Reproductive Coercion, Sexual Risk Behaviors, and Mental Health Symptoms among Young Low-Income Behaviorally Bisexual Women: Implications for Nursing Practice

Abstract

Aims and Objective

To describe prevalence of reproductive coercion, sexual risk behaviors, and mental health symptoms among women reporting lifetime sexual experiences with men and women compared to peers reporting sex exclusively with men.

Background

Reproductive coercion, a global public health problem, is understudied among sexual minority women. Violence against women remains high among women who have sex with women and men. Rates of sexual and physical violence among this population are higher than women reporting exclusive sexual partnerships with either men or women. Nurses and other healthcare providers often do not conduct comprehensive sexual histories; assumptions related to a sex partner’s gender may provide indications of broader health implications.

Design

Cross-sectional survey of low-income Black women ages 18 to 25 recruited from three community-based sites for a parent study focused on intimate partner violence and health.

Methods

We analyzed survey data from participants who reported lifetime sexual experiences with men and women (N=42) and compared their outcomes to those of women reporting sexual experiences with men only (N=107).

Results

A greater proportion of women who have sex with women and men reported experiencing reproductive coercion. Women who have sex with women and men also reported a greater number of lifetime intimate partner physical and sexual violence experiences, traded sex for resources, and had post-traumatic stress disorder symptoms.

Conclusions

Findings provide vital information that can inform nursing clinical practice, specifically related to history-taking, screening protocols, and counseling strategies for intimate partner violence and mental health among women who have sex with women and men.

Relevance to Clinical Practice

Strategies for addressing reproductive coercion and intimate partner violence as well as the health consequences among women who have sex with women and men in clinical and community-based settings should include a longitudinal understanding of sexual behavior and gender of sex partners.

INTRODUCTION

Reproductive coercion (RC), a form of intimate partner violence (IPV), is a significant public health issue and is understudied globally among sexual minority women. Over the last decade, women who have sex with women and men (WSWM), or behaviorally bisexual women, and/or women who identify as bisexual, are consistently recognized as having increased risks for a number of physical and mental health morbidities (WHO 2013a). These include sexually transmitted infections (STIs) (Champion et al. 2005), anxiety, depression, alcohol dependence (Cochran, Mays 2009), digestive problems, chronic fatigue syndrome, heart problems, and obesity (Cochran, Mays 2007). In this paper, women who report having sexual experiences with both men and women, at some point in their lifetime, will be referred to as WSWM. We use the term WSWM rather than bisexual to refer only to sexual behaviors that are sometimes incongruent with identification of sexual orientation or identity (Diamond, 2009).

Reproductive coercion is defined as behaviors “intended to maintain power and control in a relationship… behavior that interferes with contraception use and pregnancy” (American College of Obstetricians and Gynecologists, 2013). RC behaviors are associated with increased risk of HIV and STIs through diverse pathways such as: early sexual initiation, forced sex, multiple sexual partners, use of drugs and alcohol prior to and during intercourse and unprotected intercourse (Decker et al. 2014, Fontenot et al. 2014, Heintz, Melendez 2006). RC behaviors include threats to leave or have a child with a different sexual partner, forced or pressured sexual activity without contraception including condoms, or clandestine removal of condoms during sexual intercourse in an effort to impregnate the female partner (Miller et al. 2010). Young WSWM are at increased risk for pregnancy because adherence to contraceptive behaviors may change depending on the gender of their partners during a particular time period (Alexander, Fannin 2014). Additionally, emerging adult women, aged 18 to 25, continue to have high numbers of unintended pregnancies due to ambivalent childbearing motivations (Higgins, Popkin & Santelli 2012).

Globally, intimate partner violence has been found to be a consistent risk factor for unintended or unwanted pregnancy and can lead to detrimental consequences for women’s health (Pallitto et al. 2013). Unintended pregnancy has been linked to maternal mental health problems such as depression during pregnancy (Dibaba, Fantahun & Hindin 2013) as well as child mental health and behavioral problems such as aggression, delinquency, and substance use (Hayatbakhsh et al., 2011). While it has only recently been studied outside the United States, RC and its implications for women’s health have been described in other countries such as Jordan (McCleary-Sills 2013) and Côte d’Ivoire (McCauley et al. 2014a). However, we found only one study that described RC or its implications for mental and sexual health outcomes in WSWM (McCauley et al. 2014b). Specifically, the experiences of low-income Black young adult WSWM living in a United States city have not been adequately described.

Despite the existing knowledge that sexual orientation encompasses three dimensions, identity, behaviors, and attraction, researchers most often use identity to report their findings (Johns, Zimmerman & Bauermeister 2013, Bauer, Jairam & Baidoobonso 2010). This practice has created a significant knowledge gap about the lives of WSWM who do not identify as bisexual. Thus, we aimed to describe, among WSWM compared to women who report having sex with men only, RC experiences, their mental health outcomes, and sexual risk behaviors. We approached data analysis and interpretation in this paper using an intersectionality theoretical approach (Hill Collins 1990). This approach provided a foundation for examining the lives of marginalized young women whose class position, racial identity, sexual behaviors, and gender influence their health outcomes (Bowleg 2008, Bowleg et al. 2003). In this paper, we will also address the implications for nurses in clinical and community-based settings.

BACKGROUND

Reproductive coercion includes: (1) pregnancy coercion, such as threatening to harm a woman physically or psychologically (e.g., with infidelity or abandonment) if she does not become pregnant; (2) birth control sabotage, such as intentionally breaking or removing condoms; and (3) control of pregnancy outcomes such as threats related to pregnancy continuation or termination (Miller, Silverman 2010). There are significant associations between IPV and RC. For example, evidence suggests increased rates of unintended/unwanted pregnancies, pregnancy termination, contraceptive methods change, and inconsistent use of condoms are some observed negative reproductive health outcomes among women reporting RC and IPV (Fantasia et al. 2012, Miller et al. 2010, Pallitto et al. 2013). In one of the few large-scale prevalence studies examining RC and IPV among women attending California family planning clinics, Miller and colleagues (2011) found 53% of respondents reported physical or sexual violence, 19% experienced pregnancy coercion, and 15% reported birth control sabotage. To date, studies about RC and IPV have been conducted among heterosexual women and primarily in clinical settings.

Screening for IPV among women ages 14 to 46 is a highly recommended intervention strategy to reduce victimization (WHO 2013b; Moyer 2013; Singh 2014); and healthcare providers in general, and nurses in particular are often the first point of contact for women experiencing IPV. However, evidence suggests nurses and other providers are often reluctant to inquire about IPV (Miller et al. 2015, Valpied, Hegarty 2015) and obtain sexual histories (Quinn, Happell & Browne 2011). Insufficient or absence of training, discomfort about asking questions related to sexual behaviors, sexual orientation or IPV, and time constraints are some reasons given by healthcare providers for not taking a comprehensive sexual history (Althof et al. 2013) or screening for IPV (LoGiudice 2014). Nurses have significant roles in understanding and addressing IPV and RC that are shaped by multiple social inequalities present in the lives of young Black WSWM including sexual orientation/identity, gender, race, and ethnicity.

IPV and RC

According to a recent population-based study and review of the literature, sexual minority women experience IPV at similar or higher rates than heterosexual women (Walters, Chen & Breiding 2013, Edwards, Sylaska & Neal 2015). Additionally, compared to lesbians, bisexual women are more likely to report lifetime IPV experiences. Lifetime violence experiences including sexual assault have been reported by lesbian and bisexual women at rates of 16 to 85%; this represents significantly higher proportions for most types of violence (Rothman, Exner, & Baughman 2011). In addition, in a U.S.-based national study, researchers found Black women had higher adult lifetime risks of violence and prevalence of violence than White, Hispanic, and Asian American women (Moracco et al. 2007). Of note, some methodological limitations across studies prevent a full understanding of IPV prevalence in this population. For example, few studies use consistent measures to assess IPV types or explicitly evaluate the identified perpetrator’s gender (Edwards, Sylaska, & Neal 2015).

The implications for experiences of IPV on physical and mental health are well-known (Campbell 2002) and efforts to screen and develop intervention programs are emerging. However, these efforts primarily target heterosexual women (Ard, Makadon 2011). Lesbian and bisexual women face unique challenges to seeking formal support and help from healthcare providers who most often function under a heteronormative medical framework; in addition, stigma associated with sexual identity in addition to ongoing or historical experiences of IPV, may pressure women not to seek help (Carvalho et al. 2011). This is more difficult to address in racially marginalized Black lesbian or bisexual women where the nature and frequency of multiple minority stressful life experiences (sexism, racism, and heterosexism) intersect to shape their lives differently (Bowleg 2008, Bowleg et al. 2003).

Sexual Health Outcomes among WSWM

Sexual health and sexual risk behaviors among WSWM have rarely been studied (Bauer, Jairam & Baidoobonso 2010). Most existing research is comprised of convenience samples and has focused on identifying HIV/STI prevalence, condom use, and/or sexual risk behaviors in individuals who identify as LGBT or as a sexual minority (Heintz, Melendez 2006, Rothman, Exner & Baughman 2011, Herrick et al. 2013). Individuals in these studies reported higher rates of STIs, unintended pregnancies, multiple sex partners, unprotected sex, and sex under the influence of alcohol or drugs than heterosexually-identified persons. Research consistently uses identity as a marker for behavior, usually comparing sexual minorities and/or LGBT to heterosexuals (Kuyper, Vanwesenbeeck 2011, Satinsky, Jozkowski 2014).

Irrespective of whether compared by identity or behavior, women who identify as bisexual, behaviorally bisexual or WSWM have poorer sexual health outcomes and exhibit riskier sexual behaviors (WHO 2013a). For example, this group has reported higher rates of STIs compared to women who identify exclusively as heterosexual or exclusively lesbian (Schick et al. 2015, Tao 2008, Everett 2013). Additionally, heterosexually-identified women who reported same-sex relationships in the past year were found to have significantly more male sex partners (with a median lifetime of 10 male partners), engaged in more sex while using drugs, initiated vaginal intercourse at a younger age, and were less likely to use condoms when they last had vaginal sex, than women who identified as heterosexual, bisexual, or lesbian (Bauer, Jairam & Baidoobonso 2010).

Despite the higher rates of STIs reported, WSWM face great barriers in accessing and receiving health care. Researchers conducting a national, population-based survey uncovered barriers such as lack of insurance or primary care provider (Ward et al. 2014). Whereas, some studies with convenience samples found stigma and discrimination were obstacles to attaining health care services (Meyer 2003, Martin-Storey & August 2015). These barriers are intensified for low-income, Black WSWM (Bowleg 2008).

Mental Health Outcomes among WSWM

Mental health disparities between sexual minorities and heterosexuals are well documented in the literature. Recent studies indicate WSWM have significantly higher rates of depression than other sexual minorities and heterosexual women, regardless of whether compared by sexual identity or behavior (Bostwick et al. 2010, Pyra et al. 2014). In addition, bisexual youth (aged 16-25) are found to have significantly higher levels of depression, PTSD, and suicidal ideation than those older than 25 years (Ross et al. 2014).

Women who show dissonance between their sexual identity and behavior such as women who identify as heterosexual and have same-sex partners, report more psychological distress and mental health problems as a result of having to navigate different spaces: a heterosexual space associated with a more socially privileged identity, and a lesbian or bisexual space associated with stigma, discrimination, and underprivileged identity. In addition, heteronormative cultural biases that exist in most societies assume a stable sexual identity over time. When women show flexibility in sexual identity and behaviors, the result can include additional psychological distress, stigma, and rejection (Schick et al. 2012). Further, harassment due to gender non-conformity in sexual minorities was found to mediate the association between sexual minority identity and depressive symptoms (Martin-Storey, August 2015).

In summary, the health risks faced by WSWM are often overlooked in the current research literature. Perhaps, this is due to ambiguities that exist relating to how a woman chooses to identify herself (such as heterosexual, bisexual, or lesbian) and her sexual behavior. As research and practice move forward to fully examine the lives of WSWM, we should aim to separate these two concepts in an effort to understand the best ways to promote optimal health among this population. Thus, in this paper, we compared mental, physical, and sexual health risk among WSWM to these outcomes among their peers who engage in sex with men only.

METHODS

Design

This exploratory, descriptive study includes data from 149 women: 42 women (27.8% of parent study sample) who reported lifetime gender of sex partners as “equally men and women” (n=17) or “mostly men” (n=25) were categorized as WSWM. These women were compared to 107 women who reported sex with “men only” or Women who have sex with men (WSM). Participants ranged in age between 18 to 25 years old. The community-based sample was recruited from six sites: three Women, Infants, and Children (WIC) nutrition distribution centers, one community-based organization that focused on linking low-income individuals to health insurance and health services, and two youth employment/educational centers. These sites were selected in the parent study to involve young women who were not specifically seeking health services. We used flyers and word-of-mouth advertising by community members, recruitment site staff, and participants to recruit participants to the study.

Data Collection

Procedures

The institutional review board of BLINDED approved all research activities. Data were collected between February 2014 and July 2014 (6 months). Data analyses were conducted using a subsample from the parent research study (BLINDED, unpublished). A research assistant recruited women at each of the community-based sites. Women were initially screened to determine if they met study eligibility criteria: 1) self-identify as Black or African-American; 2) aged 18 to 25; and 3) report sexual activity with a man in the previous 6 months. The majority of women deemed ineligible reported sex with women only in the previous 6 months. We obtained oral informed consent due to the sensitive subject of the study. Participants completed the survey using a computer tablet at the recruitment site. After survey completion, participants were assessed for emotional distress by the research assistants and provided a resource list that included support services for IPV, health care, substance abuse, and mental health. Participants received $20 remuneration for their time.

Measures

We assessed lifetime RC and mental health symptoms using three assessment measures. RC was the dependent variable. We assessed lifetime experiences of reproductive coercion using 10 items developed by Elizabeth Miller and colleagues (Miller et al. 2011). Women who endorsed one or more of these items were classified as experiencing RC. We measured participant self-reported depressive symptoms using the Center for Epidemiologic Studies Depression Scale (CESD-10) (Andresen et al. 1994). This brief screening tool assessed for levels of past-week depressive symptoms (Range 0 – 29). Scores were summed and categorized into either not suggestive depression (score <10) or having symptoms suggestive of depression (score 10+). This cut-point was previously established in the literature as an indicator of depression symptomatology (Andresen et al. 1994). We assessed PTSD symptoms using the 4-item Primary Care Post-traumatic Stress Disorder Screening (PC-PTSD) (Prins et al. 2003). The PC-PTSD measured PTSD symptoms (intrusive re-experiencing, numbing, avoidance, and hyperarousal) in the past month. Scores range from 0 to 4 and we used a cutoff score of 3 or higher to specify clinically significant PTSD symptoms.

We assessed correlates in four domains: socio-demographics, lifetime experiences of physical and sexual violence, sexual risk behavior, and mental health symptoms. We used single items to assess demographic characteristics, including age, race, relationship status, education level, and number of pregnancies. Language in the survey referred to intimate partners as “someone you were dating or going out with”. To assess lifetime experience of physical and sexual violence, we adapted items from the Sexual Coercion (2-items) and Physical Assault (1-item) subscales of the Conflict Tactics Scale 2-Short form (Straus al. 1996, Strau & Douglas 2004). The CTS short form is comparable in concurrent validity to the full scale (.69 for physical assault scale and .67 for sexual assault) (Straus & Douglas 2004).

Sexual Risk was assessed using six items: reported anal and vaginal sex in the previous 3 months, condom use during sexual activity in the previous 3 months, inability to use condoms when she wanted to use one, STI diagnosis, and trading sex for money, drugs, shelter, gifts, or other resources.

Data Management and Analyses

De-identified survey data were exported from Excel into SPSS 23 and checked with double entry procedures. Data from the parent study (N=151) were cleaned and recoded into numerical variables in SPSS 23. Participants who did not respond to any reproductive coercion questions were excluded (N=2). Women who reported that sexual partners were “mostly men” (N=25) were combined with women who reported “equally men and women” (N=17) for the group WSWM (N=42). We conducted descriptive statistics to characterize the sample. Chi-square and t-tests were used to analyze differences in lifetime experiences of intimate partner violence, reproductive coercion, sexual behaviors, and mental health between WSM and WSWM.

RESULTS

The mean age of the sample was 21.3 years old (SD 2.10). The vast majority of the sample (79%) reported their education ended before or immediately following completion of high school and had experienced at least one pregnancy (76%). Slightly more than half (57%) of the sample reported they were dating only one person or in a serious relationship. No statistically significant differences in demographics existed between the two groups, WSWM and WSM (Table 1).

Table 1.

Demographic Characteristics

Variables	Total Sample N= 149	WSM No.(%) N=107 (71.8)	WSWM No.(%) N=42 (28.2)	Chi-Square or T-tests for difference
	N (%)	N (%)
Age (M/SD)	21.29 (2.10)	21.13 (2.15)	21.29 (2.11)	.146
Race				1.202 (2) p=.548
- Black/ African American	146 (98.0)	104 (97.2)	42 (100)
- Multiracial	2 (1.3)	2 (1.9)	0 (0)
- American Indian	1 (0.7)	1 (0.9)	0 (0)
Relationship Status				2.397 (4) p=.663
- Dating only one person/in  a serious relationship	86 (57.0)	64 (59.8)	21 (50.0)
- Single	46 (30.9)	32 (29.9)	14 (33.3)
- Married	8 (5.4)	5 (4.7)	3 (7.1)
- Dating more than one  person	6 (4.0)	3 (2.8)	3 (7.1)
- Don’t know	4 (2.7)	3 (2.8)	1 (2.4)
Education Level				7.004 (5) p=.220
- Less than 12th grade	77 (50.3)	55 (51.4)	20 (47.6)
- Finished High School	43 (28.9)	33 (30.8)	10 (23.8)
- Some College	23 (15.4)	13 (12.1)	10 23.8)
- Finished college or grad  school	4 (2.7)	3 (2.8)	1 (2.4)
- Not applicable	3 (2.0)	3 (2.8)	0
Total Pregnancies				4.1 (5)
- One	46 (30.9)	32 (29.9)	14(33.3)
- Two or more	68 (45.6)	46(43)	22 (52.4)

Participants in the WSWM group compared to those in the WSM group reported more lifetime incidents of physical IPV (71.4% vs. 39.3%; chi-square = 12.689 (df2), p=0.002) and sexual IPV without using force or threats (40.5% vs. 10.4%; chi-square =17.981 (df2), p=.000). However, there were no differences between the groups related to experiences of sexual violence using force or threats (26.2% vs. 15.0%; chi-square= 2.889 (df2), p=.236).

We analyzed the most prevalent lifetime reproductive coercion items reported among the 57 participants who endorsed at least one RC experience from the parent study (Table 3): 1) told by male partner not to use birth control (n=24; 42.1%); 2) male partner removed condom during sex (n=34; 59.6%); and 3) male partner forced or pressured sex to promote pregnancy (n=25; 43.9%), because endorsement frequencies of the other items were less than 23%. Of these three items, WSWM were more likely to have been told by a male partner not to use birth control than WSM [Chi-square= 7.107 (df2), p=.029]. There were no differences between groups in experiencing other forms of reproductive coercion. Sexual risk behavior differed between the two groups in recent condom use and trading sex for money, drugs, shelter, gifts, or other resources. WSWM were less likely to use condoms in the previous three months than reported by WSM, however, the survey did not allow participants to report the gender of their sex partners during that time period. Almost one-third of WSWM reported trading sex for resources in the previous three months and were more likely to report such activity than WSM [10.833, (df 2), p=.004] (Table 4).

Table 3.

Differences in Lifetime Reproductive Coercion between WSWM and WSM

Lifetime Reproductive Coercion
Variables	Total Sample N= 149	WSM No.(%) N=107	WSWM No.(%) N=42	Chi Square
Told by male partner not to use birth control	24 (16.3)	12(11.4)	12(28.6)	7.107 (df2) p=.029
Force or pressure by male partner to become pregnant	25 (16.8)	14(13.1)	11(26.2)	3.816 (df2) p=.148
Male partner removed condom during sex without consent so you would get pregnant	34 (23.1)	21(20.0)	13(31.0)	2.106 (DF2) p=.349

Table 4.

Differences in Sexual Risk Behaviors between WSWM and WSM

Sexual Risk Behaviors
Variables	Total Sample N= 149	WSM No.(%) N=107	WSWM No.(%) N=42	Pearson Chi- Squared
Vaginal sex (last 3 months)	139 (93.9)	101 (95.3)	38 (90.5)	1.217 p=0.270
If vaginal sex in last 3 months -never, rarely, or sometimes) used a condom for vaginal sex	93* (67)	62 (62.4)	31 (81)	6.182 p=.013
Anal sex (past 3 months)	22 (14.8)	13 (12.1)	9 (21.4)	2.063 p=.151
STI diagnosis (lifetime)	64 (43.0)	41(38.3)	23 (54.8)	3.834 (df2) p=.147
Had sex without a condom when wanted to use one (past 3 months)	58 (38.9)	39(36.4)	19(45.2)	1.740 (df2) p=.419
Traded sex for money, drugs, shelter, gifts, or other resources	21 (14.1)	9 (8.4)	12 (28.6)	10.833 (df 2) p=.004***

^*analysis from data of those that reported vaginal sex in last 3 months and one participant not included due to response “I don’ t know” (n=138)

Both groups reported poor mental health (Table 5); however, WSWM reported more positive screens for PTSD [chi-square=4.312, 9df=20, p=.038] which may concur with their higher incidence of reported IPV. In both groups, approximately 40% had a positive screen for depression; but between the groups there were no significant differences.

Table 5.

Differences in Mental Health Outcomes between WSWM and WSM

Mental Health Outcomes
Variables	Total N (%) N=149	WSM No.(%) N=107 (71.8%)	WSWM No.(%) N=42 (28.2%)	T-tests or Pearson’s Chi- Square (2- tailed)
PTSD Score [M (SD)]	1.58 (1.47)	1.40 (1.45)	2.02 (1.44)	T=−0.236 (p=.020)*
PTSD (positive screen)	94 (63.1%)	62 (57.9%)	32 (76.2%)	4.312 (p=.038)*
Depression score [M (SD)]	9.80 (5.50)	9.54 (5.75)	10.45 (4.80)	t= −0.922 (p=.363)
Depression (CES-(positive screen)	64 (43.2%)	45 (42.5%)	19 (45.2%)	.095(NS)

^*P < .05

^**P < .01

^***P< .001

DISCUSSION

We examined associations between reproductive coercion, sexual risk behaviors, and mental health outcomes among low-income emerging adult Black WSWM living in a United States city. In our analysis, WSWM and WSM had several similarities including experiencing high levels of intimate partner violence and poor mental health. However, significant differences emerged that warrant further research. Due to the exploratory nature of this study, we interpreted the results with caution.

To our knowledge, RC has limited examination as a phenomenon occurring in this population of women and findings from this study represent a new step to beginning this conversation. The only RC behavior that showed statistically significant differences between groups was being told not to use birth control by a male partner in order to get pregnant. This behavior highlights the intersection of reproductive and sexual health that is unique to the WSWM experience. Though not statistically significant, WSWM were slightly more likely to report lifetime experience of two or more pregnancies (52.4%) than WSM (43%) in the sample. This result is aligned with previous literature, finding that pregnancy is common among young sexual minority women (Herrick et al. 2013, McCauley et al. 2014b). In fact, previous studies have demonstrated that sexual minority youth involved in a pregnancy report facing social pressures to conform to heterosexist expectations of relationships (Travers, Newton & Munro 2012). Additionally, low-income Black sexual minority women may face barriers to quality and accessible sexual and reproductive health care because they are less likely to have health insurance or disclose their sexual behaviors to a health care provider (ACOG Committee on Health Care for Underserved Women 2012).

Disparities among sexual risk behaviors between the two groups indicate a renewed urgency to target this unique population. WSWM were less likely to report frequent condom use during vaginal sex. Though not statistically significant (p=.151), WSWM in this study were also more likely to have engaged in recent anal sex (21.4%) and to have had a STI diagnosis (54.8%) than WSM (12.1% vs. 38.3% respectively). This finding is aligned with recent research that shows African-American WSWM women were more likely to report having a STI and engage in heightened sexual risk-taking behaviors when compared to African-American WSW in a clinical setting (Muzney et al. 2011). Additionally, in a racially diverse sample, Ward and colleagues (2014) reported more STI diagnoses among WSWM compared to heterosexual women (Tao 2008, Everett 2013). In conjunction with the participants’ higher reported rates of IPV and RC and mental health symptoms, these findings may potentially demonstrate lack of self-efficacy to advocate for condoms or communicate with their male partners (Seth et al. 2009). This highlights the potential for power imbalances and fear of condom negotiations in these young women’s relationships with men. RC is a marker for HIV risk though because it is not direct, it is often difficult to measure.

The WSWM in this sample also demonstrated greater rates of violence experiences overall including both physical and sexual violence. In particular, the WSWM reported more experiences with sexual violence without force, indicating the prevalence of coercion within sexual interactions. The United States National Intimate Partner and Sexual Violence Survey defines sexual coercion as “…unwanted vaginal, oral, or anal sex after being pressured in ways that included being worn down by someone who repeatedly asked for sex or showed they were unhappy; feeling pressured by being lied to, being told promises that were untrue, having someone threaten to end a relationship or spread rumors; and sexual pressure due to someone using their influence or authority” (Walters, Chen & Breiding 2013). This contrasts with a definition of forced sex that includes completed penetration that may or may not be facilitated by alcohol or drugs. While WSWM experienced both types of sexual violence at higher rates than WSM, the statistically significant finding of sexual violence without force warrants further exploration. This finding highlights the importance of carefully measuring RC at the intersection of psychological abuse and other forms of IPV.

We also observed high rates of mental health symptoms in this population of young women compared to rates in a global population of youth ranging from 8% to 57% (mean percentage of 20%) (Patel et al. 2007). Specifically, the rate of reported PTSD symptoms was significantly higher when compared to WSM. The processes for stress and coping, while not measured in this sample, might be manifested in the observed differences in mental health outcomes noted among WSWM. Physical abuse (lifetime) as well as sexual abuse with and without force all correlated with PTSD and depression (all scales). This association has also been demonstrated in other studies; in addition to the stress and negative mental health outcomes related to IPV, sexual minority women and in particular WSWM might experience additional internalized stressors with respect to stigma and sexual identity/behavior incongruence (Schick et al. 2011). In this sample, we must also consider that stresses associated with poverty might also contribute to manifestations of depression and PTSD symptoms (Gilroy et al. 2015).

We can interpret these findings using intersectionality (Crenshaw 1991) and minority stress theories (Meyer 2003) as guiding frameworks for practice and future research. These theoretical perspectives recognize some WSWM contend with power imbalances, inequities, and oppressions that influence sexual decision-making. These experiences are unique to the intersectionality of their gender, race, sexual orientation, and sexual experiences (Bowleg 2012). Thus, living within these circumstances may create exposure to stressors that are consequences to intersecting environmental influences, including the WSWM’s minority status.

RC is a consequence to these environmental influences and is gender-based (Reed et al. 2010)

This community-based sample of young, Black WSWM provides insight into several reasons for the invisibility of this population in nursing practice. The young women in this study may face multiple challenges to attaining optimal health due to their status as minorities (Black race and sexual behaviors with men and women) and the low economic circumstances in which they live (Bowleg 2012). The context in which sexual relationships are formed among low-income WSWM is an important factor in determining risk profiles as evidenced by the differences in sex trading behaviors reported between WSM and WSWM (Adimora, Schoenbach 2005).

LIMITATIONS

There were several limitations to this study that should be considered when interpreting the findings. The small sample size precludes generalizability to other WSWM living in diverse geographic regions. Additionally, this study represents a secondary analysis of data collected for other purposes that did not specifically target WSWM. In the survey, we did not ask how the participant identified (heterosexual, lesbian, bisexual, queer, etc.). We also did not ask about participants’ intentions for pregnancy. We were not able to determine the gender of the partner that committed RC or IPV against the participant nor were we able to determine when the behaviors occurred. Furthermore, classification of participants as having experienced RC based on the inventory developed by Miller and colleagues (2011) has no measures for determining severity or pattern of the behaviors.

The findings represent suggestive patterns of behaviors and experiences among the sample. We relied on retrospective reporting by study participants that could influence the accuracy of responses due to limitations in memory recall and potential social desirability. Future research is needed and should include larger, probability-sampling among community-based populations. However, examining the experiences of RC among this group can likely be translated to other populations of disadvantaged women in a global context. Using intersectionality and minority stress theories as guiding frameworks, the limitations of data from this study can transcend location to ignite practice improvement initiatives and further research efforts among this often-invisible group of women.

CONCLUSION

Our findings demonstrate that WSWM are at increased risk for reproductive coercion and physical and sexual IPV compared to their peers only having sex with men. We also found higher rates of mental health symptoms, specifically PTSD symptoms among WSWM compared to WSM. Our findings suggest that greater screening for IPV and HIV using trauma-informed perspectives are necessary. RC is a contributing factor and persistent challenge to WSWM who wish to control their sexual and reproductive autonomy.

RELEVANCE TO CLINICAL PRACTICE

Based on our findings, we suggest that nursing practice approaches to WSWM should be informed by theories of intersectionality and minority stress. These theories can assist nurses to understand the interdependent processes within which sexual decisions operate and within which stress and multiple sexual identities, behaviors, and relationships are managed (Lewis et al. 2012). Within a context of historical and/or current IPV or RC experiences, nurses, therefore, should advocate for and implement systematic screening and psychoeducation for RC and mental health symptoms among women in all healthcare settings. These conversations should integrate renewed emphasis and teaching about coercion as a factor for HIV and STI transmission risk.

Our findings challenge the current trend in sexual history-taking. Nurses currently ask about numbers, types, and genders of partners in the months prior to the healthcare encounter. However, knowledge of lifetime gender of sex partners may provide new strategies for counseling and identification of RC from partners of either gender. Nurses should employ health-counseling strategies to include information about reproductive coercion as a form of intimate partner violence and its implications for negative mental and sexual health outcomes.

Using the findings from this study, a nurse might consider asking the following questions to WSWM about reproductive coercion:

Even if you are currently having or recently had sex with a woman,

1. Do you have any plans or desires to become pregnant in the next year?

2. Have any of your partners, past or present, pressured you to have sex in order for you to get pregnant? Pressure can include repeatedly asking for sex, threatening to leave you or spread rumors about you, or lying to you.

3. How do you change contraception you use depending on the gender of your sex partner?

4. Are you more or less likely to use a condom after you have had a fight with your sex partner?

We suggest that nurses should conduct a comprehensive sexual and reproductive history that avoids assumptions related to a sex partner’s gender every time a woman visits a health care provider. Non-judgmental discussions that give longitudinal insight into relationship experiences and associated consequences could provide valuable information to nurses, guiding tailored assessment and counseling strategies that meet WSWM needs.

Table 2.

Differences in Lifetime IPV Experiences between WSWM and WSM

Lifetime Intimate Partner Violence Experiences
Variables	Total Sample N= 149	WSM No.(%) N=107	WSWM No.(%) N=42	Chi- square (value (df)
Physical Violence	72 (48.3)	42(39.3)	30 (71.4)	12.689 (df2) .002***
Sexual Violence using force or threats	27 (18.1)	16(15.0)	11(26.2)	2.889 (df2) p=.236
Sexual Violence without using force or threats	28 (18.9)	11(10.4)	17(40.5)	17.981 (df2) p=.000***

What Does this Paper Contribute to the Wider Global Clinical Community?

• WSWM are at higher risk for reproductive coercion, intimate partner violence, and PTSD than women who have sex with men only.

• Health counseling strategies should include information about reproductive coercion as a form of intimate partner violence and its implications for negative mental and sexual health outcomes.

• The gender of WSWM sex partners may change throughout the life course and thus approaches to sexual history taking in health settings should include discussions that give longitudinal insight into relationship experiences.

• Healthcare providers can address the unique issues of WSWM when they are positioned within the diverse contexts of race and ethnicity, gender, class, and sexual identity or behaviors.