis may "punish" innovators who challenge existing gender roles (e.g., a girl who wants to be on a football team, a boy who wants to grow up to be a mom) by stigmatizing them.

Sensitive to critiques of diagnosis as a means of social control, authors of the DSM-IV make "clinically significant distress or impairment in social, occupational, or other important functioning" a requirement for GID diagnosis. However, distress is subjective, and widely defined. For example, subjective distress could simply be the interpersonal problems one has with parents because of one’s nonconforming behavior. It is hard to imagine any person who is somewhat different not experiencing some distress, at least related to others’ reactions to his or her non-normative behavior. A child who is taunted for being a "sissy," or who is socially isolated because she is "masculine," will clearly experience subjective distress, but the source of the distress is oppression and intolerance – not an inherent disorder. This is analogous to ethnically based taunting.

Small, non-random studies of boys with childhood GID have indicated that 75% of boys with GID may grow up to be homosexual men (DSM-IV). This may heighten concerns that parents and physicians will see GID as a precursor, or even a "risk factor," for homosexuality, and – if seen as a problem – may seek to treat it aggressively. All these have been discussed as reasons for the declassification of GID as a mental disorder. Against this, some transgender advocates argue for maintaining the classification, suggesting that the inclusion of GID in the DSM-IV may allow for insurance reimbursement and treatment for transgender people seeking psychiatric therapy or hormonal treatment. Investigation is needed to determine how such third-party payments can be appropriately authorized without unnecessarily stigmatizing children and adults for gender roles that do not fit the dominant norm.

Stress and mental health raise important practical public health issues related to HIV risk taking behavior, particularly as prevention research has begun to move from purely cognitive explanatory models of risk taking behavior to the consideration of affective and non-rational processes among gay men (McKirnan, Ostrow, & Hope, 1996; Kelly, Murphy, Sikkema, & Kalichman, 1993; NIMH, 1999. Mental health problems, including substance use and personality and psychological constructs (e.g., low self-esteem, sensation seeking) have been associated with HIV-related risk taking behavior (Bartholow, Doll, Joy, et. al., 1994; Graham, Kirscht, Kessler, & Graham, 1998; Strathdee, Hogg, Martindale, et. al., 1998; Hays, Paul, Ekstrand, Kegeles, et. al., 1997; Kalichman, Kelly, & Rompa, 1997; Diaz, Stall, Hoff, et. al., 1996; Kalichman, Heckman, & Kelly, 1996; Kalichman, Kelly, Morgan, & Rompa, 1997; Chesney, Barrett, Stall, 1998; Leviton, 1989; National Institutes of Health, 1997). Research has also described high-risk behavior as an emotion-focused coping strategy to reduce stress (Folkman, Chesney, Pollack, & Phillips, 1992).

Alcohol and drug use, in combination with emotion-focused coping, serve a disinhibiting function leading to more risk taking (Folkman, Chesney, Pollack, & Phillips, 1992; Meyer & Dean, 1995, 1998). The convergence of internalized homophobia, drug problems, and AIDS-related traumatic stress response are related to increased risk taking (Meyer & Dean, 1998). In addition, in the context of reports of increased risk for suicide among gay and bisexual men, the relationship of mental health and HIV risk taking is of particular concern. Some researchers have suggested that high-risk sexual behavior among young gay and bisexual men may be related to a sense of hopelessness and suicidal tendencies (Frances, Wikstrom, & Alcena, 1985; Odets, 1994; Flavin, Franklin, & Frances, 1986).

Whether gay/bisexual men and lesbians have higher rates of suicidal behavior has been widely debated in recent years, and research evidence is mixed (Muehrer, 1995). Studies of suicide ideation and attempts found elevated lifetime prevalences for gay/bisexual men and lesbians (e.g., D’Augelli & Hershberger, 1993; Kurks, 1991; Remafedi, Farrow, & Deisher, 1991; Schneider, Farberow, & Kurks, 1989; Pillard 1988; Bell & Weinberg, 1978; Saghir & Robins, 1973), but studies of completed suicides found no evidence for elevated rates (e.g., Rich, Fowler, Young, & Blenkush, 1986; Shaffer, Fisher, Hicks, Parides & Gould, 1995). Results of the NLHCS found that more than half the sample had had thoughts about suicide at some time, and 18% had attempted suicide (Bradford, et. al., 1994). This compares with 33% and 4%, respectively, for women in the U.S. as reported in the ECA studies (Robins & Regier, 1991; Moscicki, 1994). Several recent studies of gay youth populations continue to report alarmingly high rates of suicide ideation and attempts among gay and bisexual youth. For example, Faulkner and Cranston (1998), Remafedi et. al. (1998), and Fergusson et. al. (1999) found that rates of various measures of suicide ideation and attempts were three to seven times higher among gay and lesbian youth compared with heterosexual youth.

Both types of studies – of completed suicides and of suicide ideation and attempts – have severe methodological limitations. The former uses samples biased toward underestimating rates of completed suicides because of difficulties in post-mortem classification of sexual orientation; the latter uses samples biased toward over-reporting of suicide ideation and attempts because more vulnerable gay youth may be identified at younger ages. These problems have led members of a workgroup convened by the NIMH and the CDC to recommend epidemiological study of mental disorders and suicide as an important priority (Working Groups, Workshop on Suicide and Sexual Orientation, 1995).

Social stressors may impact body image and eating patterns among gay and lesbian adolescents. Significant literature suggests that adolescence, and the social stress that accompanies it, may place gay men in particular at increased risk for body dissatisfaction and problem eating behaviors whose onset commonly occurs during this period.

In addition to this social stress hypothesis, other research suggests a sociocultural hypothesis linking body image dissatisfaction and eating disorders with the cultural valuation of a thin body (Silberstein, Mishkind, Streigel-Moore, et. al., 1989; Hefferman, 1994). According to this view, emphasis on thinness normally placed on feminine bodies may also influence gay men, who are more likely to demonstrate atypical gender role behavior (Fichter & Daser, 1987), and may place greater cultural emphasis on appearance (Herzog, Newman, & Warshaw, 1991; Siever, 1994). In contrast, a decreased emphasis on physical appearance among lesbians may be protective against the development of weight concern, body dissatisfaction, and disordered eating behavior (Siever, 1994; Beren, Hayden, Wilfley, & Grilo, 1996).

Clinical samples assessing whether homosexuality is overrepresented among patients with eating disorders has found sexual orientation to be a significant predictor of eating disorders among men, though not among women. A study of the 135 men treated for eating disorders at Massachusetts General Hospital from 1980 to 1994 showed 27% of all men reported homosexual orientation and 42% of bulimic patients identified as homosexual or bisexual (Calat, Camargo, & Herzog, 1997). These results are consistent with other smaller studies, which found association between male homosexuality and eating disorders, showing that a higher proportion of men who were anorectic or bulimic were gay than would be expected. (Herzog, Norman, Gordon, & Pepose, 1984; Schneider, & Agras, 1987; Robinson & Holden, 1986; Fichter & Daser, 1987).

Nevertheless, it is difficult to assess the validity and generalizability of these conclusions. With an overall prevalence of less than 1% for bulimia and 3% for anorexia (APA, 1994; Hall, Dalahunt, & Ellis, 1985), eating disorders are rare, and men account for less than 10% of total cases (APA, 1994; Carlat & Camargo, 1991). In addition, research on the association of eating disorders and homosexuality has used varying definitions of both eating disorders and homosexual tendencies, sometimes equating lack of heterosexual experience or atypical gender behavior with homosexuality (Fichter & Daser, 1987). The effect of comorbid psychiatric disorders (e.g., depression) is largely ignored in these studies. Differences by gender in reporting of eating habits may also bias results. For example, women are more likely than men to label eating a large amount of food as a "binge" (Calat & Camargo, 1991). To the extent that this bias in reporting style is also characteristic of gay men, it may lead to over-reporting and increased detection of anorexia and bulimia among gay men as compared with heterosexual men. Heterosexual men may be more reluctant than homosexual men to seek help for eating disorders because the disorders are perceived, and thus stigmatized, as "feminine" (Calat & Camargo, 1991). Finally, while many clinical studies report an association with male homosexuality, several found no such association (Turnbull, Freeman, Barry, et. al., 1987; Pope, Hudson, & Jonas, 1986; Herzog, Newman, & Warshaw, 1991).

Community studies – which have typically sampled college students or participants in gay and lesbian organizations – have often found significant association between sexual orientation and negative body image among gay men (Silberstein, Mishkind, Striegel-Moore, et. al., 1989; Beren, Hayden, Wilfley, & Grilo, 1996; Herzog, Newman, & Warshaw, 1991), but it is impossible to draw clear conclusions from them. These studies typically used small samples of easily recruited subjects, and their generalizability to the general gay population is questionable. Also, some studies show that heterosexual men with eating disorders have similar problems in body image, exercise, and dieting behavior (Olivardia, Pope, Magweth, & Hudson, 1995). Among women, while being female was related to concerns about dieting and being fat, lesbians were less likely to be preoccupied with weight and body image than heterosexual women (Siever, 1994; Brand, Rothblum & Solomon, 1992).

The Minnesota Adolescent Health Survey (Neumark-Sztainer, Story, Resnick, & Blum, 1998), a survey of more than 30,000 Minnesota students (grades 7 — 12), was the only one to use a random community sample. In that study, homosexual boys were more likely than heterosexual boys to report a poor body image (28% vs. 12%), frequent dieting (9% vs. 6%), binge eating (25% vs. 11%), and purging behaviors (e.g., vomiting: 12% vs. 4%) (French, Story, Remafedi, Resnick, & Blum, 1996). Homosexual girls, by contrast, were more likely than heterosexual girls to report a positive body image (42% vs. 21%), though no less likely to report disordered eating behaviors such as binge eating or purging. (Rogers, Resnick, Mitchell, & Blum, 1997). While random, the sample did identify adolescents of high school age or younger. Because many gay men and lesbians do not know or disclose their sexual orientation until older, it may be difficult to generalize the results to gay men and lesbians more broadly.

It is important to note that significant gender differences related to eating problems hold true for gay and lesbian populations (Saewyc, Bearinger, Heinz, et. al., 1998). Young boys (15%) in the Minnesota sample were less likely than young girls (27%) to rate themselves low on body image. A greater proportion of older females (38%) than males (22%) considered themselves to be overweight. In each group more girls than boys reported dieting in the last year, at all levels of frequency. Despite increased risk among homosexual men compared with heterosexual men, women (both lesbian and heterosexual) are at greater risk for body dissatisfaction and disordered eating.

Like members of other minority groups, LGBT populations are not passive recipients of stigma and discrimination but engage in active coping to counter the ill effects of negative social stressors. Such positive coping has been shown to be beneficial to members of minority groups (Shade, 1990). Minority status is related not only to stigma and discrimination (stress) but also to structural resources, such as group solidarity and cohesiveness, that protect minority members from the adverse mental health effects of social stress (Crocker & Major, 1989; Kessler, et. al., 1985). Even exposure to antigay violence, while creating a crisis with potential adverse mental health outcomes, also presents "opportunities for subsequent growth" (Garnets, et. al., 1990). Jones and colleagues (1984) described two functions of minority-related coping, cohesiveness and support: 1) to allow stigmatized persons to experience social environments in which they are not stigmatized by others; and 2) to provide support against the negative evaluation of the stigmatized minority group. Social evaluation theory suggests another plausible mechanism for minority coping. In-group support can provide a reappraisal (Lazarus & Folkman, 1984) of the stressful environment, yielding it less injurious to psychological well being. Through reappraisal, the minority group validates its experiences (Thoits, 1985; Crocker & Major, 1989).

Oetjen and Rothblum (in press) used a standardized measure to examine the effect among lesbians of factors consistently cited in the women and depression literature (relationship status, relationship satisfaction, social support from friends, and social support from family), and two unique factors (outness and relationship status satisfaction), to determine their ability to predict depression among lesbians. Perceived social support from friends, relationship status satisfaction, and perceived social support from family were found to be significant predictors, accounting for 17.8% of the variance in depression. Among those lesbians in committed relationships, social support from friends was the only predictor, accounting for 5.8% of the variance in depression. A similarly designed study among Canadian lesbians found lower depression rates among women in relationships, lesbians who had more social support from friends and family, and those who were more open about their sexual orientation. Some 38% of the variance in depression scores could be predicted from the four variables (Ayala & Coleman, in press).

In addition to HIV, men who have sex with men (MSM) are at increased risk of certain sexually transmitted diseases relative to their heterosexual counterparts (Council on Scientific Affairs, 1996; Harrison & Silenzio, 1996; Ungvarski & Gorssman, 1999). The stigma associated with sexually transmitted infections in general and with homosexual acts make accurate estimates of the prevalence rates for these conditions in MSM almost completely unavailable. STDs for which homosexually active men are at risk include urethritis, proctitis, pharyngitis, prostatitis, hepatitis A (HAV) and B (HBV), syphilis, gonorrhea, chlamydia, herpes, genital warts, and HIV infection. MSM are believed to be at generally increased risk for these infections, although additional data on patterns of infection and definitive prevalence data are needed. In addition, high rates of HIV infection among MSM put them at danger for chronic or life threatening complications from sexually transmitted infections that are harmless or self-limiting in HIV-negative individuals, including cytomegalovirus, herpes, and anal cancer associated with strains of the Human Papillomavirus. In spite of availability of vaccines to prevent hepatitis A and B, and recommendations calling for vaccination of sexually active MSM against these viral infections, rates of vaccination for these diseases remain low.

After sustained declines in rates of STDs in gay and bisexual men since the beginning of prevention programs aimed at HIV/AIDS, recent data from many U.S. and international cities indicate that rates of unprotected anal sex and STDs may be increasing (Centers for Disease Control (6), 1999, Community Disease Report Weekly, 1999). STD clinics and public health departments report increases in rectal gonorrhea rates among MSM in San Francisco (CDC, 1999), as well as increases in syphilis rates among MSM in Seattle. Community-based clinics reported epidemic rates of HAV in 1998 and 1999 among gay men in New York, Boston, Atlanta, and various cities outside the U.S. (Wolfe, in press). These increased STD rates are coincident with the decline in the number of MSM reporting that they "always used condoms," and an increased proportion of MSM reporting unprotected anal sex. This points to a clear need for renewed efforts to reduce the rates of STD and HIV transmission among MSM, as well as the development of new surveillance techniques to assess the efficacy of these efforts.

No known gynecological problems are unique to women who have sex with women, and none are believed to occur more often in lesbians than in bisexual or heterosexual women. STDs appear to be less common in women who identify as lesbian and in women who are sexually active only with women than in either heterosexual women or gay men. This may be due in part to a relative epidemiological isolation of this group from men and the lack of penile-vaginal intercourse.

Human Papillomavirus and bacterial vaginosis have been shown to be transmissible between women and do occur in lesbians (Marrazzo, Koutsky, Stine, et. al., in press; Berger, Kolton, Zenilman, et. al., 1995). Candidiasis and Trichomonas vaginalis infection do occur in lesbians and appear to be transmissible between women (Degen & Waitkevicz, 1982). Women who are sexually active only with women appear to have a lower incidence of syphilis and gonorrhea than any other group except those who have never been sexually active. Infections with chlamydia or herpes virus disease appear to be less common in lesbians who have been sexually active exclusively with women, but all are theoretically transmissible (Johnson, Smith & Guenther, 1987; Robertson & Schachter, 1981; Degen & Waitkevicz, 1982). Hepatitis A, amebiasis, shigellosis, and helminthism also have a low prevalence in these women. Hepatitis B and C occur only when other risk factors are present (Walter & Rector, 1986; William, 1981).

Epidemiologic studies on alcohol and other drug abuse have rarely asked about sexual orientation. Or, if they have, they exhibited serious methodological flaws when focusing on gay and lesbian populations. Early studies, for example, recruited gay and lesbian subjects in bars, a sample which not surprisingly found higher rates of heavy alcohol and drug use than among the general population (Fifield, Lathan & Phillips, 1977; Lohrenz, et. al., 1978, Saghir & Robins, 1973). These and other opportunistic samples have frequently put prevalence of excessive or problem drinking among gay men, lesbians, and bisexuals near 30%, as compared to 10% among the general population (Paul, Stall & Bloomfield, 1991).

Subsequent studies, recruiting subjects via other means, have generally disputed claims that alcoholism among gay men and lesbians reaches 30%. McKirnan and Peterson (1989), surveying readership of a Chicago gay newspaper, found rates of heavy drinking among gay men and lesbians to be comparable to those in the general population. Bloomfield (1993), using random phone surveys in the San Francisco area, reported that there were no significant differences in levels of drinking and bar-going behavior between lesbians and bisexual women and their heterosexual counterparts. Skinner (1994) found higher rates of drinking among lesbians than among women in a geographically matched sample. Stall and Wiley (1988), comparing alcohol use patterns of heterosexual and homosexual males in San Francisco, found no significant differences in quantity and frequency of alcohol consumption overall, though differences at the extreme patterns of use were noted. Gay and bisexual men were approximately twice as likely to be heavy drinkers or abstainers as heterosexual men. Lesbians and bisexual women, too, report higher rates of abstention than heterosexual women in some studies (Bloomfield, 1993).

Sound data about substance use among lesbians is even more scarce, though McKirnan and Peterson (1989) found that rates of marijuana and cocaine use were higher among lesbians than among heterosexual women, and that differences between light and heavy use were smaller than those found in the general population. Skinner and Otis (1996), surveying gay men and lesbians, found few gender differences in substance use, though lesbians in some age groups reported somewhat higher rates of marijuana and cocaine use, as well as higher rates of smoking. Gay men reported higher rates of use of inhalants such as amyl or butyl nitrite ("poppers"), of hallucinogens, and of illicit drug use overall. LGB health researchers and advocates have called for greater research into the effects of recreational drugs common in nightclubs and dance parties (e.g., ketamine, MDMA, etc.), as well as for research into the potential and on-going consequences of chronic pharmacologic manipulation of neurotransmitters (Abrams to Leshner, 1999).

Belief in a direct causal relationship between substance abuse and unsafe sexual behavior has caused the phenomenon of substance use to be better documented among gay men since the advent of HIV (Stall, McKusick, et. al., 1986). Though the assumption that substance use causes unprotected sex has been critically challenged , a wide body of evidence suggests that there is an association between recreational drug use and high-risk sexual behavior, and that gay men use particular drugs more often than their heterosexual counterparts. Stall and Wiley (1988) found gay men significantly more likely to have used marijuana and psychedelics in the last six months, three times as likely to have used barbiturates, five times as likely to have used MDMA, and 58 times more likely to have used "poppers." It is important to note, however, that while this study found gay men more likely to have used more drugs, it did not demonstrate higher levels of dependence or addiction.

Adolescent and young adult gay men and lesbians appear to be especially at risk for substance abuse, and with important health consequences. With the exception of popper use, differences in drug use between heterosexual and homosexual men noted by Stall and Wiley (1988) were largely attributable to higher rates of drug use among the youngest cohort. Data from the 18-to-25-year-old cohort of gay men in the Trilogy Project (Skinner & Otis, 1996) found rates of 87.1%, 78.5%, 31.5%, 57.0%, and 33.3% respectively, for alcohol, marijuana, cocaine, inhalants, and hallucinogens, with an overall rate of 87.1% for any illicit drug use. Young lesbians, too, report overall rates of substance use in excess of 80%. Given the association between substance use and a variety of adverse health consequences for which young gay men and lesbians are at increased risk – including HIV infection and suicide– greater attention is needed to the problem and prevention of substance abuse by young gay men and lesbians.

Finally, though data is not yet conclusive, research suggests that a number of illicit substances may speed replication of HIV or have an immunosuppressive effect, a fact which may impact significantly on the health of the disproportionate number of gay men who are at risk for or are infected with HIV (Seage, 1992; Bagasra & Pomerantz, 1993). Further, adverse interactions have been documented between recreational drugs and other medications. Poppers, for example, cause potentially fatal drops in blood pressure when combined with the common erectile dysfunction medication sildenafil citrate (Viagra). Following the death of a British gay man in 1997, community publications alerted gay men to the possibility that Ritonavir, a common anti-HIV medication, might inhibit the liver’s ability to process the drug MDMA, boosting levels of the recreational drug to potential fatal levels (Wolfe, in press).

Evidence suggests that gay men and lesbians may perceive themselves to be at increased risk for alcoholism and substance abuse, that they have increased need for drug and alcohol treatment, and that they face particular barriers in accessing such treatment. McKirnan and Peterson (1988, 1989) found that while heavy drinking patterns did not differ significantly by sexual orientation, gay men and lesbians reported rates of alcohol problems nearly twice as often as heterosexuals. Drinking rates among gay men and lesbians do not appear to decrease with age as quickly as they do in heterosexual populations (Skinner, 1994, McKirnan & Peterson, 1988; Stall & Wiley, 1988; Bradford & Ryan, 1994). Gay and lesbian Alcoholics Anonymous meetings have become the largest special interest group within the self-help fellowship (Paul, Stall & Bloomfield, 1991). Assessments of alcohol and drug treatment facilities have documented lack of staff training in treatment issues for gay and lesbian alcoholics, and few or no gay staff (Hellman, 1991; Garnets, et. al., 1991), in spite of evidence that gay, lesbian and bisexual clients are more willing to attend treatment programs that address gay issues and less likely to comply with treatment from homophobic mental health providers (Paul, Stall & Bloomfield, 1991; O’Hanlan et. al., 1997)

Effective research into the treatment needs of gay men and lesbians should include enhanced understanding of the role of subcultural factors in forming and influencing patterns of both substance use and sexual behavior. Such factors, including the reliance on bars for socialization, stress caused by discrimination, and targeted advertising by liquor companies in gay and lesbian publications have been documented, but remain inadequately understood (Hughes, 1997; Nardi, 1991).

Adverse health effects of tobacco use among gay and lesbian populations are similar to those among the general population. Evidence suggests, however, that the rates of tobacco use among sexual minority men and women may exceed those of the general population, ultimately leading to increased rates of tobacco-related disease.

As with surveys of alcohol use among gay men, lesbians and bisexuals, studies of tobacco use in these populations tended to use non-random samples, often drawn from bar patrons, and to report rates of tobacco use sharply higher than those of their heterosexual counterparts (Stall et al., 1999). Unlike studies of alcohol use, however, later more representative studies of tobacco use seem to support earlier reports that the prevalence rate is strikingly higher among gay men than in the general male population. Using a household-based sample, Stall, et. al. (1999) found 41.5% of gay adults to be smokers, a rate far in excess of the national rate of tobacco use by men generally (28.6 %) (Centers for Disease Control, 1994). DuRant (1998) found that adolescent males who engage in same-sex sexual behavior also have increased rates of tobacco use relative to their peers, and that higher numbers of male sexual partners correlated with a higher frequency of tobacco use, as well as higher rates of drug use, victimization, and use of violence.

As with alcohol use, representative studies of lesbian tobacco use have yet to be completed. Cardiovascular diseases (e.g., heart, stroke, arteriosclerosis), however, represent the leading causes of death for women in general. Existing non-random studies indicate that lesbians may smoke more and have a higher body mass index than heterosexual women, and thus may be at higher risk for cardiovascular disease and cancers (Bradford, Plumb, White & Ryan, 1994; White & Dull, 1997; Moran, 1996). Because lesbians appear to seek health care less often, they are less likely to receive blood pressure and cholesterol screening, further compounding their risk. In addition, comparisons between young gay men and lesbians (Skinner and Otis, 1996) have found that lesbians actually smoke more than their gay counterparts, furthering concern about their risk and the need for greater research in this area.

A report on the response of the criminal justice system to bias crimes concluded that lesbians and gay men are among the most frequent victims of hate violence in the United States and as a group they "are probably the most frequent victims" (Finn & McNeil, 1987). Acts of aggression, denigration, and violence against LGBT people have been documented in a variety of settings including: schools and colleges, the armed services, jails and prisons, at homes, in work places, and in public places (APA, 1997; Bradford, et. al., 1994; D’Augelli, 1989; Garnets, et. al., 1991; Herek, 1989; Levine & Leonard, 1984; Rothblum, 1994; Wolfe, 1998; Wooden & Parker, 1982).

IV. A. Overview and Definitions

The scientific literature on gender identity, sex, and sexuality lags significantly behind the gender-focused scholarship in history, literature, anthropology, philosophy, cultural studies, gender studies, and other disciplines (Elliott, 1998). Researchers and scholars in the latter fields have undertaken theoretical analyses and presented extensive cross-cultural and historical evidence marking the narrow horizons of western and particularly U.S. definitions of gender (Chauncey, 1994; Devor, 1997; Herdt, 1994; Ekins & King, 1996; Feinberg, 1996; Gagne & Tewksbury, 1998; Garber, 1992; Garfinkel, 1967). This scholarship has followed the lead of transgender activists who have facilitated a shift in perspective, away from notions of gender pathologies to ones of gender "non-conformity" and the problems inherent in narrow and rigid societal definitions of gender.

Gender is now more commonly understood as having a strong cultural definition in addition to precise biological and extensive psychosocial components. The public health of transgender populations should be pursued with this interdisciplinary definition in mind. Thus, the health issues related to gender non-conformity may no longer be interpreted as confirmation that transgender identities or behaviors are inherently pathological. Rather, the health issues are defined as either a normal component of various stages of change specific to transgender individuals – concerns that may be related to, but are neither rooted in nor caused by, an individual’s gender identity – or the result of prejudice, discrimination, and other culturally-based stressors (Cohen, et. al., 1997; Gagne, 1997; Israel & Tarver, 1997; Mason-Schrock, 1996).

In 1999 the American Public Health Association proposed a series of policy statements focused on the public health needs of transgender populations. In their statement, "The Need for Acknowledging Transgender Individuals within Research and Clinical Practices," the Association:

1. Urges the National Institutes of Health and the Centers for Disease Control and Prevention (as well as individual researchers and health care workers) to categorize MtF [Male to Female] and FtM [Female to Male] transgendered individuals as such and not equate them with gay men or lesbians (unless as appropriate to an individual’s sexual orientation in their preferred gender) as well as acknowledging the variation that exists among transgendered individuals.
2. Urges researchers and health care workers to be sensitive to the lives of transgendered individuals and treat them with dignity and respect, and not to force them to fit within rigid gender norms. This includes referring to them as the gender with which they identify.
3. Urges researchers, health care workers, the National Institutes of Health, and the Centers for Disease Control and Prevention to be aware of the distinct health care needs of transgendered individuals. and
4. Urges the National Institutes of Health, and the Centers for Disease Control and Prevention to make available resources that will enable a better understanding of the health risks of transgendered individuals, especially the barriers they experience within health care settings.

Transgender is an umbrella term describing a number of distinct gender positions and identities including: transsexual, transvestite, transgenderist, androgyne, intersex, hermaphrodite, and the states of crossgender, crossliving and crossdressing (American Public Health Association, 1999; Israel & Tarver, 1997). As gender possibilities are elaborated and cultural definitions of gender and sex are further expanded, additional identities are likely to emerge, though current definitions include the following:

~ American Educational Gender Information Service (AEGIS)

~ Gender Public Advocacy Coalition (GenderPAC)

~ The Harry Benjamin International Gender Dysphoria Association (HBIGDA)

~ International Foundation for Gender Education (IFGE)

~ Intersex Society of North America (ISNA)

IV. B. Barriers to Care

Prejudice against transgender individuals is pervasive. There is a long-held view on the part of U.S. medical providers and researchers, as well as the public at large, that transgenderism is pathological. This, in itself, constitutes one of the most significant barriers to care. As a result of this labeling, transgender individuals have under-utilized public health and social services. A survey of transgender men and women in San Francisco reported that many in the population are chronically underserved with regard to basic medical and psychological support services. Transgender individuals frequently resort to self-medication with black-market hormones or visit irresponsible practitioners who promote hormone administration and silicone treatments without appropriate medical follow-up. Few resources exist that address their special needs or provide necessary consumer education and regular medical follow-up (Asian AIDS Project, 1995; San Francisco Human Rights Commission, 1994).

Transgender individuals have also been stigmatized by prominent and well-established gay and lesbian movements and organizations which marginalize transgender issues or erroneously conflate sexuality with gender and consequently define transgender concerns as a subset of those confronted by gays and lesbians. In addition to the prejudice experienced because of perceived sexual orientation, many transgender individuals are subject to prejudice based on their gender. For example, mental health providers and researchers have noted that male to female (MtF) transgender participants were not prepared to experience the harassment faced by women. In the female role, these individuals feared more for their physical safety. Not "passing" in the crossgender role and being "read" as transgender further increased the fear of assault (Bockting, et. al., 1998).

Under the guidance of a small group of transgender activists and their supporters, organized efforts to forcefully challenge stigma, discrimination, and prejudice are underway. Members of transgender communities have become increasingly public, and work to dispel myths and prejudices through an increasing number of publications and public appearances. Community-based transgender organizations are receiving increasing attention, and representatives of these organizations continue to work with public health, medical, and mental health researchers to question and refine foundational assumptions about gender definitions and gender-based diagnoses. Transgender celebrities and positive representations of transgender individuals and communities are increasingly prevalent, sparking debates about gender identity necessary to significantly alter prejudice and stigma.

As is the case with other marginalized and oppressed minorities, the HIV/AIDS epidemic has focused attention on transgender health issues and difficulties in accessing care, and begun to bring community advocates, medical providers, researchers, legal scholars, legislators, and policy makers together to address the needs of transgender communities. Still, health concerns of HIV positive transgender individuals remain largely unexplored at many of the nation’s leading AIDS service organizations, including those based in or primarily serving gay communities.

One of the most significant barriers is that most health professionals lack the necessary knowledge about transgender identity and sexuality to respond adequately to their patients. As a result, patients are required to educate health care providers repeatedly about transgender issues (Bockting, et. al., 1998). These experiences underscore the stigma and contribute to some of the most significant health deficits faced by transgender individuals – feelings of shame, low self-esteem, isolation, loneliness, anxiety and depression (Prieur, 1990). While problems in themselves, these feelings also increase risk for other major health problems, including HIV infection and suicide.

Social and economic marginalization frequently accompanies the transgender experience. Rejected by family and community, with reduced educational and employment opportunities because of the harassment faced in both settings, transgender men and women are commonly subject to discrimination, homelessness, unemployment, and poverty. Many are unable to afford basic medical and mental health services. Furthermore, a disproportionate number of these individuals are people of color, HIV-positive, and/or youth, thereby increasing the likelihood they are socially and medically underserved (Israel & Tarver, 1997). At times, these hardships force individuals into trading sex for services. Obtaining appropriate care is particularly difficult under these circumstances.

Although a large body of evidence exists demonstrating that treatments are available and successful in many cases, public and private insurers specifically exclude coverage for treatment on the grounds that the treatments are either cosmetic or experimental (Seil, 1996). Transgender individuals, even as they are diagnosed with mental disorders, are denied legal protection such a diagnosis ordinarily provides. Although Gender Identity Disorder (GID) "causes clinically significant distress or impairment in social, occupational, or other important areas of functioning" (American Psychiatric Association 1994), individuals with GID are specifically excluded from the Americans with Disabilities Act and thus do not receive its benefits or protections (Gordon, 1991; Israel & Tarver, 1997).

As a result of the financial barriers to legitimate quality medical care, many transgender individuals, particularly those within minority subpopulations, are targeted as desperate victims by unscrupulous care providers who offer hormones, silicone injections (illegal in the U. S.), aesthetic and genital reassignment surgeries or other services without informed consent, appropriate medical administration, or follow-up (Israel & Tarver, 1997). Effective and safe transgender hormone administration can be provided only by a licensed physician (M.D. or D.O.), with individual treatment plans based on routine testing of hormone levels and knowledge of the patient’s health history and predisposing physical characteristics. For many, this basic service is unaffordable.

U.S. attitudes toward removal of financial barriers to transgender care are different from those found in a number of other countries, such as Great Britain, Holland and Australia. In these countries, genital reassignment surgery is paid for by national health insurance and is accepted as a standard of care. U.S. cases dealing with Medicaid funding for genital reassignment surgery have usually ruled solely on the question of whether a state can promulgate a regulation absolutely excluding such funding. Judgements for plaintiffs in such cases have not guaranteed funding for future applicants but merely ensured that the requests will be evaluated on an individual basis (Gordon, 1991). The final judgements are not always favorable; moreover, the costly legal process is beyond the means of the vast majority of the transgender population.

Various locations in the U.S. have begun to adopt a model of care pioneered outside the country, where transsexual individuals are diagnosed and treated by specialists, sometimes in multidisciplinary gender teams and according to established standards of care for hormone treatment and surgery (Cohen-Kettenis & Gooren, 1999). As documented in section IV. G., below, a large body of research has demonstrated that genital reassignment surgery is effective for many transsexuals. These findings are reflected in the changing attitudes of health professionals toward surgery (Gordon, 1991). New perspectives among health care providers combined with improved procedures for surgical outcomes argue against the definition of the procedure as either experimental or cosmetic.

IV. C. Special Populations

The health-related needs and experiences of transgender youth have received very little examination. For the most part, the literature does not distinguish these populations from lesbian and gay youth. Scientific investigation of the specific and unique needs of transgender youth is urgently needed.

Most transgender youth remain invisible. With few exceptions, they do not publicly crossdress or appear feminized or masculinized. They have a significant investment in appearing indistinguishable from their non-transgender peers in order to avoid physical and emotional abuse. The isolation keeps most youth from essential mental health and medical care until a crisis occurs. The consequences of such crises are enormous because they are not just medically destabilizing, but usually entail rejection and further isolation from family and peers. As a result of family and social abandonment, many transgender youth encounter victimization through homelessness, drug use, and prostitution (Israel & Tarver, 1997; Kral, et. al., 1997).

Once in care, transgender youth likely experience unpleasantness similar to or greater than that experienced by lesbian and gay youth. In an unpublished 1995 study conducted by Blanco (reported in Kreiss & Patterson, 1997) LGBT youth in Washington State were asked to assess their access to health care and the quality of care they received. The study found that 66% of youth stated that their health provider had never brought up issues of sexual orientation. Many received inappropriate treatment and health education based on their provider’s heterosexual assumptions and ignorance of their true sexual orientation. We can only speculate that gender identity issues were either confused with sexual orientation or not noticed at all.

Transgender advocates have reported that transgender individuals who are incarcerated or institutionalized are victimized (including rape and physical abuse) as a result of their gender-identity issues. Failure to protect these individuals has resulted in incidents of suicide and homicide. Victimization also included withholding hormones from individuals. Withdrawal of hormones has been linked to an increased risk of self-mutilation (auto-castration), heightened clinical depression, behavioral difficulties, illegal drug use, and suicide attempts (Israel & Tarver, 1997).

As with the general population, transgender persons of color are more likely to be economically disadvantaged and face disproportionately higher rates of victimization, unemployment, substance abuse, HIV infection, prostitution, and other difficulties. Transgender persons of color also report a loss of community identity when their gender identity becomes known. In African American, Asian, Pacific Islander, or Latin American contexts, for example, heterosexual males and females commonly stereotype gay males and lesbians as no longer a part of their ethnic community because they assume that all people of color are or should be heterosexual. This ostracism carries over to transgender individuals (Israel & Tarver, 1997).

IV. D. Medical and Research Institutions

Probably as a result of the view that transgender individuals are pathological and the concomitant medicalization that proceeds from a diagnosis of adult GID, the majority of research in the field has been conducted on transsexuals with a particular emphasis on surgical interventions. Within this group, the vast majority of studies have been conducted on MtF transitions with little written about FtM procedures and outcomes. Lack of literature specifying issues faced by other transgender subpopulations – such as transvestites and androgynes – suggests that providers may be ignorant of the health concerns of these individuals. Furthermore, the literature emphasizes genital reassignment surgery and pays little attention to individuals who select hormone treatment but do not also elect surgery.

IV. E. Standards of Care

Licensure in a traditional health field is not enough to guarantee proficiency in gender issues (Israel & Tarver, 1997). It is necessary to include a Gender Specialist as part of a treatment team, particularly throughout the process leading up to and following from genital reassignment surgery. As with definitions and cultural understandings of trangender issues, standards of care – and efforts to assess them – are evolving.

The Harry Benjamin International Gender Dysphoria Association (HBIGDA) developed guidelines for caregivers and patients regarding the steps of transsexual transition in 1979. Most recently revised in 1990, these guidelines are called the Standards of Care (SOC) and are widely known by transsexual individuals and generally followed by workers in this field. These guidelines provide clinicians with a method for selecting those transsexuals most likely to benefit from genital reassignment surgery. The standards reflect the consensus of the medical community experts in this field. However, a recent survey of the policies and criteria currently used by European and North American gender-identity clinics that use the HBIGDA standards of care found that only nine out of 19 clinics that responded to the survey adhered to the HBIGDA standards entirely (Petersen & Dickey, 1995). As a result, post-surgical care is unevenly provided. This includes pain management and care for other complications such as infection and excessive hemorrhaging.

Discussion about whether to surgically alter ambiguous genitalia is an increasingly controversial issue and represents the most acute conflict between our cultural and biological definitions of gender. Intersexuality – biologically variant sexuality known more commonly as hermaphrodism – disturbs the conventional distinction between male and female persons so fundamental to self-identification and social ideology, particularly in the West (Chase, 1998). This dynamic forces consideration of institutional practices and behaviors that have been seen as problematic by many transgender individuals.

It is conservatively estimated that one in 2,000 newborns are found to have ambiguous external genitalia, and that 100 to 200 pediatric surgical reassignments are performed in the U.S. annually. Globally, thousands of these procedures have been done since the practice was institutionalized in the 1950s with the intention of precluding the traumatizing stigma of not having a clearly defined male or female physiognomy (Kipnis & Diamond, 1998; Preves, 1998). It has been standard practice to recommend surgery for infants with ambiguous genitalia. The parents of these patients are told to raise them without ambiguity. As a result, many adults who have had these operations in infancy have never been candidly informed of their medical histories (Elliott, 1998; Kipnis & Diamond, 1998). The majority of intersex conditions are found to be physiologically benign. Some conditions do require surgical or hormonal intervention for the physical health of the individual such as where the elimination of urine or feces is difficult due to the intersexed nature of the body (Diamond & Sigmundson, 1997; Kessler, 1998).

Kipnis and Diamond (1998) have identified a number of limitations to clinically managing intersexuality. First, the line that decisively and non-arbitrarily separates male from female is unclear. Second, the psychosocial development of a gender is not alterable in the same manner as the physical genitalia. Third, it is not possible to predict confidently the gender that an intersexed newborn will settle into during adulthood.

In a recent and ongoing study, Reiner (1997) tracked six boys who had lost their penises in infancy by accident or through surgery and were being reared as girls. These children behaved more like boys than girls and, in two cases, the children autonomously changed gender and assumed male roles (Reiner, 1997). In follow-up interviews with a cohort of intersex adults, Preves (1998) found that many who had genital surgeries emphasized that the very operations that were intended to assuage feelings of difference only served to highlight their stigma. Far more extensive follow-up studies are required to determine the clinical benefits and harms of infant surgery on intersexed individuals. The ethical implication of concealing information from the patient also requires extensive further analysis (Nelson, 1998). A review of current standards of care for intersexed individuals is urgently required.

IV. F. Mental Health Needs

Gender dysphoria, feelings of incongruity with the physical gender assigned to one at birth, is endemic to trangendered individuals. Individuals experiencing concerns about their gender identity may suffer social isolation, emotional anguish, distorted self-image, and even misdiagnosis by health professionals. Once an individual has self-identified transitional goals or has established a self-defined transgender identity, she or he is no longer considered to be gender dysphoric.

The Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) offers three diagnostic categories in the area of Gender Identity Disorders: Gender Identity Disorder, Transvestic Fetishism and gender identity confusions that occasionally occur during Schizophrenic episodes (American Psychiatric Association 1994). Revisions in the original definitions of GID in the DSM-III (American Psychiatric Association, 1980) demonstrate a move toward defining only the emotionally distressed subgroup of transgender individuals as exhibiting a bona fide mental disorder, while partially validating the fact that many transgender persons do not evidence gender-associated confusion or emotional distress.

Nevertheless, the diagnoses of Gender Identity Disorders and Transvestic Fetishism are still considered pejorative by many in the transgender community, as well as problematic for gay men and lesbians. Both see diagnosis of GID as potentially stigmatizing of non-normative gender identity behavior, in much the same way that homosexuality was pathologized prior to its removal from the list of mental illnesses by the American Psychiatric Association.

While research is vastly inadequate, transgender persons experience mental health difficulties that are not much different from anyone who experiences major life changes, relationship difficulties, chronic medical disorders, or significant discrimination on the basis of a minority background. The expression of mental health issues is likely to include stresses that result from gender-related issues (Israel & Tarver, 1997). Mental health problems include adjustment disorders, anxiety disorders, post-traumatic stress disorders, and depression.

Some studies have established incidence of severe personality disorders, psychosis and mental illness among transgender persons above the norm. The methodologies – particularly the lack of controlled comparisons – are questionable. These studies also tend to underestimate the impact of societal stressors such as sexual assault, violence, abandonment by family and other support persons, substance abuse, loss of opportunities, and legal harassment. Much more rigorous research is required to determine the actual incidence of mental health disorders in these populations.

Israel and Tarver (1997) suggest that transgender persons frequently avoid treatment for depression because it is widely believed that in order for treatment to be effective, both gender identity and depression issues must be addressed at the same time. The concern is that the gender issues will be assumed to play a causal role and will be judged negatively. Because of these and other factors, hormone-associated depression may be underdiagnosed.

A specific subset of gender-related depression and disturbance derives from the practice of assigning gender in utero or at birth. Some individuals are never able to find a sense of gender congruency after having been surgically altered without appropriate presurgical counseling or informed consent. This is particularly true in situations where parents made final surgical decisions without the child’s understanding the full consequences of such procedures. Intersex or hermaphroditic individuals who are unable to adjust in an assigned gender role are often mistakenly classified by physicians as having a severe psychiatric disorder and thus unnecessarily victimized (Israel & Tarver 1997).

The most severe transgender-related crisis is that of an attempted, or completed, auto-castration or genital mutilation. This occurrence is most common among transsexuals and transgenderists, although cross-dressers have acted it out as well. The stronger the transgender identity, the higher the possibility that this may be attempted by those who do not perceive the availability of support services or the existence of viable solutions to their circumstances. A study of a cohort of transgender individuals who applied for services at gender identity clinics reports that breast mutilation was attempted by 2.4% of the females, and genital mutilation by 9.4% or the males (Dixen, et. al., 1984). Similar studies also report that a history of suicide attempt was given by 19%-25% of participants, a history of drug or alcohol abuse was obtained from 30%, and a history of psychiatric hospitalization from 11% (Buhrich, 1981; Dixen, et. al., 1984; Kuiper & Cohen-Kettinis, 1988; Verschoor & Poortinga, 1988).

Psychological evaluation has shown that cross-gender reassignment treatment increases the wellbeing of both MtF and FtM transsexuals (Kuiper & Cohnen-Kettenis, 1988), but there are still a high number of deaths due to suicide in MtF transsexuals (van Kesteren, et. al., 1997). A very small number of transgender individuals have been diagnosed with eating disorders (Surgenor & Fear, 1998).

The impact on the partners of transgender individuals has received almost no attention. There is evidence that changes related to gender identity do impact partners and could unsettle relationships causing significant emotional stress to both individuals (Doctor & Prince, 1997).

When initially introduced to their partner’s transgender status, spouses, partners, and significant others often question their own sexual orientation, at the same time expressing the same concerns with respect to their partners.

Victimization is widely recognized in the literature, with the implication that virtually every transgender person is likely to experience some form of victimization as a direct result of his or her transgender identity or presentation. The precise prevalence of victimization in the lives of transgender individuals is not reported in any of the literature reviewed here, and the link between these experiences and mental health disorders such as Post Traumatic Stress Disorder have not been adequately recorded. This work is urgently needed.

Victimization includes subtle forms of harassment and discrimination as well as blatant verbal, physical, and sexual assault. The last may include violent beatings, rape, and even death. The majority of assaults against transgender persons never become matters of record. This situation exists because transgender individuals have little societal support or access to legal recourse, particularly where they are disenfranchised. Not only is sexual violence common against MtF transgender individuals, the fact that incidents are rarely prosecuted indicates how unimportant this crime is thought to be (when it happens to a transgender person) by the criminal justice system (Feinberg, 1996).

Israel and Tarver (1997) report that victimization is common in institutional settings. Rape systematically occurs in jails and prisons. Guards and administrators either ignore these events or even encourage this activity because keeping violent criminals in "relationships" tends to make populations more manageable. As a result, transgender individuals frequently become victimized by aggressive prisoners or are pressured into providing sexual services for one or more individuals. Rape under similar circumstances is also prevalent in institutionalized mental health populations.

During puberty a young person is likely to ask questions about gender and sexuality. Most mental health professionals are not familiar with gender issues and are likely to confuse these with questions of sexual identity or sexual orientation thereby misdiagnosing the youth. It must be noted that gender and sexuality are separate identity issues.

Almost all of the research on transgender youth combines them with gay and lesbian youth in the catchall nomenclature, lesbian-gay-bisexual-transgender (LGBT). Consequently, little is known about the prevalence of mental health disorders among transgender youth. In their small study of psychological functioning in 29 adolescent transsexuals, Cohen et. al. (1997) found little evidence that as a group adolescent transsexuals evidence greater mental health disorders than the norm. They did, however, find some evidence of lower levels of reality testing in adolescent transsexuals than in non-transsexuals. The etiology or implications of this are unclear. The overall findings support the idea that there are psychological differences between adolescent and adult transsexuals who are anticipating genital reassignment surgery.

In addition, it is assumed that like lesbian, gay, and bisexual youth, transgender youth are at increased risk for low self-esteem, depression, suicide (Remafedi, Farrow & Deisher, 1991), substance abuse, school problems, family rejection and discord, running away, homelessness, and prostitution (Kruks, 1991; Remafedi, 1990; Savin-Williams, 1994). Transgender youth are likely to be the victims of social stigma, hatred, hostility, isolation, and alienation, as are gay and lesbian youth (American Academy of Pediatrics Committee on Adolescence 1993), and to experience higher rates of substance use and suicide ideation than their heterosexual, gender-conforming peers.

Finally, the apparent greater extent of psychopathology seen in adult transsexuals may be the result of increasing and enduring stress related to coping with gender identity conflicts. These mental health findings, along with the aesthetic benefits attendant on early genital reassignment surgery, argue in favor for at least beginning hormone treatment during puberty. Transsexuals who have been treated early at the Amsterdam Gender Clinic pass very easily as a member of the opposite gender (Cohen-Kettenis & Goozen, 1997). As a result, these individuals may suffer less stress as adults. Further research on the relationship between the age at which an individual requests genital reassignment surgery and long-term outcomes would be extremely beneficial to both consumers and medical providers.

IV. G. Genital Reassignment Surgery

Transsexualism is relatively rare. In DSM-IV-R, prevalence is reported as one per 30,000 for males and one per 100,000 for females (American Psychiatric Association, 1987). In the Netherlands in 1986, it was reported as one per 18,000 for males and one per 54,000 for females (Eklund, et. al. 1988). The discrepancy may, in part, be because many transsexuals are drawn to the Netherlands for treatment because care is covered by national health insurance.

The term "transsexualism" did not appear until 1949. No official diagnostic criteria for transsexualism appeared until the DSM-III (American Psychiatric Association 1980). Its inclusion under psychosexual disorders along with ego-dystonic homosexuality and the paraphilias reflected a lack of knowledge of the etiology of the disorder, as well as its course and impact on society. In DSM-III-R (American Psychiatric Association 1987), patients were subtyped as being homosexual, heterosexual, and asexual, on the basis of genetic or anatomic gender. This strictly biological viewpoint is at odds with the patient’s self-identity. Transsexuals found the labels offensive and at variance with their self-concepts (Bradley, et. al., 1991).

After the well-publicized case of Christine Jorgensen in 1952 revealed that genital reassignment surgery was possible, patients started requesting the procedure. By conservative estimates, since the late 1970s approximately 10,000 genital reassignment surgeries have been successfully performed worldwide. This figure includes approximately 4,500-6,000 surgeries performed on U.S. residents (Landen, et. al., 1996; Midence & Hargreaves, 1997).

Because reimbursement for and performance of genital reassignment surgery has been a subject of controversy, evaluation of its effectiveness is important. Empirical support confirming the expectation that better post-treatment outcomes result from the HBIGDA guidelines is badly needed. Unfortunately, studies evaluating the indispensability of components of the currently employed procedures are nonexistent (Cohen-Kettenis & Gooren, 1999; Midence & Hargreaves, 1997). Since the 1960s, 70 follow-up studies have been published on the transgender patients’ satisfaction with their genital reassignment, and all but one concluded that the surgery was satisfactory (Seil, 1996). Psychological functioning as measured on the MMPI has been shown to improve after surgery (Fleming, et. al., 1981; Mate-Kole, et. al., 1990).

Complications for transgender women with biological risk factors or for those who have had excessive estrogen therapy include the sequelae of thromboembolism (blood clots in the legs, lungs, eyes, brain – i.e., strokes – or other organs). In their study of mortality and morbidity in transsexual subjects, van Kesteren et. al. (1997) recorded a 20-fold increase in the occurrence of venous thromboembolism than occurs in the general population. Transdermal oestradiol administration may considerably reduce the risk of venous thromboembolism (van Kesteren, et. al., 1997).

There are reports in the literature of a number of MtF transsexuals developing breast carcinoma upon estrogen administration (Ganley & Taylor, 1995; Kirk, 1996). Other complications include: liver disease; heart disease, including myocardial infarction; high blood pressure; sterility; mood changes; and decreased sexual desire (Kirk, 1996). Smoking increases the risk of blood clots with estrogen therapy, particularly after age 40. These complications are also evident in the use of progesterones.

Side effects for other hormone therapies such as Spironolactone, Flutamide, and Cyproterone Acetate include weakness; fatigue; decreased appetite; weight gain; headaches; and gastrointestinal disturbances.

The major risks associated with excessive doses of testosterone are: increased cholesterol and higher lipid levels; heart disease, including myocardial infraction; mood changes; male pattern baldness; acne; and cessation of menses within three months of continual usage. Smoking increases the risk of coronary heart disease in individuals using testosterone (Israel & Tarver, 1997).

IV H. HIV/AIDS

Very limited research is currently available on HIV/AIDS risk and prevalence in transgender populations. The majority of the literature focuses on the impact the AIDS epidemic has had on transsexual sex workers.

Given that a noticeable number of transgender females engage in commercial sex work, as a group they are considered at increased risk for infection and transmission of HIV and STDs. As noted previously, the cost of genital reassignment surgery is very high and rarely covered by insurance, so commercial sex work is one of the means of obtaining the money for pre-operative transsexuals (Pang, et. al., 1994).

Increased prevalence rates of HIV (Elifson, et. al., 1993; Modan, et. al., 1992), syphilis, and hepatitis (Elifson, et. al., 1993) among transgender commercial sex workers compared to female commercial sex workers may be the result of certain sexual practices specific to transgender individuals. Often beginning their sexual practices prior to genital reassignment surgery, transgender females engage in anal intercourse (Pang, et. al., 1994; Tsoi, 1990). After surgical reassignment, transgender individuals are at risk for contracting HIV/STDs through both vaginal and anal intercourse (Pang, et. al., 1994). Transgender commercial sex workers are at the bottom of the hierarchy of prostitution; that is, they work in the least desirable locations, earn the least money, and are stigmatized or ridiculed by male and female commercial sex workers (Cohen, 1980; Garber, 1992). Therefore, they are inclined to engage in unprotected sex because of client demand, and the prospect of additional money paid for unprotected sex (Asian AIDS Project, 1995; Boles & Elifson, 1994; Gattari, et. al., 1992).

Other common risk factors include multiple sexual partners, irregular condom use, and drug and injecting needle use (Boles and Elifson, 1994; Elifson, et. al., 1993; Galli, et. al., 1991; Gattari, et. al., 1991; Inciardi & Surratt, 1997; Modan, et. al., 1992; Ratnam, 1996; Tirellie, et. al., 1991). A study investigating HIV risk behaviors among 53 "transvestite" commercial sex workers in Atlanta revealed that 68% of the sample were HIV-positive. HIV seroprevalence rates were higher among those transvestites who had engaged in receptive anal sex and used crack cocaine than those who had not (Elifson, et. al., 1993).

One study examining HIV knowledge and risk behavior in a transsexual sample not limited to sex workers reports that 24% of sexually active respondents engaged in receptive anal sex and 19% of those reported condom use; 4% reported drug use (Avery, et. al., 1995, reported in Bockting, et. al., 1998).

Those transgender people receiving hormone therapy who do not have health insurance to cover it and who cannot afford expensive hormone pills, or who have side effects from taking hormone pills, inject hormones (Nemoto, et. al., 1999). One of the factors encouraging needle use and consequently increasing the likelihood of needle sharing to inject hormones and/or illicit drugs is the belief that injecting hormones is more efficacious than taking pills (Nemoto, et. al., 1999).

Focus on transgender populations or their sexual partners is largely absent from most HIV/AIDS interventions, whether for prevention or the provision of care. Local and national transgender organizations have largely failed to publicize HIV-related safe-sex materials or even discuss HIV concerns. Brockting et. al. (1998) report that transgender individuals find that existing prevention education is not inclusive of transgender people and often makes assumptions about sex and gender that are not applicable to their situation. In addition, transgender persons involved in the injection of black market hormones and silicone may be neither identify as drug users nor perceive their behavior as injecting drug use, and hence are unlikely to identify themselves as at risk despite frequent needle sharing. Prevention education needs to target this specific risk and improve access to specialized medical care that promotes responsible use of hormones.

The usual sense of urgency for surgery felt by those waiting for genital reassignment surgery can be heightened by the presence of HIV infection. Research suggests that this group has a greater level of anxiety, hopelessness, and loneliness than matched controls (Kok, et. al., 1990). Patients in this situation may be reluctant to express their true distresses for fear of jeopardizing their chances of surgery (Pang, et. al., 1994).

Providing surgical and other support services for an HIV-positive transgender individual introduces numerous questions and uncertainties, particularly because to date, there have been no recommendations available to provide informed guidance to all concerned. Concerns regarding hormone administration, aesthetic surgery, and genital reassignment surgery are the common denominators that set HIV-positive transgender individuals apart from the remaining HIV-positive population. No other HIV-positive population is subject to so many surgical procedures. Studies on the effects of hormones on the progression of HIV/AIDS are needed, and the risks and potential benefits of genital reassignment surgery need to be clarified. A special task force of the Harry Benjamin International Gender Dysphoria Association has been formed for this purpose (Bockting, et. al., 1998).

Because of a lack of adequate clinical research on the side effects of aesthetic surgical procedures or of genital reassignment surgery on HIV-positive individuals, it is unclear whether to endorse or disqualify individuals from seeking these procedures. Although no formal studies exist at this time, some practitioners believe that hormone administration in HIV-positive transgender individuals may have immune-enhancing effect. Hormone-induced side effects, particularly those affecting psychological well being, may be exacerbated in HIV-positive individuals. There is little information available on the potentially compounded side effects of simultaneously undergoing hormone and HIV drug therapies (Israel & Tarver, 1997).

As with mental health, studies have tended to conflate sexual orientation and gender identity, and HIV/AIDS prevalence is unknown for transgender youth. Given high rates of homelessness and substance use among LGBT youth, high rates of sex work among transgender individuals, and high rates of HIV risk among homeless or runaway LGBT youth, transgender youth are likely to be at greatly elevated risk for HIV infection.

V. A. Overview

The public, government officials, health care providers, and researchers have only recently begun to recognize the many important, yet largely overlooked, links among health, sexual orientations, and transgender identity (Solarz, 1999; Scarce, 1999; Ryan & Futterman, 1998; Joint Policy Committee, APHA, 1998; Council on Scientific Affairs, American Medical Association, 1996). In order to improve the health of the U.S. population as a whole these concerns can no longer be neglected. However, as researchers have begun to hypothesize and study these links, unresolved methodological difficulties have presented serious challenges to the collection of reliable and valid information.

These difficulties include defining, measuring, and sampling individuals using largely critically unexamined and socially constructed categories, as well as sampling rare and hidden populations concerning sensitive topics (Solarz, 1999; Sell, 1997; Gonsiorek, Sell, & Weinrich, 1995). Not surprisingly, these difficulties mirror similar problems that have been recognized and examined related to the classification of people based upon race and ethnicity. To understand the very real needs of individuals as related to their sexual orientations and transgender identity, a similar effort must be made to the one that has examined race and ethnicity.

While there are many challenges surrounding the collection of data concerning sexual orientation and transgender identity, four require immediate attention:

• defining the populations to be studied
• constructing valid and reliable measures of sexual orientations and transgender identity representative of these definitions;
• sampling rare and hidden populations; and
• sampling and studying sensitive topics.

These four areas are often ignored or left unresolved, and the resultant research has therefore shown significant variation in how sexual orientations and transgender identity are defined, measured, and sampled (Israel & Tarver, 1997; Shively, Jones, & DeCecco, 1985; Sell & Petrulio, 1996). For example, a review of published public health research articles that sampled homosexuals, lesbians, gays, and/or bisexuals between 1990 and 1992, found that research publications: rarely provided a conceptual definition of the population they sampled; used a range of incompatible methods and measures of sexual orientation to identify and select participants; sampled from settings representative of dramatically different populations; and rarely used probability sampling (Sell & Petrulio, 1996). This is not to say that none of the studies reviewed were methodologically sound, nor that the studies did not produce important results. To better understand and monitor the public health concerns of lesbian, gay, bisexual, and transgender people, steps must be taken to standardize definitions, measures, and methods. Each of the above four topics is briefly reviewed here.

V. B. Defining the Populations

Different definitions and measures of sexual orientations have been proposed and used to develop study populations since the 1860s when sexual orientations first gained widespread research interest (Sell, 1997). In fact, many different terms were used to label sexual orientations before the terms "heterosexual," "homosexual," "bisexual," "gay," and "lesbian" slowly came into widespread use from the 1920s through the 1960s. Unfortunately there is still no general consensus on the definitions of these terms, although each includes components of at least one of three dimensions: 1) sexual orientation identity; 2) sexual behavior; and/or 3) sexual attraction (Laumann, et. al., 1994). For example, one study might define sexual orientation as a form of identity (as self-identified heterosexual, homosexual, bisexual, gay, or lesbian), while another defines it as gender choice in sexual partners, and yet another as the gender of those to whom one is sexually attracted (Sell, 1997).

Within each of the above three dimensions there is even further variation. One researcher might define sexual behavior as any relationship between two people resulting in sexual arousal (not necessarily including physical contact), while another researcher’s definition may specify physical contact resulting in orgasm. Consensus is required to develop valid and reliable measures.

Recent national studies estimating the percentage of the population that falls into each of the three broad dimensions of identity, behavior, and attraction show that 1 to 4 % of the population identifies as lesbian or gay, 2 to 6 % of the population reports some same-sex behavior in the previous five years, and up to 21 % of the population reports same-sex attraction at least once in adulthood (Sell, Wells & Wypij, 1995; Laumann, et. al., 1994; Billy, et. al., 1993; Smith, 1991; Harry, 1990; Fay, et. al., 1989). Therefore, depending upon how it is defined and measured, 1 to 21 % of the population could be classified as lesbian or gay to some degree, with the remainder classified as bisexual or heterosexual to some degree.

Defining the term "transgender" is even more treacherous and less examined than sexual orientation. Israel and Gianni (1997) point out that "the term ‘transgender’ has become the word of choice for both professionals and consumers when referring to individuals or the community as a whole." The term itself, however, encompasses a number of populations and communities including transsexuals, cross-dressers, and the intersexed. See section IV, Transgender Health Concerns, for full definitions of the sub-groups within the transgender population. Each sub-population presents definitional challenges.

V. C. Measuring the Populations

There is also much confusion about measures of sexual orientation and transgender identity. This is not surprising considering that valid measures should first and foremost be based upon conceptual definitions of the populations in question and no such definitive definitions exist (Streiner & Norman, 1989; Sudman, 1976).

Existing measures of sexual orientation range in complexity from simple dichotomous measures in which the subject reports they are or are not heterosexual or homosexual, to more complex scales as developed by Kinsey et. al. (1948); Klein et. al. (1985); Shively and DeCecco (1977); and Sell (1996) (for a description of each see Sell, 1997). There exists no consensus and virtually no literature discussing when and where each of these measures should be used, therefore, their use and value for research is uncertain at best. Only common sense and the health question before the researcher provides any guidance as to which measure, if any, should be chosen. For example, questions that pertain to biomedical pathogenesis, disease prevalence, health care access, or prevention efforts may all require different dimensions of sexual orientation to be measured. For instance, self-identification may be best for studying access to health care, sexual behavior may be best for studying STDs, while sexual attraction may be best for examining some mental health related issues. How well each of the existing measures captures these dimensions of sexual orientation must be taken into consideration.

Measures to identify transgender individuals are, not surprisingly, less well described and developed. While measures of sexual orientation generally assess identity, behavior and/or sexual attractions, the term "transgender" has been constructed and is usually defined as a form of identity. That is, an individual is transgender if they choose to identify as such. Self-identification as transgender then generally serves as a marker for one of the related communities, such as transsexual, crossdresser or intersexed individual. Of course, most researchers wanting to examine the health of transgender populations would want to further identify individuals into each of these categories, or perhaps single out one of these specific populations for study (Israel & Tarver, 1997).

Measures of sexual orientation and transgender identity must also take into account racial, ethnic, and age differences among research participants, which may affect measure validity and reliability (Solarz, 1999; Ryan & Futterman, 1998). Substantial variation exists across racial and ethnic groups concerning the social acceptability of exact orientations and identities, and consequently the reporting and understanding of these constructs across communities. The terminology to discuss human interactions similarly varies and must be taken into consideration when developing and choosing measures of sexual orientation and transgender identity.

V. D. Sampling Rare Populations

In addition to the above research challenges, researchers studying relationships among health, sexual orientations, and transgender identity are faced with the difficulty of sampling and studying rare populations. Rare populations, often geographically dispersed and hidden, present certain methodological and financial challenges to researchers wanting to construct samples, and in particular representative samples (Sell, Wells & Wypij, 1995; Martin & Dean, 1990; Sudman, Sirken & Cowan, 1988; Sudman, 1976). To a larger extent, unlike in the areas of defining and measuring the populations as discussed above, there is research outside the field of sexual orientations and transgender identity that can be examined, modified, and applied to the construction of research samples where appropriate and helpful (Lee, 1993; Renzeti & Lee, 1993).

The methods that have been used most often to study LGBT populations include:

• List Sampling — when a sample is derived from a list such as members of an LGBT organization. Sometimes the entire list is used for this selection of subjects, while at other times the list may be sampled.
• Multipurpose Sampling — when a sample constructed for another purpose is expanded to study issues related to LGBT health. For example, a study designed to examine women and cancer, by stratifying the sample by sexual orientation, can be expanded to examine the relative risks of various cancers by sexual orientation.
• Screening Sampling — when a larger general population sample is screened in order to identify LGBT people for inclusion into the research study. Because the populations are rare, it is sometimes necessary to screen study participants by sexual orientation or transgender identity using a pre-screening to determine inclusion.
• Network or Snowball Sampling — when the researcher identifies members of the populations of interest or key informants who then identify other members of the populations who are consequently contacted and included in the study. These additional individuals can then be asked to identify additional participants and so forth, resulting in a "snowballing" effect.
• Outcropping Sampling — when individuals of rare populations are identified and recruited into the sample at locations frequented by the population. For example, lesbian and gay pride events, lesbian and gay bars, or lesbian and gay neighborhoods.
• Advertising Sampling — when the researcher places advertisements to recruit subjects. The advertisements can be placed in newspapers and/or periodicals catering to the population, or can be posted in locations frequented by the population. For example, transgender subjects may be recruited through postings on transgender-related web pages or publications.
• Servicing Sampling — when a service is offered to the study subject as a method of recruitment. For example, transsexuals may be recruited by offering free hormone-replacement counseling, or gay men may be recruited by offering free or reduced price hepatitis vaccinations.

The above methods of course can and often are mixed and matched to construct samples. Each of the methods introduces biases, too numerous to discuss here, into the study that must be addressed when interpreting findings. Despite these biases, however, the above methods are generally used because they are feasible considering the limited resources generally available to study these populations.

V. E. Sampling Sensitive Topics

The final concern to address when studying LGBT health is the sampling and studying of "sensitive" topics. Sieber and Stanley (1988) define sensitive research as "studies in which there are potential consequences or implications, either directly for the participants in the research or for the class of individuals represented by the research." Research on LGBT health, by any definition, must be considered sensitive.

The revelation of sexual orientation or transgender identity by study subjects can be difficult because of cultural taboos and because some subjects may have unresolved issues relating to their sexual orientation or gender identity (Ryan & Futterman, 1998; Sell, Wells & Wypij, 1995). The subjects may also place themselves at risk for violence and discrimination if responses are not kept confidential (Lee, 1993). The revelation of sexual orientation by study participants may also imply the conduct by respondents of certain sexual behaviors classified as criminal in some jurisdictions in the U.S. (Hunter, Michaelson & Stoddard, 1992). Research concerning how to conduct studies on sensitive topics must therefore be refined and examined in the context of research on LGBT health (Lee, 1993; Renzeti & Lee, 1993). In fact, the sensitive nature of LGBT health affects the entire research process from the formulation of the research question, to the design and conduct of the study, to the publication and dissemination of the results. These problems, in addition to presenting methodological challenges, can present ethical, political, and legal challenges that must be addressed by the researcher. This document, however, only attempts to address some of the methodological challenges.

The following is a brief review of methods that have been shown to successfully assist the conduct of research on sensitive topics and specifically the development of survey questions:

• Loading questions — Loading questions refers to the process of biasing a question to influence the subject’s comfort with providing a response. For example, a question can be phrased to imply, using an authoritative source, that a certain behavior is common or socially acceptable. The question can also be worded to assume that a respondent has participated in specific behaviors forcing the subject to respond in the negative if they have not (Lee, 1993).
• Familiar words — Using familiar words has been shown to assist in understanding and increasing comfort with questions. Familiar words are those commonly used by the subject or the population to describe the sensitive topic being explored by the researcher (Bradburn & Sudman, 1979).
• Long questions — Short questions are almost always preferable to long questions, however, long questions can be used to provide memory clues to the respondent or to give the respondent more time to recall past experiences on sensitive topics (Bradburn & Sudman, 1979).
• Embedded questions — Questions that are sensitive can be embedded into the questionnaire in such a way that it will decrease any threats they may pose. For example, a question about same sex sexual behavior is less threatening in a survey concerning sexual behavior in general. Because questions about past behavior or practice tend to be less threatening than current behavior or practice, questions about the past can be asked first. Questions concerning the present would then only be asked if a respondent reports "ever" expressing the behavior (Lee, 1993).

In addition to paying particular attention to the wording and placement of questions, researchers can modify other aspects of the research process to better examine sensitive topics. The following are some of the techniques described elsewhere that can be used: randomized response; nominative techniques; and microaggregation techniques (Lee, 1993; Duffy & Waterton, 1984; Bradburn & Sudman, 1979; Boruch & Cecil, 1979).

Finally, when studying sensitive topics, assuring research subjects that their responses will be kept confidential can improve both response rates and the validity of responses. Confidentiality assists with the research process by convincing respondents that they can trust the researcher. The process of assuring confidentiality can be complex, however every researcher studying LGBT health should be aware of procedures to do so and must take them seriously (Boruch & Cecil, 1979).

V. F. Summary

Despite the challenges of defining, measuring, and sampling sexual orientation and transgender identity discussed above, researchers are forging ahead with studies that provide important information concerning the links between health, sexual orientation, and transgender identity as well as providing valuable insights into the conduct of such research (Israel & Tarver, 1997; Meyer & Colten, 1999; Faulkner & Cranston, 1998; Ramefedi, et. al., 1998; Binson, et. al., 1995). The most impressive work has perhaps been done in conjunction with the Nurse’s Health Study and the Women’s Health Initiative, both of which have now included items of sexual orientation in their data collection (Solarz, 1999). The field experience of these researchers should be examined to provide a framework for the conduct of future studies and to begin to resolve the challenges presented in this paper.