Gynecologic Issues for Lesbians

Both health care professionals and the medical research establishment have historically ignored the so-called sexual minorities, including lesbians. Much of what was known was derived from anecdotal experience and volunteer surveys.^{1, 2} Over the last decade, empirical studies with more methodological rigor have begun to describe the health status of lesbians more accurately.

As a basic principle, most health concerns of lesbians are no different from those of heterosexual women. However, there are unique factors in four areas. First, the sexual behaviors of the lesbian include all those available to the heterosexual woman except for penile–vaginal intercourse, and she needs to know if these behaviors place her at risk for disease. This question is most urgent regarding human immunodeficiency virus (HIV) transmission risk. Second, if she wants to become a parent, she will have a more difficult time accessing the usual options for achieving that goal. Third, while most research in lesbian health has focused on gynecologic issues, a broader area of research has begun to look at primary care issues among lesbians including risks for chronic illness, mental health issues, etc. Finally, lesbians live in a unique psycho-social-economic milieu.³ There is the possibility that each new clinician from whom she seeks care will be indifferent or hostile simply because of her sexuality; her access to health care and insurance may be limited if partner’s benefits are not available; sources of health care may present their services in a way that is not welcoming or effective for lesbians; and lesbians may view the stages of reproductive life differently from heterosexual women, which may affect the preventive choices they make.

Prevalence

A person’s sexual orientation and preferences are often considered to be unchanging over time, but this is often not the case. Men and women who may identify and behave as heterosexuals early in their lives may later self-identify as gay or lesbian, and a similar switch can occur from early homosexual to exclusively heterosexual behavior; persons who are bisexual may have a much more varied behavioral and relationship history.⁴ Thus, the notion of a stable 'prevalence' for lesbians is of questionable meaning. Realizing the limitations of this concept, the survey by Laumann and colleagues⁵ found that between 1% and 4% of adult women in the United States identify themselves as lesbians.

Etiology

The etiology of lesbianism is not known, just as the etiology of heterosexual identity is not understood. Genetic, prenatal, hormonal, and early childhood influences each may have a role. Child abuse and emotionally distant parents have not been found to play a role.

Sexual Practices

Except for activities that require a penis, lesbians have available to them the same potential repertoire of sexual practices as heterosexual couples. The most frequent activities appear to be mutual masturbation and oral–vaginal sex.^{5, 6, 7} The use of sex toys, including vibrators and dildos is common. The prevalence of anal contact varies widely among studies, with a prevalence of 10% in the more recent Diamont survey.⁷

Compared to gay men and heterosexual couples, lesbians as a group appear to have less frequent sex, and fewer lifetime partners.

Research Methodological Issues

Before discussing the various specific health issues, it is important to recognize the challenges inherent in doing research among stigmatized, minority populations such as lesbians and other sexual minorities. One of the major goals of the current research agenda is identifying methods that will overcome these difficulties. The Institute of Medicine published a landmark report in 1999 that comprehensively details these issues, and that calls for more scientific investigation in these areas.⁸

The primary challenge is the identification of unbiased, representative samples. The randomly selected population-based sample, the gold standard of epidemiologic research, has been considered impossible to obtain. Consequently, most investigators have used volunteer or convenience samples, despite the unavoidable bias this introduces. The use of convenience samples is preferable to the use of volunteers, and there are several large surveys using this type of sampling.^{9, 10} This type of study is valuable for generating but not testing specific hypotheses. While larger samples are preferable, size alone does not overcome the inevitable biases. Investigators are therefore obligated to describe the demographic characteristics of the sample, and to avoid overgeneralizing their conclusions.

The assumption that women will not willingly disclose their sexual orientation to investigators has been tested in a small number of studies in which study participants have been asked to self-identify. In one large clinical trial of coronary heart disease risk in young adult men and women, 12% reported having had at least one same-sex partner, suggesting that people will provide this type of information in surveys.¹¹ Among 96,000 women between the ages of 50 and 79 enrolled in the National Institutes of Health (NIH)-funded Women’s Health Initiative, 0.6% identified as lesbian and 0.7% as bisexuals, confirming that older women will respond to these questions as well.¹² In a pilot study within the Nurse's Health Study cohort, fewer than 1% of the sample declined to answer a sexual orientation question.¹³ Standard sampling methods such as random digit dialing and household surveys have been shown to be effective in identifying sexual minorities, so that the prospect of gaining representative information is now a reality.^{14, 15, 16}

The disorders of most interest are those for which sexual behavior is related to risk: vaginal infections, sexually transmitted diseases (STDs), cervical neoplasia, and HIV infection.¹⁷ The prevalence of sexually transmitted infections in a group of lesbians will generally reflect the rate of bisexual activity, and the rate of nonmonogamy among women in the group. A majority of lesbians have a past history of heterosexual activity, and many have used oral contraceptives and have been pregnant.^{9, 18} Although the medical literature contains little information based on the direct study of lesbians, reasonable clinical decisions can be made by combining this limited empirical evidence with theoretical considerations and with extrapolations of information about female-to-male transmission. The recommendations in this chapter reflect this approach.

Vaginitis

Bacterial Vaginosis

This infection is common in lesbians, with some studies showing rates of 30% or more among women seeking routine health care. Lesbians who seek care at STD clinics have been found to have approximately twice the rate of bacterial vaginosis compared to heterosexual women attending the same clinics.^{19, 20} There continues to be controversy as to whether bacterial vaginosis is sexually transmitted. In one study of monogamous lesbian sexual partners, there was a high concordance for the presence or absence of this infection among the pairs, supporting the possibility of sexual transmission between women²¹ and a cross sectional study found that bacterial vaginosis risk was associated with an increasing number of female partners.²⁰ The mechanism for such transmission has not been established, although one proposed hypothesis is the sexual practice of repetitive cunnilingus, which is common among lesbians.²² Given the possibility of sexual transmission, it would be reasonable to advise female sexual partners to be examined and treated as appropriate. The treatment of bacterial vaginosis in a lesbian is the same as for a heterosexual woman.

Candida albicans Vulvovaginitis 

Most Candida infections are not the result of sexual transmission, and the prevalence of this infection is similar in lesbians. However, Candida can be acquired by direct contact with an infected person, and the risk of acquiring Candida increases in lesbians with more female partners.²³ Some heterosexual women with recurrent infections have a male partner who harbors the organism on the genital skin. Transmission between two women could, therefore, also be possible during vulva-to-vulva contact or transmission from fingers to vagina. To minimize the chance of transmission between women, it is prudent to advise the partner of the infected woman to avoid direct contact with the partner’s vulva or vagina during active infection. If fingers or hands come in contact, the partner should avoid placing her hand in her own vaginal area.

Trichomonas vaginalis Vaginitis

The primary route of transmission for Trichomonas vaginal infection is heterosexual intercourse. Two cases of probable female-to-female transmission have been reported.^{24, 25} In each case, a Trichomonas infection was diagnosed in both partners, and there was no history of recent heterosexual contact. In each case, transmission was thought to be via mutual masturbation. Although Trichomonas is fairly fastidious, it can survive in the discharge outside the vagina for at least several minutes. Female partners should therefore avoid contact with infected vaginal secretions until treatment is completed.

Gonorrhea, Chlamydia, and Syphilis

None of these has ever been reported in a lesbian who has never had heterosexual intercourse.^{9, 17, 26, 27} Therefore, the risk of transmission between women must be negligible. Chlamydia and gonorrhea, transmitted via infected semen, primarily infect the cervix, which is not easily accessible to the female partner. Even if a partner’s fingers picked up organisms, direct inoculation onto her own cervix would have to occur quickly. Although infection of a female partner is unlikely, it is again prudent to adhere to vaginal abstinence until treatment is completed.

Viral Sexually Transmitted Diseases

Herpes

Although herpes type 1 and type 2 are traditionally associated with oral and genital infections, respectively, each viral subtype can infect at either location. Transmission requires direct contact between infected and uninfected mucous or skin. Women with oral herpes may, therefore, transmit this infection to a female partner’s vulva via oral contact. The virus can also be transmitted via an intermediary, such as a finger. Johnson and coworkers found that of the 1921 lesbians surveyed, 3% reported having had a genital herpes infection, including 1.5% of the women who had never had heterosexual intercourse.⁹ Among lesbians seen in STD clinics, herpes infections are common.^{19, 28, 29} Marrazzo et al found that the finding of HSV-1 was more common as the number of female partners increased, whereas HSV-2 was associated with a history of male partners.³⁰ The key to avoiding transmission is to avoid direct contact with any active lesions on either the lips or the external genitalia. Once the lesion is scabbed over, the risk of transmission should be quite low.

Human Papillomavirus and Condyloma

Subtypes 6 and 11, which are responsible for most condyloma, infect squamous epithelium and are nearly impossible to eradicate completely. Transmission is poorly understood. Direct contact with infected surfaces, which usually occurs in the context of sexual activity, is probably the most common route of transmission for the human papillomavirus (HPV) subtypes that cause condyloma.

In theory, transmission of the HPV virus should be as 'efficient' between two women as that between a man and a woman. Marrazzo and colleagues³¹ found a prevalence of 30% for any kind of HPV type among 149 lesbians, and 19% among the 21 women who had never had heterosexual sexual activity, suggesting that sexual transmission does occur between women. Thus, condyloma are not uncommon among lesbians, although the prevalence may be lower than among comparable heterosexual women.^{19, 28}

Cervical Neoplasia

The majority of women who currently identify themselves as lesbian have a history of past sexual contact with men, and thus, are at risk for cervical dysplasia; abnormal Pap smears have been identified among women who have no such history, although the risk in this group is undoubtedly low.^{19, 28, 32, 33, 34} Marrazzo et al found positive HPV DNA tests in 13% of a sample of 248 women who had had sex only with women in the previous year.³⁵

Several small studies have found that lesbians, when compared to heterosexual or bisexual women, are less likely to get regular Pap smear screening.^{9, 12, 35, 36} Rates of screening are lower among younger lesbians with less education.³⁷ The urgent concern is the evidence that many lesbians who are at high risk for cervical neoplasia because they engage in heterosexual sex or are HIV infected, fail to recognize their risk, and are not screened appropriately.³⁸

The 1988 consensus recommendation of the American College of Obstetricians and Gynecologists, the American Cancer Society, the National Cancer Institute, the American Medical Association, the American Nurses’ Association, the American Academy of Family Physicians, and the American Medical Women’s Association was that all women who are or who have been sexually active should have an annual Pap smear, and that after three consecutive normal annual smears, further screening should be determined by individual risk factors. This approach is appropriate for lesbians; presence of the usual risk factors (HPV infection, smoking, HIV infection) would justify annual screening.³⁹

Hepatitis

Hepatitis C was found to be more than seven times more common in the lesbian subjects in a large case control study of women seen in a STD clinic.¹⁹ Risk factors included having had sexual contact with a homosexual or bisexual man or a partner who used intravenous drugs. HIV-infected women are also at higher risk.⁶ There is one well-documented case of probable transmission of hepatitis A between two women, suggesting that any infection that is transmitted via the oral–fecal route can be shared between women.⁴⁰

HIV

In 1990, the Centers for Disease Control (CDC) first reported on acquired immune deficiency syndrome (AIDS) cases among the newly devised category of women who have sex with women (WSW).⁴¹ Between 1980 and 1989, 79 cases of AIDS had been reported in this group, and 95% of these cases were intravenous drug users. Subsequent studies have confirmed that for a lesbian, the major risk factors for acquisition of HIV include personal intravenous drug abuse, bisexuality, and having sex with homosexual men.^{6, 42, 43, 44} In fact, lesbians who have one or more of these risk factors appear to have higher rates of HIV positivity than heterosexual women matched on the same risk factors.

Most, but not all,⁴⁵ studies have found that lesbians in these high-risk groups have often been found to be less likely to practice safe sex.^{19, 46, 47} For example, in a study of more than 280 HIV-infected women, all of whom had a traditional risk factor for HIV, 44 also reported a main female sexual partner over the preceding three years. Cleaning of sex toys before sharing was practiced by all but one woman, but the use of a dental dam (the functional equivalent of the condom for oral sex) was half as low, with only 26% who practiced oral sex usually using, and almost half never using dental dams. Overall, the use of barriers by lesbians was half that of infected heterosexual women.⁴⁸

The risk of transmission of HIV between female sexual partners is not known, but is thought to be quite low. The CDC has never reported a case thought to be the result of female-to-female transmission. There were no seroconversions in a prospective 6-month study of serodiscordant women sexual partners,⁴⁹ and no cases identified in a STD clinic survey that included over 130 women who had sex exclusively with women.⁴³

Three possible cases have been reported. The most convincing (and instructive) case is reported by Marmor and coworkers.⁵⁰ They describe a bisexual woman who developed typical acute HIV symptoms six weeks after a lesbian relationship with a woman whose AIDS was secondary to intravenous drug use. The sexual encounter included digital and oral contact with the infected partner’s vagina and anus. Both occurred during the menses of the infected woman. Finally, both women had vaginal bleeding from traumatic sexual activity. The infected woman had tested positive for HIV two years later, and had no other apparent risk factors. The other two cases rely on symptom history rather than serology for the diagnosis; it is not clear in either case that other risk factors were absent.^{51, 52}

In theory, female-to-female transmission should be the least efficient way to transmit the virus, because female-to-male transmission is thought to be less common than male-to-female or male-to-male transmission. Although the virus has been found in vaginal fluid, there are antibodies in the vagina that may limit transmission. Saliva also contains the virus, but exposure to saliva has not been proven as a route of transmission.

It appears that prevention efforts would be best directed toward lesbians who engage in the behaviors identified as risk factors. In addition to the standard advice to use condoms when having sex with men and avoiding needle sharing, these women should be advised to take measures to minimize exposure to the vaginal secretions of women partners via sex toy use and receptive oral sex.^{16, 48} Avoiding exposure to blood (from menses or genital tract trauma) is also advisable. The barrier methods recommended include the use of dental dams for oral–genital or oral-anal contact; condoms on dildos and sex toys; and latex finger cots during vaginal or anal penetration.

Many lesbians raise children who are the product of previous heterosexual relationships, and an increasing number want to start families in a lesbian relationship.⁵³ Many books are available that include useful and comprehensive discussions of the many issues involved in this decision.^{54, 55, 56} Women interested in this course should also be advised to talk to other lesbians who have had children. Adoption is a desirable option, but domestic adoptions of healthy newborns are difficult for lesbians to arrange. As a result, many lesbian families pursue international adoption, or children with special needs. Women who inquire about this option should be advised to consult with persons knowledgeable about the adoption situation locally.

Women wishing to be biologic mothers should be advised that donor insemination through a medical or other licensed facility is optimal. The essential procedure (placing semen in the vagina) does not require any special medical expertise or equipment and can be arranged without medical or legal consultation; however, this method is not advisable for a number of reasons. First, the donor may later desire involvement with the child beyond the original agreement, and although sometimes this works out well, it often may not. Second, the donor may decide to stop donating semen without notice. Finally, and perhaps most importantly, the woman cannot be sure that the donor is infection-free, particularly from HIV. HIV transmission via artificial insemination (one in the United States) has been reported, with an estimated risk of less than 0.5% per exposure with infected semen.⁵⁷ Eskenazi and colleagues⁵⁸ studied 88 women who were inseminated between 1979 and 1987. Half of the women received fresh semen from gay or bisexual men living in San Francisco before 1982. All women in the study were tested for HIV; none were positive.  Ross et al surveyed a group of lesbians and bisexual women, and shared their suggestions with assisted reproductive services who were interested in being more lesbian-friendly.⁵⁹

Not all medical providers are willing to provide insemination services for lesbians. The basis for this policy is sometimes stated as concern for the outcome of the children born into a lesbian household.  A number of studies have specifically focused on role development and social relationships. A meta-analysis of 18 studies concluded that there are no differences between straight and lesbian families in “parenting style, emotional adjustment or sexual orientation of the children.”⁶⁰ A review of data regarding children conceived by donor insemination and then raised by two-parent families found no adverse consequences of having homosexual parents.⁶¹

While the literature to date shows that children raised in lesbian households are not more likely to have adverse psychological, developmental, or educational outcomes than those raised by heterosexual parents,⁶² Stacy and Biblarz⁶³ observed that it seems theoretically implausible that there would be no differences at all. To test this hypothesis, they performed their own analysis of 21 studies published through 1998, and found a number of interesting differences, which they point out are essentially ignored by the authors of the studies. For example, young children being raised by lesbians are less likely to restrict their play to gender stereotypic activities; older girls are more likely to have interest in traditionally masculine occupations; and children of both genders are more likely to report homoerotic thoughts, although there is not a difference in reported sexual orientation. In conclusion, the authors suggest that there might be much to gain in the understanding of families and parenting styles if these differences were examined instead of glossed over.

A model for those who choose to provide lesbians with insemination services is described by Brewaeys and associates.⁶⁴ This clinic in Brussels, which has allowed lesbian insemination since 1982, has a specific clinic screening protocol. Before being accepted as clients, lesbians must: accept their lesbianism; have a supportive social environment; plan to seek male role models for the child; and have a currently stable relationship.

Lesbians who become pregnant appear to be model patients. Harvey and colleagues⁶⁵ surveyed 35 women who became pregnant as a result of artificial insemination. All obtained prenatal care within 16 weeks of conception; 90% attended childbirth classes; and 80% breastfed for at least six months. Of interest, 90% disclosed their sexual orientation to their provider.

Chronic Diseases

There is growing evidence that lesbians, as a group, have a profile of risk factors that may predispose to cardiovascular disease, diabetes, and various cancers. The factors that have been identified so far include higher rates of tobacco and alcohol use,^{12, 37, 66, 67, 68} obesity;⁶⁹ lower rates of parity and thus less breastfeeding, and less use of oral contraceptives.¹²

This combination of factors could lead to higher rates of cardiovascular disease, diabetes, and cancers of the breast, colon, endometrium, and ovary. However, there are few data examining the relative rates of these diseases between heterosexual and homosexual women. The largest sample to date is from the Women’s Health Initiative Study, in which almost 100,000 women are being followed prospectively to study a variety of postmenopausal health issues.¹² The participants were asked to define themselves as heterosexual (90,578); bisexual (740); lifetime lesbians (264), adult lesbians (309), or never having had sex as an adult (1420). Based on data collected at the time of enrollment, the lesbians and bisexual women were more likely to report the risk factors of obesity, smoking, and had a higher rate of alcohol use.

Overall, there were no significant differences in the rates of any of the disease conditions assessed, and differences described here must be interpreted accordingly. Rates of 'any cancer' were highest for the bisexual women, while the rates were similar for the other groups. Breast cancer was reported by each of the nonheterosexual groups more often than the heterosexual women, although the trend was not significant. Interestingly, the highest rates of cervical cancer were reported by the bisexual women, and unexpectedly, lifetime lesbians. Lifetime lesbians also reported the highest rate of colorectal cancer. There were no cases of endometrial cancer among the lifetime lesbians, also unexpected given their higher rate of nulliparity. In the category of cardiovascular disease, adult and lifetime lesbians reported more myocardial infarctions, but had the same rate of hypertension, and fewer strokes as compared to the bisexual and heterosexual women. Although the sample sizes for the nonheterosexual groups are the largest published to date in this age group, they are quite small from an epidemiologic perspective, and a finding of no difference must be interpreted with caution.

Mental Health Issues

Alcohol and Substance Abuse

The older literature suggested that lesbians have a high rate of alcohol abuse, and indeed, bars had an important place in lesbian culture. Most recent studies, conducted in general population samples, have found that more lesbians than heterosexual women use alcohol, but the rates of heavy drinking are similar to those of the heterosexual controls.^{12, 67, 70} Two studies found higher rates of heavy drinking among lesbians compared to controls.^{37, 66}

Substance use among lesbians appears to be more common than among heterosexual women. Data from the Trilogy Project, a longitudinal study of more than 1000 young adult gay men and lesbians, found higher rates of marijuana but not cocaine use compared to a general population survey.⁷⁰ Incidence rates of alcohol and drug dependency over one year were compared between the exclusive heterosexuality (n = 9714) and men and women reporting same gender partners (n = 194) from the 1996 National Household Survey of Drug Abuse; lesbians were more likely than heterosexual women to report drug or alcohol dependency.⁷¹ A 2006 survey of over 2,000 lesbians in California found that 33% had used marijuana in the past year, and 16% reported a serious problem drug use at sometime in their life.⁷² Lesbians who seek treatment tend to have more severe addiction problems, and more co-diagnoses than heterosexuals.⁷³

Depression and Anxiety

The actual rates of depression and anxiety among lesbians is not clear, with data varying widely among differently designed studies, but the literature taken as a whole suggests that lesbians have higher rates of mental health problems.  Population based studies provide the best data. A national probability survey in the US showed a higher rate of anxiety disorders among lesbians compared to heterosexual women,⁷⁴whereas a similarly designed study in the Netherlands showed higher rates of depression, but not anxiety, among lesbians compared to heterosexual women.⁷⁵ This same study also found that lesbians had a two-fold rate of at least two lifetime mental health diagnoses. The risk of depression and other mental health problems, including risk of suicidal ideation, appears to be increased among lesbians who are not 'out'⁷⁶and those who have experienced bullying and discrimination.⁷⁷ 

Of most concern has been the suggestion that homosexual adolescents are at higher risk for depression and suicide than their heterosexual peers. The commonly quoted risk for suicide attempts of between 20% and 40% for gay youths, or that 30% of all completed adolescent suicide victims are gay, are probably overestimates.⁷⁸ In a large representative survey of over 3,000 high school age young people, adolescent boys who were identified as gay were four times more likely than heterosexual boys to have attempted suicide, whereas girls who identified as gay did not have higher suicide attempt rates.⁷⁹

Intimate Partner Violence

Lesbians are at risk for intimate partner violence. The methods required for identification and intervention may differ from those that are effective with heterosexual women.⁸⁰ In a review of 19 studies addressing this issue, Burke and Follingstad⁸¹ conclude that the rates and risk factors for violence in lesbian and gay male relationships are similar to those in heterosexual ones, and that risk factors, other than gender, are similar. In a national probability sample, the prevalence of intimate partner violence was higher between gay men, and lower between lesbian partners compared to heterosexual couples.⁸² Helfrich and Simpson surveyed providers of services to battered women, and make suggestions about how these services could be made more lesbian-friendly.⁸³

Access

A 2006 study based on a national probability sample found that lesbians, but not gay men, were less likely than heterosexuals to have health insurance and to have seen a health care provider in the last year, and more likely to have unmet health care needs due to inability to pay for care.⁸⁴ Consistent with this data is a review of seven controlled studies that comprise data from 11,000 women which found that the rate of mammograms and pelvic examination frequency was lower between the lesbian and bisexuals.⁶⁷ Stevens¹ suggested three factors that may explain the reduced financial circumstances of many lesbians: (1) inadequate insurance coverage because of the inability to be covered by a partner’s health insurance; although this option is now sometimes available, it is far from universal; (2) low family income because women are paid less than men; and (3) inability to take advantage of inexpensive routine gynecologic care, which in many communities is provided by family planning clinics that serve only women who need birth control.

Lesbians who do seek regular care appear to have screening rates similar to those of care-seeking heterosexual women. Koh⁴⁵ compared screening testing among 524 lesbians, 143 bisexual women, and 637 heterosexual women recruited from a clinic setting and found that the rate of cholesterol testing was the same, the rate of HIV testing was higher among the lesbians and bisexuals, and mammography and Pap smear rates were the same. In the Women’s Health Study, the women who identified themselves in one of the two lesbian categories had the higher rates of mammography, and same rates of Pap smears, hemoccult, and the same hormone replacement therapy use as the heterosexual women.

Experiences With Health Care Providers

In her review of 20 years of literature regarding health care providers’ attitudes toward lesbians (nine studies), and lesbians’ experiences in health care encounters (19 studies), Stevens¹ concluded that despite some progress, problems still existed as of 1990.  

Positive attitudes on the part of providers are detectable by lesbian patients, and have good consequences on health care. Based on a survey of almost 7000 lesbians, Diamant and colleagues reported that disclosure of lesbian status to the health care professional was associated with a higher likelihood of getting a Pap smear, supporting the notion that openness is associated with better health care.⁸⁵ An analysis of lesbians who received regular health care revealed that a major predictive factor was 'coming out' to the provider.⁸⁶

Based on their experiences 'coming out' in other parts of life, some lesbians may find it difficult to disclose their orientation to a new provider. Stevens¹ points out that there are substantial personal costs to continually maintaining nondisclosure. If disclosure to a new person is considered, a complex monitoring process must occur to determine if the disclosure is safe. Despite the risks, many lesbians do choose to disclose to health care providers. In two large surveys, half of the lesbian respondents had disclosed to at least one physician. Smith and colleagues⁸⁷ provide additional details. When asked what effect disclosure would have on their care, 38% said it would 'hinder', 31% thought 'no effect', and only 22% said it would improve their care. Yet, disclosure was usually unsolicited; only one fourth of the women who disclosed did so in response to a direct question from the physician.

In summary, there is both theoretical and empirical evidence to suggest that disclosure to health care providers (perhaps especially physicians) is considered risky, yet it appears that a large number of lesbians do so despite the risk. Why? Johnson and Guenther⁸⁸ suggest three possible reasons: a need for appropriate health care; a desire for an enhanced relationship with the provider; or a political belief that society’s attitudes will change only if gay persons 'come out'.

A need for appropriate health care might overcome fear of adverse interpersonal consequences. A lesbian with a vaginal infection will want to know if she can transmit it to her female partner, and if so, during what specific sexual activities. More urgently, she may want to discuss her risk of acquiring HIV infection and how to reduce this risk. Many lesbians now want to have children, and must successfully negotiate the process for insemination, and later, prenatal care and childbirth. The lesbian with a life-threatening illness needs to include her partner in discussions with the provider, and to make known the power-of-attorney to be used should she be unable to speak for herself.

The other two motivations for disclosure are less obvious but no less important. An honest, respectful, mutually satisfactory relationship between the provider and the patient leads to better health outcomes. Sexual orientation and intimate relationships are each integral parts of a person’s identity. When it is necessary to withhold these aspects of the self from the provider, trust cannot develop. Finally, individuals who are comfortable being 'out' may disclose purely for political purposes. This strategy may, in fact, be useful. Physicians who have knowingly treated homosexual patients have been shown to be less likely to be homophobic.

The literature is unanimous in its recommendations to health care providers who choose to improve services to lesbians:

1. Educate yourself about the medical aspects of problems relevant to lesbians. This chapter provides a starting point for that task.
   
2. Use inclusive language. Questions about marriage, sexual activity, and birth control are often asked in a way that assumes the respondent is heterosexual. Instead of “Are you married?” try “Do you have an ongoing relationship?” Rather than “What do you use for birth control?” ask “Are you in need of birth control?” By allowing an opening for the nontraditional answer to these questions, you create an atmosphere in which the lesbian can more confidently disclose.
   
3. Keep the patient’s disclosure of her sexual orientation confidential. Some clinicians use a personal code in the medical record so that they remember from visit to visit. If there is need to write the sexual orientation in the record, or if another provider should be informed, inform the woman why.
   
4. Include the significant other, or other support system of the patient, in the same way you would the partner of a heterosexual woman. This is easy to do in your own office; it may still occasionally be necessary to arrange exceptions to 'family only' policies in the hospital setting.
   
5. Finally, examine your own attitudes, and if you are not comfortable caring for lesbians, refer to another physician.