Despite evidence that sexually transmitted infections (STIs) can be spread between female sex partners,1–3 there is relatively little literature on sexual behaviors and associated risk for STIs among women who identify as lesbian or bisexual, or who have sex with women, herein referred to as sexual minority women (SMW). Lower perceived risk of STIs,4 lower use of barrier protection,1,4,5 lower use of reproductive health care compared with heterosexual women,6,7 and lack of knowledge by primary providers about appropriate screening8 may all lead to undetected STIs among SMW. Several observational studies have demonstrated higher rates of STIs among women who have sex with women compared with women who have sex with only men.9–13 Furthermore, cross-sectional analysis of US adolescents and young adults found that bisexual women had significantly higher odds of having a bacterial STI than did heterosexual women4 and that women who identified as “mostly heterosexual” or bisexual women had significantly higher odds of STI history compared with “completely heterosexual” women.6 In addition, previous research has suggested that SMW may have higher rates of drug use and heavy alcohol use.9,12,14,15 Because substance use, particularly before or during sex, has been found to be associated with increased sexual risk behaviors among women treated at urban health clinics16,17 and STI diagnosis,18 elevated rates of substance use may impact sexual behavior and STI outcomes among SMW.
However, differences in classification methods make it difficult to estimate STI prevalence among SMW. Studies of STIs and associated risk factors have often combined lesbian and bisexual women or examined sex of current sex partners independently of sexual identity, despite evidence of discordance between sexual identity and sexual behavior, and fluidity of sexual identity and partner sex over time.19 In contrast, this study uses both sexual identity and partner sex to distinguish between risks associated with identification as a sexual minority and the sex of partners.
Our objectives were to describe STI prevalence among women of different sexual identities and to determine if substance use, sexual behaviors, and partner characteristics varied by sexual identity, and whether such differences, if present, were associated with differences in STI prevalence. An understanding of the unique sociocultural and behavioral factors that influence STI risk among SMW is important for reducing transmission and progression of serious STI sequelae20 and for developing effective prevention messaging.
MATERIALS AND METHODS
Participants and Data Collection
This was a cross-sectional analysis of data collected from 669 women who accessed sexual health care services at the Howard Brown Health Center walk-in STI clinic. Howard Brown Health Center primarily serves lesbian, gay, bisexual, and transgender (LGBT) people, but the walk-in clinic is a high-volume, low-cost, confidential STI testing and treatment center used by many outside the LGBT community. The analysis was restricted to the first clinic visit between January 2009 and December 2011, excluding 69 subsequent visits by 53 women. Similar associations were observed when we included these visits in the analysis. We also excluded 22 women who did not report their sexual identity, which was the primary explanatory variable. All study methods and data analysis protocols were reviewed and approved by the institutional review board at Howard Brown Health Center.
Outcome Variables. STI diagnoses were the primary outcomes of interest. The standard clinic protocol included screening for urogenital gonorrhea (GC), chlamydia (CT), and syphilis. Diagnoses of GC or CT were based on nucleic acid amplification testing of urine specimens using Strand Displacement Amplification with the Becton-Dickinson ProbeTec assay (BD Diagnostic Systems, Sparks, MD). We did not include syphilis in this analysis because there were very few cases of syphilis among women at our clinic during the study period. Bacterial vaginosis, genital warts, genital herpes, pelvic inflammatory disease (PID), and trichomoniasis were evaluated based on clinical signs and symptoms. Papanicolaou tests and human papillomavirus DNA testing are not performed at the walk-in clinic, so only the prevalence of HPV-related genital warts, not HPV infection, could be determined. Information on clinical diagnoses was extracted from patients’ electronic medical records (Centricity Practice Solution; GE Healthcare).
Explanatory Variables. Behaviors were measured using the risk assessment used in routine STI surveillance. Patient characteristics collected at each visit included age, race/ethnicity, STI history, and sexual orientation. For the 90 days before the clinic visit, patients reported the number and sexes of sex partners, whether they had anonymous sex partners, whether they met partners via the internet, and how often they used condoms or barrier protection for oral, vaginal, and anal sex. Alcohol and drug use was assessed in the past year: substances listed were cocaine, ecstasy, heroin, marijuana, methamphetamines, poppers, nonprescribed prescription drugs, club drugs, and “other.” Patients also reported their most recent sex partner’s sex, race/ethnicity, the partnership type (main or casual), relationship type (monogamous or nonmonogamous), and whether they had unprotected oral, vaginal, or anal sex with that partner. In addition, patients reported whether they or their partners used drugs or alcohol at last sex. All data were based on self-report.
Alcohol use, substance use, and individual and partner substance use at last sex were categorized as any versus none. Condom and barrier use was dichotomized as always versus not always using barriers for vaginal or anal sex separately. Patient age was collected as a continuous variable, and means were compared for analysis. Those who had more than 1 recent sex partner were compared with those with 1 or fewer recent sex partners. Women who reported having only male sex partners in the last 90 days were categorized as women who have sex with men, those who only reported female sex partners were categorized as women who have sex with only women, and those who reported both male and female sex partners were categorized as women who have sex with women and men.
Three STI-specific variables were created: (1) laboratory-confirmed diagnosis of GC or CT, (2) laboratory-confirmed or clinical diagnosis of any STI at clinic visit (including GC, CT, bacterial vaginosis, genital herpes, genital warts, syphilis, trichomoniasis, or PID), and (3) self-reported history of STI (any vs. none). Sexually transmitted infection diagnoses were combined to improve statistical power despite potentially different epidemiologic profiles and infection mechanisms because we did not have large-enough sample size to examine each infection separately and because the analysis was exploratory in nature.
We compared sexual behavior, substance use, and STI diagnoses by sexual identity. Differences in nonnormally distributed continuous variables were assessed using Kruskal-Wallis or Wilcoxon rank sum tests. Bivariate analysis of categorical variables with sexual identity was performed using χ2 tests or Fisher exact test, where appropriate, with statistical significance determined at the P < 0.05 level.
Logistic regression was used to calculate univariable and multivariable odds ratios, controlling for age (as a continuous variable) and race/ethnicity (as a categorical variable, with white race as the referent category). Logistic regressions contained the 3-category sexual identity variable as the primary explanatory variable with heterosexual as the referent category. Models were adjusted for age and race because these variables were independently associated with many of the behavioral risks and because age, race, or age, and race were significant variables in multivariate models. No variables were identified as effect modifiers, so interaction terms were not included in the final regression models. All statistical analyses were conducted using SAS statistical software 9.2 (SAS Institute, Inc, Cary, NC).
Characteristics of Study Population
Of the 669 women who sought sexual health services at the walk-in STI clinic between 2009 and 2011, 9.3% identified as lesbian, 15.3% as bisexual, and 75.4% as heterosexual (Table 1). There were no significant racial/ethnic differences by sexual identity. The mean age among lesbian women was 32 years versus 29 years among heterosexual and bisexual women (P = 0.03).
Bisexual women were significantly more likely to report more than 1 recent sexual partner (68%) than heterosexual (43%) or lesbian women (42%; Table 1). There was discordance between sexual identity and reported sex partner sex: 32.8% of lesbians reported male partners in the last 90 days, and 2.5% of heterosexuals reported female partners. Most (58.6%) lesbian women had exclusively female partners in the last 90 days, whereas 20.7% of lesbians had only male partners (Table 2). Bisexual women were more likely than heterosexual women to report anonymous partners in the past year (31.0% vs. 15.7%, P = 0.002). In addition, bisexual women were less likely than lesbian and heterosexual women to describe their most recent partner as their main partner or to report monogamy with their most recent partner. Few women of any sexual identity reported exchanging sex for drugs or money.
Sexual minority women had 2 times the odds of using drugs in the last year, compared with heterosexual women (Table 3). Bisexual women had the highest odds of drug use overall (adjusted odds ratio [AOR], 2.44; 95% confidence interval [CI], 1.54–3.88) and at last sex (AOR, 5.35; 95% CI, 2.28–12.5). Twenty-three percent of bisexual women’s last sex partners used substance use at last sex, significantly more often than heterosexual or lesbian women (Table 1).
Alcohol use (measured as any use in the past year) was nearly ubiquitous, ranging from 85.4% for heterosexuals to 96.2% among lesbians. However, individual and partner alcohol use at last sexual encounter was significantly more common among heterosexual and bisexual women than among lesbian women (Table 1).
Nearly all women reported sex (oral, anal, or vaginal) in the last 90 days, and differences by sexual identity were not statistically significant. Lesbian and bisexual women were statistically similar in regard to recent vaginal sex and were significantly less likely to report having vaginal sex than heterosexual women (88.0% vs. 94.6%; P = 0.01). Consistent condom use during vaginal sex was low and did not vary by sexual identity. Bisexual women were more likely to report anal sex than lesbian or heterosexual women, but they had the highest rate of condom use among women reporting anal sex (Table 1).
STI Positivity and Self-Reported History of Diagnosis
Overall positivity for laboratory-confirmed GC was 1.5% and that for CT was 5.2%; combined positivity was 6.3%. Gonorrhea and CT positivity was significantly higher among women 25 years or younger (10.7%), compared with those older than 25 years (4.4%; P = 0.002). Gonorrhea and CT positivity did not differ by sexual identity. There were also no significant differences by sexual identity in reported symptoms in the last 90 days (Table 1). Regardless of sexual identity, all women for whom information was available on partner sex (n = 576; 86.1% of the sample) with a laboratory-confirmed or clinical STI diagnosis at the clinic visit reported recent male sex partners, although these results should be interpreted with caution because of small sample size. Of the characteristics collected from each patient, only age and partner sex were significantly associated with STI diagnosis at visit (Table4). This association between male sex partners and infection remained when we included other clinically diagnosed and laboratory-confirmed STIs (Table 2). One case of PID and 1 case of bacterial vaginosis were diagnosed in women who had sex with women and men; all other cases of PID, bacterial vaginosis, genital herpes, genital warts, syphilis, and trichomoniasis occurred among women who had only male sex partners.
Self-reported history of STI also varied. Less than half (43.4%) of bisexual women and 42.6% of heterosexual women reported ever being diagnosed as having an STI versus 18.3% of lesbian women, a statistically significant difference (P = 0.001). Lesbian women had significantly lower odds of ever being diagnosed than did both bisexual (AOR, 0.31; 95% CI, 0.14–0.66) and heterosexual women (AOR, 0.31; 95% CI, 0.16–0.60).
As expected, there was variation in sexual behaviors and partner characteristics by sexual identity. Norms and expectations of sexual minority communities may affect the behavior and partner choices of lesbian and bisexual women. However, sexual identity was not as powerful a predictor of STI risk as sex of partners. Collecting patient information on both sexual identity and sex partners is necessary to fully understand risks, but eliciting accurate disclosure can be difficult, as SMW may anticipate heterosexist or homophobic treatment from providers. Diversity training for providers may be useful for increasing sensitivity and awareness about LGBT health issues. Provider knowledge and comfort with sexual minorities can increase patient comfort, disclosure, and care.21 Knowledge of patients’ sexual identity is helpful not only in providing culturally competent care but also in correctly identifying risks associated with social prejudice and stress, such as substance use and mental health outcomes.22,23 Assessing partner characteristics, which have been shown to influence STI risk both here and elsewhere,24–26 enables appropriate STI and pregnancy risk assessment and patient education.
Behavioral data and STI history diagnosis were based on self-report and may be subject to recall or social desirability bias. Not all patients answered all questions, and these missing data both reduced sample size and introduced potential bias, although missingness did not differ by sexual identity, age, or race.
Patients’ ages ranged from 16 to 66 years, but the overall mean age was 30 years. This is a primarily adult sample, so although age was controlled for in multivariable analysis, our results may not be generalizable to populations of younger women, who typically have higher rates of STIs and different behavioral risk profiles.
In addition, these data are cross sectional and do not provide information on temporal relationships between exposure and outcome variables. The sample is solely from STI walk-in clinic patients, so STI positivity and related behaviors may not be generalizable to women in other settings. Women were only routinely screened for urogenital GC and CT and syphilis, so estimates of genital herpes, genital warts, and trichomoniasis are likely underestimated because of incomplete reporting of clinical diagnoses in the electronic medical records and lack of routine screening for trichomoniasis and HPV infection. Owing to low sample sizes, risk factors for individual STIs could not be estimated. Estimates from other cross-sectional studies of SMW at sexual health clinics suggest prevalences ranging from 1% to 14.3% for genital herpes,2,7,12,23,27 0.5% to 7.5% of genital warts,2,27 7.9% to 33% of bacterial vaginosis,2,3,12,27 1.4% to 5.2% of trichomoniasis,2,3,27 and 0.7% to 1.2% of syphilis.3 Trichomoniasis, genital herpes, and genital warts may be sexually transmitted between women,2,28–30 and so underestimation of these xinfections, which are more commonly transmitted between women who have sex with women than GC and CT, may underestimate the overall STI burden among SMW. Conversely, because SMW have higher odds of unmet medical needs, particularly regular Papanicolaou tests,6,31,32 they may be more likely to be first diagnosed at an STI walk-in clinic than sexual majority women, and thus, this sample may overestimate STI prevalences among SMW.
Strengths of the study include the heterogeneous racial and ethnic makeup of the sample. In addition, we were able to characterize women according to both sexual identity and sex of partners and to compare positivity within these groups, unlike previous studies that have combined SMW into a single category or used only one method of measuring sexual behavior or identity. Howard Brown Health Center is known as an LGBT health care organization, which may have enabled greater patient disclosure regarding nonheterosexual behavior and identities.
We found significant differences in behaviors and STI diagnoses between lesbian and bisexual women, suggesting that combining SMW into a single category may mask important differences in behavior and disease prevalence.
Consistent with other literature,2 we did not find evidence of transmission of GC and CT among women who reported exclusively female sex partners, although our results should be interpreted in the context of small sample size. Regardless of sexual identity, women with recent male sex partners were at risk for infection. The discordance between sexual orientation and sex of partners suggests that irrespective of sexual orientation, women who are sexually active with men should be screened for STIs. We observed multiple cases of GC and CT among lesbian women 25 years or younger, suggesting that age-driven screening recommendations should be used regardless of reported sexual identity.33
In this sample, less than half of women of any sexual identity consistently used barrier protection during sex, suggesting a need for interventions to increase barrier method use. Intervention messaging should be culturally sensitive and inclusive toward SMW.
In addition, nearly half of lesbian and bisexual women reported using drugs in the last year. Bisexual women and their partners had particularly high rates of drug and alcohol use during sex. The health of SMW may benefit from further study of the underlying causes and consequences of higher substance use rates.
1. Marrazzo J, Coffey P, Bingham A. Sexual practices, risk perception and knowledge of sexually transmitted disease risk among lesbian and bisexual women. Perspect Sex Reprod Health 2005; 37: 6–12.
2. Bailey J., Farquhar C, Owen C, et al. Sexually transmitted infections in women who have sex with women. Sex Transm Infect 2004; 80: 244–246.
3. Marrazzo J, Koutsky L, Handsfield H. Characteristics of female sexually transmitted disease clinic clients who report same-sex behavior. Int J STD AIDS 2001; 91: 591–597.
4. Kaestle C, Waller M. Bacterial STDs and perceived risk among sexual minority young adults. Perspect Sex Reprod Health 2011; 43: 158–163.
5. Rowen T, Breyer B, Lin T, et al. Use of barrier protection for sexual activity among women who have sex with women. Int J Gynecol Obstet 2013; 120: 42–45.
6. Charlton B, Corliss H, Missmer S, et al. Reproductive health screening disparities and sexual orientation in a cohort study of US adolescent and young adult females. J Adolesc Health 2011; 49: 505–510.
7. Bauer G, Jairam J. Are lesbians really women who have sex with women (WSW)? Methodological concerns in measuring sexual orientation in health research. Women Health 2008; 48: 383–407.
8. Farquhar C, Bailey J, Whittaker D. Are lesbians sexually healthy? A report of the “Lesbian Sexual Behavior and Health Survey.” London, Soc Sci Res Papers No 11, Faculty of Huam.
9. Bauer G, Jairam J, Baidoobonso S. Sexual health, risk behaviors, and substance use in heterosexual-identified women with female sex partners: 2002 US National Survey of Family Growth. Sex Transm Dis 2010; 37: 531–537.
10. Bevier P, Chiasson M, Heffernan R, et al. Women at a sexually transmitted disease clinic who reported same-sex contact: Their HIV seroprevalence and risk behaviors. Am J Public Health 1995; 85: 1366–1371.
11. Bell A, Ompad D, Sherman S. Sexual and drug risk behaviors among women who have sex with women. Am J Public Health 2006; 96: 1066–1072.
12. Fethers K., Marks C., Mindel A., et al. Sexually transmitted infections and risk behaviors in women who have sex with women. Sex Transm Inf 2000; 76: 345–349.
13. Everett B. Sexual orientation disparities in sexually transmitted infections: Examining the intersection between sexual identity and sexual behavior. Arch Sex Behav 2013; 42: 225–236.
14. Gonzales V, Washienko KM, Krone MR, et al. Sexual and drug-use risk factors for HIV and STDs: A comparison of women with and without bisexual experiences. Am J Public Health 1999; 89: 1841–1846.
15. Burgard S, Cochran S, Mays V. Alcohol and tobacco use patterns among heterosexually and homosexually experienced California women. Drug Alcohol Depend 2005; 77: 61–70.
16. Chadambuka A, Chimusoro A, Maradzika J, et al. Factors associated with contracted sexually transmitted infections among patients in Zvishavane urban Zimbabwe. Afr Health Sci 2011; 11: 535–542.
17. Hutton H, McCaul M, Santora P, et al. The relationship between recent alcohol use and sexual behaviors: Gender differences among sexually transmitted disease clinic patients. Alcohol Clin Exp Res 2008; 32: 2008–2015.
18. Cook R, Clark D. Is there an association between alcohol consumption and sexually transmitted diseases? A systematic review. Sex Transm Dis 2005; 32: 156–164.
19. Diamont A, Schuster M, McGuigan K, et al. Lesbians’ sexual history with men: Implications for taking a sexual history. Arch Intern Med 1999; 159: 22730–22736.
20. Mehta S. Gonorrhea and chlamydia in emergency departments: screening, diagnosis, and treatment. Curr Infect Dis Rep 2007; 9: 134–142.
21. Rankow EJ, Tessaro I. Mammography and risk factors for breast cancer in lesbian and bisexual women. Am J Health Behav 1998; 22: 401–410.
22. King M, Semlyen J, Tai S, et al. A systematic review of mental disorder, suicide, and deliberate self harm in lesbian, gay and bisexual people. BMC Psychiatry 2008; 8: 70.
23. Reisner S, Mimiaga M, Case P, et al. Sexually transmitted disease (STD) diagnoses and mental health disparities among women who have sex with women screened at an urban community health center, Boston, MA, 2007. Sex Transm Dis 2010; 37: 5–12.
24. Staras S, Cook R, Clark D. Sexual partner characteristics and sexually transmitted diseases among adolescents and young adults. Sex Transm Dis 2009; 36: 232–238.
25. Gorbach P, Drumright L, Holmes K. Discord, discordance, and concurrency: Comparing individual and partnership-level analyses of new partnerships of young adults at risk of sexually transmitted infections. Sex Transm Dis 2005; 32: 7–12.
26. Javanbakht M, Gorbach P, Stirland A, et al. Prevalence and correlates of rectal chlamydia and gonorrhea among female clients at sexually transmitted disease clinics. Sex Transm Dis 2012; 39: 917–922.
27. Skinner C, Stokes J, Kirlew Y, et al. A case-controlled study of the sexual health needs of lesbians. Genitourin Med 1996; 72: 277–280.
28. Johnson S, Smith E, Guenther S. Comparison of gynecologic health care problems between lesbians and bisexual women: A survey of 2,345 women. J Reprod Med 1987; 32: 805–811.
29. Sivakumar K, De Silva A, Roy R. Trichomonas vaginalis
infection in a lesbian. Genitourin Med 1989; 65: 399–400.
30. Kellock D, O’Mahoney C. Sexually acquired metronidazole-resistant trichomoniasis in a lesbian couple. Genitourin Med 1996; 72: 60–61.
31. Buchmueller T, Carpenter C. Disparities in health insurance coverage, access, and outcomes for individuals in same-sex versus different-sex relationships, 2000–2007. Am J Public Health 2010; 100: 489–495.
32. Kerker B, Mostashari F, Thorpe L. Health care access and utilization among women who have sex with women: Sexual behavior and identity. J Urban Health 2006; 83: 970–979.
33. Centers for Disease Control and Prevention. Sexually transmitted disease treatment guidelines, 2010. MMWR 2010; 59 (No. RR-12): 44–49.