|Home | About | Journals | Submit | Contact Us | Français|
A growing body of research documents mental health disparities among women who have sex with women (WSW) compared to women who have sex with men only (WSM). However, there remains a dearth of research exploring these indicators alongside sexually transmitted diseases (STDs) and WSW sexual health.
A retrospective chart review was conducted of all female patients (n=368) screened for STDs between July 2007 and December 2007 at an urban community health center in Boston, MA. De-identified electronic medical record data (e.g., demographics, psychosocial, sexual health) were analyzed and linked to STD positivity. Women who did not have sexual behavior documented in their medical chart (n=58) were excluded from this analysis. Bivariate and multivariable logistic regression procedures examined sexual and psychosocial health indicators, including sexual preference.
Twenty-seven percent of participants were WSW (17% WSW only and 10% WSW/M). Overall, 5% of WSW were diagnosed with a new STD (HPV, anogenital warts, genital herpes, PID) and 17% had a history of a prior STD. In multivariable models adjusting for demographics, WSW were disproportionately more likely to have mental health and psychosocial issues noted in their medical records, including: a clinical diagnosis of depression, anxiety, and PTSD, history of suicide attempts, and inpatient psychiatric/mental health treatment. However, WSW were significantly less likely than WSM to engage in “high risk” HIV/STD sexual behavior. In a final multivariable model, same sex behavior was not associated with a different likelihood of being diagnosed with an STD, compared to opposite sex behavior. However, WSW diagnosed with STDs were at increased odds of having bipolar disorder and utilizing outpatient mental health counseling services compared to WSW without STDs. WSW with a history of STDs were at increased odds of having attempted suicide in the past, utilizing both outpatient and inpatient mental health treatment services, and having a history of injection drug use compared to WSW without a history of STDs.
WSW with STDs may have serious psychosocial problems. Further research is warranted to better understand the relationship between sexual behavior and health, as well as to guide the development of interventions to ameliorate health disparities among WSW, particularly in the psychosocial domain.
Relative to women who have sex with men only (WSM), women who have sex with women (WSW) have been shown to be disproportionately affected by a variety of psychosocial and physical health factors.1-6 For example, WSW have been shown to be disproportionately affected by mood disorders and increased psychological distress, most commonly depression and anxiety.7-14 WSW have also been shown to use alcohol and illicit substances to a greater degree than WSM, and are at increased risk of having alcohol and drug dependency disorders.9,13,15-24 Moreover, elevated rates of tobacco use have been observed in WSW compared to WSM.15, 23-26
Despite these documented health disparities, however, a dearth of research to date has examined these psychosocial factors in relation to sexually transmitted diseases (STDs) among WSW. Prior research about WSW sexual health indicates that up to 44% of WSW have a lifetime history of one or more STDs, depending on the sample characteristics and the STDs studied.27-37 Among cross-sectional studies examining current STD diagnosis, genital human papillomavirus (HPV) and bacterial vaginosis have been shown to be relatively common among WSW, with prevalence ranging from 3% and 19%,29-32 while the prevalence of gonorrhea, Chlamydia, syphilis, and HIV have been shown to be relatively low among WSW, ranging from 1% and 3%.27,28,30,37 Although some prior research has documented significantly lower rates of STDs among WSW compared to WSM,32,35 other studies have found increased rates of STDs among WSW.34,36 Further research is warranted to investigate WSW sexual health in conjunction with health disparities in order to understand who is most at risk for STDs.
The purpose of this study was threefold: (1) to explore STD prevalence among Massachusetts area women who underwent STD screening at an urban community health center, by sexual behavior; (2) to describe differences the demographics, behavioral risk factors, substance use, mental health diagnoses, and chronic disease between WSW and WSM in the sample; and (3) to understand the relationship between health disparities and STDs among WSW.
A retrospective chart review was conducted of all female patients (n=368) who were screened for STDs for a six month period between July and December 2007 at Fenway Health (FH), the largest freestanding community healthcare and research facility serving the needs of the lesbian, gay, bisexual, and transgender community in the greater Boston area.38 FH provides comprehensive medical and mental health services—including but not limited to primary care, gynecology, outpatient mental health therapy, and addiction services such as outpatient substance abuse treatment and acupuncture detoxification—to approximately 12,000 female patients annually.38 All women who came into the clinic and were screened for the following STDs were included in this sample: HIV, Chlamydia, gonorrhea, syphilis, human papillomavirus (HPV), genital herpes, pelvic inflammatory disease (PID), anogenital warts, and trichomoniasis. The study was approved by the FH Institutional Review Board.
FH uses Centricity® EMR (GE Healthcare), an electronic medical record (EMR) system that enables physicians and clinical staff to document patient encounters, streamline clinical workflow, and securely exchange clinical data with other providers, patients, and information systems. Data were extracted from patients' EMRs, including: demographic characteristics (age, race/ethnicity, education, employment status, health insurance type, children under age 18); current substance use [alcohol (5+ drinks per week), tobacco, marijuana, crack/cocaine, club drugs (crystal methamphetamine, GHB, ecstasy, ketamine), injection drug use (IDU)]; mental health clinical diagnoses [DSM-IV-TR Axis I: depression, anxiety, PTSD, SUD, bipolar disorder, adjustment disorder, other (not specified); DSM-IV-TR Axis II: personality disorders]39; psychosocial factors (history of sexual, physical, emotional abuse; history of inpatient psychiatric hospitalization; outpatient mental health utilization); history of any STD diagnosis (through key word searches of the patients' electronic medical records as well as laboratory reports).
Depending on presenting symptoms, risk of exposure, and clinician assessment, test types used for each STD were the following: HIV: Serum EIA HIV-1/2 antibody test reported as reactive or non reactive. Chlamydia: Endocervical swab - Gen-Probe® reported as detected or not detected. Gonorrhea: Endocervical swab - Gen-Probe® reported as detected or not detected. Syphilis: Rapid Plasma Reagin test (RPR), with positive tests confirmed by the Microhemagglutination assay (MHA-TP). HPV: Clinical diagnosis and HPV DNA testing on PAP smear - reported as high risk or low risk detected or not detected. Genital herpes: cultures using Remel Micro Test M4 RJ transport media, and clinical diagnosis for HSV 1/2. PID: clinical diagnosis and Gen-Probe® reported as detected or not detected. Anogenital warts: clinical diagnosis by examination. Trichomoniasis: wet mount. In addition to STDs, the prevalence of other gynecologic infections, specifically bacterial vaginosis and yeast, were captured through vaginal swab - BBL™. All tests were processed by Quest Diagnostics (Cambridge, MA).
Women who tested positive (or reactive) or who were found to have an STD by clinical examination (HIV, Chlamydia, gonorrhea, syphilis, HPV, genital herpes, PID, anogenital warts, or trichomoniasis) were classified as diagnosed with an STD on the date of their visit; those whose tests or clinical exams were classified as negative were not. The number of positive (or reactive) test results divided by the total number of positive and negative test results among women were used to calculate STD positivity in this population. Indeterminate (n=1) and contaminated (n=1) test results were excluded from these analyses.
To address patterns of screening for STDs among women, reason for STD testing was documented by the clinician at time of testing as follows: (1) STD symptoms (patient having symptoms suggestive of STDs); (2) STD exposure (sexual partner diagnosed with STD in the previous 3 months); (3) sexual risk behavior in the prior 3 months (perceived “high risk” sex i.e., sex with a man without a condom, condom breakage, anonymous sex partners); or (4) routine STD screening (e.g., complete physical exam, HIV follow-up appointment, entering a new relationship, or patient preference for regular screening).
Data extracted from patient electronic medical records were entered into a study database and analyzed with SAS® version 9.1.3,40 where statistical significance was determined at the p<0.05 level. De-identified electronic medical record data were analyzed and linked to STD positivity results. The univariate distributions for each of the variables relevant to the analysis were examined. Data was analyzed examining proportional differences using chi-square tests and, where cell sizes were small, Fisher's exact tests.
Through key word searches of the patients' electronic medical records as well as laboratory reports where relevant, the following primary outcomes were dichotomously assessed: current substance use; mental health clinical diagnoses; psychosocial factors; and history of STD diagnosis.
The primary predictor of interest was sexual behavior. Women who reported sexual behavior with women at their study visit were categorized behaviorally as WSW, regardless of whether they also reported sex with men; those who reported only sex with men at their study visit were categorized as WSM. Women who did not have sexual behavior documented in their chart (n=58) around the time of STD screening were excluded from this analysis.
For all variables, bivariate logistic regression analyses were conducted to establish statistically significant parameter estimates with the outcomes of interest. A separate multivariable logistic regression model was constructed for each outcome, with WSW as the primary predictor of interest. Age, race/ethnicity, and health insurance were all adjusted for in all multivariable models, given their documented profound impact of health care utilization.41-44
Bivariate logistic regression procedures were used to assess whether WSW diagnosed with an STD on their date of visit and WSW with a history of STDs were more likely to have psychosocial issues, as compared to WSW who were not diagnosed with an STD on the date of their visit and those with no STD history.
Table 1 provides an overview of the total sample (demographic, behavioral, and psychosocial, characteristics) by sexual behavior (WSW and WSM). Overall, 27% were WSW (17% WSW only and 10% WSW/M). WSW were significantly older in age compared to WSM (mean age WSW 32 vs. WSM 29; t-test statistic=3.34; p=0.001). WSW were also more likely to be employed (OR=9.24; p=0.002) and less likely to be students (OR=0.23; p=0.04) compared to WSM.
Five percent of WSW were diagnosed with a new STD (HPV, anogenital warts, genital herpes, PID) on the day of their visit compared to 17% of WSM (Table 2). Notably, 17% of WSW had a history of one or more STDs, compared to 19% of WSM (Table 1).
WSW were significantly more likely to have used alcohol (5+ drinks per week) (OR=2.41; p=0.002) relative to WSM.
WSW were significantly more likely to have a clinical mental health diagnosis (meet DSM-IV criteria for depression, anxiety, PTSD, SUD, bipolar disorder, adjustment disorder, other (not specified)39) than WSM (OR=2.58; p=0.0003). Moreover, being a WSW was independently associated with a greater odds of the following: depression (OR=2.86; p<0.001), anxiety (OR=2.57; p=0.002), PTSD (OR=4.74; p=0.008), and adjustment disorders (OR=3.46; p=0.05). WSW were also more likely to have a personality disorder that met DSM-IV criteria39 (OR=5.70; p=0.05) compared to WSM.
Relative to WSM in the sample, WSW were more likely to have a history of sexual abuse (OR=2.83; p=0.01), emotional abuse (OR=2.23; p=0.03), attempted suicide (OR=3.93; p=0.01), and inpatient mental health treatment (OR=4.50; p<0.001), as documented by a clinician in their medical record.
WSW were less likely to be diagnosed with a new STD (OR=0.32; p=0.04) compared to WSM women; however there was no observed difference between WSW and WSM women with respect to STD history. WSW were more likely to report routine screening as the reason for STD testing (OR=2.32; p=0.01) and less likely to report STD symptoms (OR=0.41; p=0.02) and “high risk” sexual behavior (OR=0.29; p=0.003) in comparison to WSM.
In multivariable models adjusting for patients' age, race/ethnicity, and health insurance status, WSW were disproportionately affected by mental health and psychosocial issues: any clinical mental health diagnosis (AOR=3.45; p<0.001), depression (AOR=3.56; p=0.004), anxiety (AOR=3.34; p=0.008), and PTSD (AOR=12.34; p=0.01), history of suicide attempt (AOR=14.87; p<0.001), and inpatient psychiatric/mental health treatment (AOR=11.55; p=0.004). However, WSW were less likely than WSM to engage in high risk HIV/STD sexual behavior (AOR=0.24; p=0.03).
WSW diagnosed with a new STD on date of their visit were more likely to have a diagnosis of bipolar disorder (OR=12.73, p=0.04), to have a history of outpatient mental health treatment (OR=23.50; p=0.003), and to have a STD history (OR=15.38; p=0.02) compared to WSW who were not diagnosed with an STD during their clinical screening.
Compared to WSW with no STD history, WSW with a history of STDs were more likely to have a history of suicide attempt (OR=10.00; p=0.02), inpatient psychiatric/mental health treatment (OR=12.21; p=0.002), and outpatient mental health treatment (OR=6.86; p=0.01), and to have used injected drugs (IDU) during their lifetime (OR=15.09; p=0.009).
While STD rates were lower among WSW compared to their WSM counterparts, consistent with prior research,27-37 5% of WSW in this sample were diagnosed with a new STD on the day of their visit and 17% had a history of one or more prior STDs. Although WSW were less likely than exclusively heterosexual women to engage in “high risk” HIV/STD sexual risk behavior (i.e., sex with men without a condom or latex barrier, condom breakage, anonymous sex partners in the past 3 months), no significant difference was observed in STD diagnosis by sexual orientation after adjusting for age, race/ethnicity, and health insurance. Coupled with the knowledge that many STD infections in women are asymptomatic45-47 and untreated HPV infections can cause Pap smear abnormalities, genital warts, and cervical cancer,28,29 findings from this study suggest that culturally competent screening, diagnostic, and treatment services are indicated for WSW.
WSW were more than three times as likely to have a clinical mental health diagnosis relative to the WSM women in this sample. Consistent with prior research, WSW were especially disproportionately affected by depression, anxiety, and PTSD.7-14 Consistent with prior research on psychosocial health disparities, WSW were more likely to have a history of past suicide attempt(s)13,48 and increased mental health utilization (e.g., inpatient psychiatric/mental health treatment).17,49 Prior research suggests that stressors associated with being WSW, such as leading a marginalized life, hiding one's sexuality, facing verbal, emotional, or physical abuse, or stigma, may contribute to increased rates of mental health diagnoses among women.2-6,50
However, contrary to previous research on psychosocial health disparities, no significant differences on alcohol use, 9, 13, 16-20, 22, 23 tobacco use,15, 23-26 and history of sexual abuse or emotional abuse51were found in this sample between WSW and WSM women after adjusting for age, race/ethnicity, and health insurance. Due to the documented co-occurrence of mental health diagnoses and alcohol use among lesbians,8,26 and PTSD and history of sexual or emotional abuse, 52 it could be that mental health diagnoses are the domain of greatest concern among WSW. Additional research is warranted to examine the pathways to psychosocial disparities, in particular to understand the influence of factors salient to WSW status such as disclosure, stigma, and stress which may affect mental health.5
Prior research with women has documented significant associations between a variety of psychosocial factors and STDs;53-58 however, to date this association has not been adequately documented among WSW specifically. In the current study, WSW diagnosed with a new STD on the date of their visit were at increased odds of having bipolar disorder, utilizing outpatient mental health services, and having an STD history. WSW with a history of STDs had an increased odds of having attempted suicide in the past, utilizing both outpatient and inpatient mental health treatment services, and having a history of injection drug use. Findings suggest that future research examining “intertwined syndemics”59-61 with larger samples of women by sexual behavior may provide valuable data to guide the development of behavioral and sexual health interventions with WSW. In particular, future studies would benefit from examining whether greater numbers of psychosocial health problems are associated with high-risk sexual behavior and STD acquisition, similar to prior research studies among men who have sex with men.59
This study has limitations to consider when interpreting findings. First, symptom status was only determined for patients who were screened for an STD. We could not determine if some symptomatic patients were empirically treated without testing. Ideally, all symptomatic patients would be screened for an STD, but this could not be ascertained by the retrospective chart review used in this study. Second, 15% of women in the FH population did not report gender of their sexual partners, and this “unknown” sexual behavior group had the highest STD rates but were excluded from the study (N=58). The categorization of sexual behavior by WSW and WSM is consistent with the literature suggesting that health disparities are less likely to be related to the gender of one's sexual partner, but may be more related to stigma and/or homophobia that WSW experience, irrespective of whether they have sex with both women and men.3,5 However, future studies with larger samples might benefit from examining differences by WSW, WSW/M, and WSM, differentiating exclusively lesbian and bisexually active women. Third, the primary reasons that women presented for care at FH were not captured (i.e., primary care, mental health, HIV testing specifically). Those patients who received outpatient mental health treatment at Fenway had more information in their medical chart, especially concerning psychosocial history (e.g., history of abuse, suicide attempt, etc.) due to the extensiveness of mental health intake form. Thus, since Fenway is known to provide culturally competent care to sexual and gender minority persons, it is possible that a disproportionate percent of female clients seeking mental health services were WSW. Fourth, data are constrained by the study design itself (i.e., retrospective chart review), which has several limitations including incomplete documentation, missing charts, information that is unrecorded, difficulty interpreting information found in the documents (e.g., jargon, acronyms), problematic verification of information and difficulty establishing cause and effect, and variance in the quality of information recorded by medical professionals.62-66 Lastly, but importantly, lack of statistical power remains a major limitation given the small cell sizes present in bivariate logistic regression procedures used to assess whether WSW diagnosed with an STD on their date of visit and WSW with a history of STDs were more likely to have psychosocial issues, as compared to WSW who were not diagnosed with an STD on the date of their visit and those with no STD history. However, given the dearth of research on this topic, we felt that it was important to take a preliminary look at the psychosocial issues associated with STD diagnosis and history among WSW despite the small sample size and resulting statistical imprecision. Additional research is warranted with larger samples of women.
Limitations notwithstanding, results suggest that all sexually active women, regardless of sexual orientation and behavior, should be routinely screened for STDs. Assumptions that WSW are at low or negligible risk for STDs and other gynecological infections may be premised upon infrequent screening, ignorance of lesbian sexual practices, and/or the discomfort that WSW may feel concerning the disclosure of their sexual identity and behavior.27,35,67 Although some patient charts did not document sexual behavior or sexual identity, the majority (85%) of the sample had known gender of sexual partners, suggesting that clinicians are both talking to women about their sexual behaviors and documenting these interactions in patient charts, though this may be biased since the sample came from Fenway Health, a clinic specializing in LGBT care. Continued provider training is warranted to ensure appropriate screening, diagnosis, and care screening for STDs and other gynecologic infections among WSW. Moreover, research with larger samples is warranted to look more carefully at sexual behavior and sexual health among women, including disease prevalence rates, reasons for screening, and treatment. Future studies would benefit from examining a broad range of health disparities among WSW, especially mental health and psychosocial issues.
Thank you to the Medical and Mental Health Departments at Fenway, and D'hana Perry and Benny Vega who assisted with medical record extraction.
Author Note: The project described was supported by The Center for Population Research in LGBT Health at The Fenway Institute and by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) under Award Number R21HD051178. Some of the investigator time on this project was also supported by grant number R03DA023393 from the National Institute on Drug Abuse (NIDA) (PI: M. Mimiaga).The content is solely the responsibility of the authors and does not necessarily represent the official views of the NICHD, NIDA, or the National Institutes of Health.