Abstract
Reproductive coercion (RC) is a type of intimate partner violence that includes birth control sabotage (BCS). We explored the perceived intent behind BCS to refine RC measurement, using a mixed-methods design with a clinic-based sample of Latina women (13 interviews; 482 surveys). Women perceived partners used BCS for reasons beyond pregnancy promotion. Specifically, 16.8% of participants reported any past-year RC; this decreased to 9.5% when asked if their partner used BCS with the sole intent of getting them pregnant. RC measures and assessment should separate behavior from intent in BCS questions to not underestimate the prevalence and to guide clinical response.
Reproductive coercion (RC), the interference with women's autonomous reproductive health decision-making usually by a partner, has emerged over the last decade as a distinct constellation of behaviors that influence women's reproductive health (Miller, Decker, McCauley, Levenson, et al., 2010; Miller, Decker, McCauley, Tancredi, et al., 2010; Moore et al., 2010). Three domains of potentially harmful behaviors make up RC: (a) pregnancy coercion, (b) controlling the outcome of a pregnancy either by coercing or preventing abortion, and (c) birth control sabotage (BCS), which may include condom refusal, surreptitious removal of a condom during sex, poking holes in or damaging condoms, forcibly removing an IUD, or throwing away or preventing access to birth control, among other behaviors. Reproductive autonomy, the ability to control decisions about contraceptive use and pregnancy timing and outcome (Upadhyay et al., 2014), is threatened by the experience of RC. A number of negative reproductive health outcomes are associated with RC, such as unintended pregnancy (Miller, Decker, McCauley, Tancredi, et al., 2010), sexually transmitted infections (Northridge et al., 2017), multiple partners (Katz et al., 2017), condomless sex (Capasso et al., 2019; Fasula et al., 2018), partner rape resulting in pregnancy (Basile et al., 2018), and other forms of partner violence. RC is a type of intimate partner violence (IPV) and is often accompanied by very severe forms of IPV and other coercive behaviors characterized as sexual, financial, or social coercion (Bagwell-Gray et al., 2021; Grace, Perrin, et al., 2020; PettyJohn et al., 2021; Swan et al., 2021; Tarzia & Hegarty, 2021).
Studies of men and their motivations for RC behaviors are limited, but evidence points to desire for connection (many times due to housing instability or impending incarceration) and biological kinship or legacy, perpetuation of narratives of dominance and control, and expression of physical attraction (Alexander, Sanders et al., 2019; Nikolajski et al., 2015; Thaller, 2017). Regardless of motivation, the intent to promote or discourage pregnancy is inherent in the first two domains (pregnancy coercion and controlling the outcome of a pregnancy). BCS behaviors, however, involve actions that may be intended to promote pregnancy or may have other intentions, thus, the inclusion of intent in questions on BCS affects measurement (e.g., “Has your partner done [behavior] so that you would get pregnant?” as opposed to “Has your partner done [behavior]?”). Recent studies have begun adapting RC measures to distinguish intent from behavior, reasoning that a woman might not be aware of her partner's intent, and that intent may be irrelevant, as the behavior is still potentially harmful regardless of intent (Katz et al., 2017; Katz & LaRose, 2019; Katz & Sutherland, 2017). Whether pregnancy-promoting intent should be considered an essential part of the definition of RC is currently a subject of debate among RC researchers (Tarzia & Hegarty, 2021).
RC measures covering the three RC domains were initially developed by Miller et al. (Miller et al., 2011; Miller, Decker, McCauley, Tancredi, et al., 2010) and included 9–11 items, with reliability coefficients in subsequent studies ranging from 0.66 (Sutherland et al., 2015) to 0.91 (Willie et al., 2017) and the majority between 0.70 and 0.80 (Dick et al., 2014; Holliday et al., 2017; Kazmerski et al., 2015; Miller et al., 2014, 2016; Tancredi et al., 2015). In 2017, a refined short-form using item response theory resulted in the Reproductive Coercion Scale, consisting of five items (McCauley et al., 2017). Other studies have used a variety of adaptations and new measures to study RC, ranging from 1 (Samankasikorn et al., 2019) to 24 (Liu et al., 2016) original items, and 3 (Rosenfeld et al., 2017) to 14 adapted (from Miller, Decker, McCauley, Tancredi, et al., 2010) items (Clark et al., 2014), as well as five original items designed for adolescents (Nemeth et al., 2019) and an 8-point RC frequency scale (Cronbach alpha = 0.69) (Fleury-Steiner & Miller, 2020). Most of these studies include some measure of BCS; some ask about BCS behavior in general while others ask about BCS that has occurred with the sole intent of promoting pregnancy. The focus on intent behind BCS may underestimate true RC prevalence.
RC has been studied in a wide variety of populations including Black women (Alexander, Willie, et al., 2019; Capasso et al., 2019; Willie et al., 2017), women with HIV (Anderson et al., 2017), adolescents (Dick et al., 2014; Hill et al., 2019; Northridge et al., 2017), college students (Grace, Perrin, et al., 2020; Katz et al., 2017; Sutherland et al., 2015; Swan et al., 2021), young couples (Willie et al., 2019), women veterans (Rosenfeld et al., 2017), sexual minority women (Alexander et al., 2016; McCauley et al., 2015), and women who use drugs (Perry et al., 2018), but few studies focus on Latina women (Grace et al., 2020a, 2020b). The authors, however, acknowledge that Latina ethnicity does not preclude their representation in some of the previous research. Some aspects of the lives of Latina women may be represented within these populations, a core foundation of intersectionality theory (Crenshaw, 1989).
Lifetime prevalence of RC in population-based studies is approximately 8% (Basile et al., 2019; Black et al., 2011). Latina women are known to experience higher rates of RC compared to white women (Basile et al., 2019; Miller et al., 2014; Sutherland et al., 2015) and also have higher rates of some reproductive health outcomes associated with RC compared to white women, such as unintended pregnancy (Finer & Zolna, 2016), sexually transmitted infection and associated fertility complications (Chambers et al., 2018), and use of less effective contraception (Dehlendorf et al., 2014). Complex reasons underlie these health disparities, including racism and discrimination, levels of acculturation, healthcare access barriers and quality of care, and mistrust of birth control related to historical abuses (Dehlendorf et al., 2014; Pew Research Center, 2021; Roncancio et al., 2012). Prevalence of RC in community samples of Latina women ranges from 14% (Clark et al., 2014) to 17% (Miller, Decker, McCauley, Tancredi, et al., 2010; Sutherland et al., 2015). Focusing on RC in Latina women may help reveal some sociocultural contexts that may explain some of these associations (Grace et al., 2020a, 2020b). Thus, the purpose of this mixed-methods study was to explore with Latina women how they perceived intent behind BCS behaviors and examine implications for RC measurement in research and clinical practice.
Methods
This was an exploratory sequential mixed-methods study, conducted between May 2017 and August 2018 at four locations of a Federally Qualified Health Center (FQHC) serving low-income, primarily immigrant residents in the Washington, DC area. This methodology allows for in-depth exploration of themes qualitatively, which then informs quantitative data collection.
Data Collection
Qualitative Methods
For the qualitative phase, a purposive sample of participants was recruited using several methods. Because of the known association between RC and IPV, women were referred by social services providers if they reported IPV during a routine screening and expressed interest in the study. Women were also asked by clinic providers and staff if they were interested in talking to a researcher about a study when they sought health services. Flyers advertising the study were posted in several community locations. Eligible women were between the ages of 18 and 45, self-identified as Latina, Hispanic or Spanish, and answered “yes” to any of the lifetime RC screening questions (Miller, Decker, McCauley, Tancredi, et al., 2010). Researchers screened women for study eligibility by phone or in-person. Qualitative analysis was ongoing during the recruitment process, and recruitment continued until thematic saturation was reached, after 13 interviews (Sandelowski, 1995).
The primary researcher and bilingual research assistants conducted qualitative interviews in English and Spanish. Interviews were conducted at the health clinics, participants’ homes, community locations, and by telephone, based on participant preference, and lasted approximately 1 hour. Interviews were audio recorded with participant consent, and recordings were transcribed by the primary researcher or by a professional Spanish translator, who also translated the transcripts into English. Interviews were semistructured and followed a suggested guide that covered topics related to RC and IPV.
Quantitative Methods
The quantitative survey underwent extensive field testing prior to implementation. Eligible women were between the ages of 15 and 45, self-identified as Latina, Hispanic or Spanish, and had a dating or sexual partner in the past year. Research assistants who were fluent in Spanish and English distributed flyers in clinic waiting rooms and interested women were screened for eligibility and completed the survey in Spanish or English on a tablet computer with audio-assistance capability. Survey data were collected at three locations of the same FQHC.
Measures
Measures included demographic characteristics such as age, education, nativity, employment, and years in the United States. Demographic variables included age, education (“What is the highest level of education that you have completed?,” categories condensed due to small cell sizes), nativity (“Were you born in the United States?”), employment (“Do you currently have a paying job?”), and years in the United States (“How long have you lived in the United States?”, categories condensed due to small cell sizes). The measures were translated and back-translated and tested with cognitive interviewing.
Reproductive Coercion
RC was measured with 13 yes/no questions derived from adaptations of the Reproductive Coercion Scale questions (Miller et al., 2011; Miller, Decker, McCauley, Tancredi, et al., 2010). Questions were adapted to isolate pregnancy-promoting intent from coercive behaviors, based on data from the qualitative phase of the study and recent literature (Katz et al., 2017; Katz & Sutherland, 2017). Four questions assessed pregnancy coercion, five questions assessed BCS, and four questions assessed controlling the outcome of a pregnancy. BCS items included: “Has someone you were dating or going out with ever:” (a) taken off the condom while you were having sex?; (b) put holes in a condom or broken a condom on purpose while you were having sex?; (c) taken your birth control (like pills) away from you or kept you from going to the clinic to get birth control? (d) made you have sex without a condom so you would get pregnant?” or (e) told you not to use birth control. The behavior “Telling a woman not to use birth control” is typically considered under the domain of pregnancy coercion (McCauley et al., 2017) but because the intent behind this behavior may also be unclear, it may be reasonable to consider this a component of BCS. Positive responses to BCS questions or telling a woman not to use birth control were followed with the question, “Is this person trying to get you pregnant?” RC was defined as a positive response to any item, including BCS and telling a woman not to use birth control, regardless of response to the question about partner's intent.
Data Analysis
Qualitative Analysis
Transcripts were entered into Dedoose®, a web-based qualitative analysis program (SocioCultural Research Consultants, 2016). Using qualitative descriptive methodology (Sandelowski, 2000, 2010; Vaismoradi et al., 2013), the first author read each transcript multiple times to verify accuracy and gain an understanding of the overall responses to the interview, ensuring confirmability. A codebook of a priori codes was developed within the constructs of RC behaviors and harm reduction strategies that were aligned with the literature and was expanded during the analysis with emerging codes. After preliminary readings, detailed reading of each transcript and line-by-line coding was completed by two researchers, who independently coded all transcripts, applying the codebook. We compared our coding to resolve any discrepancies and discussed emerging themes throughout the coding process. Through pattern coding, codes were examined and grouped into categories, themes and theoretical constructs (Braun & Clarke, 2006; Sandelowski, 2010; Vaismoradi et al., 2013).
Quantitative Analysis
Descriptive statistics (mean values, standard deviations, and frequencies) were used to describe the sample, the prevalence of types of BCS behaviors, and the prevalence of perceived pregnancy motivation associated with BCS behaviors. Independent sample t-tests and chi-square were used to examine differences between those who had and had not experienced BCS (regardless of perceived intention) on risk factors and covariates.
Mixed-Methods Analysis
Qualitative and quantitative data were triangulated in a table to visualize areas of congruence and dissonance.
Ethical Review and Informed Consent
The study was approved by the Johns Hopkins Medicine Institutional Review Board (IRB00129418). Qualitative interview participants provided oral informed consent covering the topic of the study, confidentiality, and the voluntary nature of the study including that they can decline to answer any question, prior to completing a survey of their demographic information and participating in the interview. They were offered a $20 gift card to thank them for their time. Quantitative survey participants reviewed tablet-based survey/questionnaire informed consent covering the nature of the questions, confidentiality, and the voluntary nature of the study including that they can decline to answer any question. To thank them for their time, they had the opportunity to enter a raffle for one of twenty $50 retail gift cards. Research Assistants received standardized human subjects research ethics training as well as IPV advocacy training including safety assessment, technology safety, IPV resource referrals, and suicidality protocols.
Results
Key demographic characteristics and RC prevalence for the two samples are presented in Table 1. Participants who experienced any BCS behavior were younger than those who did not (27.33 vs. 30.99, p < .001) and more likely to be born in the United States (20.0% vs. 7.3%, p < .001).
Demographic Characteristics of Latina Women Seeking Care at a Federally Qualified Health Center in the Washington, DC Area; Qualitative and Quantitative Samples and Prevalence of Birth Control Sabotage Experience in Each.
*By t-test or chi-square.
Qualitative Results
Qualitative interviews with 13 women revealed that women perceived the intent behind BCS behaviors in a variety of ways. These perceptions differed based on the type of BCS.
Motivations Ascribed to Condom Refusal/Removal
When condom refusal or removal was described, it was never attributed to pregnancy-promotion motivation. Rather, one participant's (ID# 54; age 30) partner refused to use condoms due to dislike for the method: He, in reality he never liked using them [condoms], but I would tell him he had to sometimes when I didn’t get the injection. But he told me no, that he didn’t like them.
And another participant (ID# 8; age 30) perceived that her partner surreptitiously removed the condom during sex simply for his own pleasure: I’ve had like someone say they’ve put on a condom and then try to enter you and… they’re like oh it fell off… I’ve had that happen before… For the male I think it's just they’re obviously just thinking about their own experience or their own pleasure. I don’t think they’re thinking about the consequences at all, I think they're just thinking about the act of it.
Another participant (ID# 61; age 29) believed her partner's condom refusal was due to “machismo”, a Spanish word used to describe a social behavioral pattern, sometimes considered a cultural norm, of masculinity and overt displays of power and control. I wanted him to use condoms, he didn’t want to, and I had just given birth. So I told him, then we’re not going to have relations without, I had just had a baby… you can’t have me like that because you’re going to get me pregnant. But he was always machista like that…
Instances of sexual assault were also described by participants in their stories of partner refusal to use condoms. While these instances did result in pregnancy, the participants’ descriptions of the events centered around the violent act of assault, and they did not perceive intention to cause a pregnancy. An example was one participant (ID# 61; age 29): So they told me to use condoms as birth control. He was really mad because he didn’t want to put it on. And if he didn’t want to, we would fight all night until he manages to have relations with me. But not with my consent.
Condom refusal and removal puts women at clear risk for sexually transmitted infection as well as unintended pregnancy, regardless of whether the intention was to cause either of these outcomes.
Pregnancy-Promoting Motivations Ascribed to Other BCS Behaviors
Other tactics of BCS such as tampering with, telling a woman not to use, restricting access to, hiding or throwing away birth control, were attributed to a range of motivations which included pregnancy promotion. One participant's (ID# 61; age 29) partner, in addition to refusing condoms, also told her not to use birth control, and this she attributed to pregnancy-promoting intention: He would say mami, the kids are getting big, let's have a baby… And he was always… “don’t take that, mami. Don’t take that”, he would say. I can’t stop taking some form of birth control.
Another participant (ID# 35; age 24) described her partner hiding oral contraceptives which she clearly perceived as motivated by a desire for pregnancy: He hid them from me. He hid the pills, threw them away. Since he had a daughter, he wanted a son, and he wanted me to get pregnant… I didn’t want to have another baby… but he said that he wanted me to have another, and then later he told me he didn’t want to pay for my injection, for birth control. So, then I asked him why, he told me that it's because he wants to have another baby.
One participant's (ID# 54; age 30) partner told her not to use birth control out of a desire to achieve pregnancy as well: Sometimes he would tell me not to go to the appointment, when it was time to get the injection he would say, “what are you going for?” That “this is hurting me a lot.” And that “it's better to have the children you’re going to have.” And, well… I didn’t want to, to tell the truth.
Another participant's (ID# 51; age 29) partner also told her not to use birth control and locked her in the house to prevent her from accessing it, with the explicit intention of causing a pregnancy: We would fight a lot about me getting the shot for planning. He wanted me to be one of those, filled with babies. No? Like that I just had more and more babies… I didn’t want a baby… what I wanted, like I’ve said, was to finish my high school. I was in ninth grade… When we got married, well, things were different, and then he didn’t want me to go to school, didn’t want me to use family planning, wanted me completely to be a housewife… He would lock me in the house, he wouldn’t let me out to get the shot. Many times my mother had to come for me, to get me because he was, he had me locked up… It was so much, so much violence, screaming, he got terrible when I went to get the shot. Every three months.
One participant's (ID# 35; age 24) partner also prevented her from accessing oral contraceptives due to a desire to test his belief that he was infertile: I would tell him that no, I have to take my pills, and he would say “No, what for?” …he thought that he couldn’t have children, according to him, because he did it with other women and the other women didn’t get pregnant… he told me he wanted to know if he could have children or not… So he hid the pills… he threw them away.
And another participant (ID# 13; age 26) also interpreted her partner's sabotage as intention to cause pregnancy, though this was interpreted positively: Whenever he knew it was time for me to go to my gyno and get a refill on the birth controls he found a way for me to miss my appointment. And then he knew after 3 missed appointments they will no longer see me, and I would have to look for a new doctor. So I guess just to ensure that I wouldn’t get the pills anymore he kind of did that, but I don’t think it was just in a malicious way I just think that he really wanted to grow with me and become a family with me.
These women's partners were very clear that the intention behind their behaviors was for the woman to get pregnant.
Nonpregnancy-Promoting Motivations Ascribed to Other BCS Behaviors
BCS behaviors were also attributed to other intentions. For example, one participant's (ID# 61; age 29) partner told her not to use birth control reportedly out of concern for her health and mistrust of birth control: I said, “They put in an implant, it's something new, I want to try it. And it doesn’t hurt you.” He says, “No, it's going to hurt you, you’re going to end up sterile.”
Another participant (ID# 32; age 35) described her partner taking the money her mother had provided for her to get birth control, to spend on alcohol and drugs: She said to me if you’re sure you don’t want to have more children, then I’ll give you the money. And she gave me money and then what he did, he took the money for his vices.
Partner motivations in these cases were purported to be reasons other than pregnancy promotion, but in each case women experienced infringements on their ability to prevent pregnancy.
Quantitative Results
Almost one in seven (13.7%) of the 482 women in the study experienced any of the five BCS behaviors in the past year (Table 2). Of these 66 participants, 34.8% perceived that their partner was trying to get them pregnant and the remainder (65.2%) did not perceive any pregnancy-promoting intent. The behavior that was most often perceived as pregnancy-promoting was being told not to use birth control (45.7% of the 35 women who experienced this behavior). The least likely behavior to be perceived as pregnancy-promoting was being forced to have sex without a condom (29.4%).
Types of RC Experienced a in the Past Year by Domain, With and Without Perceived Intent to Impregnate, Among the Quantitative Sample (n = 81).
Not mutually exclusive, that is, women can experience more than one type of RC.
Row percentages.
Includes RC domains of pregnancy coercion, controlling outcome of pregnancy, and birth control sabotage.
Overall, 16.8% of participants (n = 81) experienced past-year RC in any of the three domains of RC behaviors (BCS, pregnancy pressure, controlling the outcome of a pregnancy) when using questions that are not specific to intent. When limited to RC experiences for which participants perceived pregnancy intent, the past-year RC prevalence was 9.5%.
Mixed-Methods Results: A Comparison of Qualitative and Quantitative Findings
Participants did not report their partners poking holes in or breaking condoms on purpose in qualitative interviews. However, all other BCS behaviors were reported at least once (Table 3). Condom removal during sexual activity and condomless forced sex were never perceived as motivated by pregnancy promotion in the qualitative data. Among women who reported these behaviors on the survey, only 30% perceived pregnancy-promoting intent. Preventing partner's birth control access was mentioned seven times in interviews, and in five of these, it was perceived to be motivated by intentions for pregnancy. According to survey data, 60% of participants who experienced this behavior perceived pregnancy-promoting intent. Women described being told not to use birth control five times in interviews, and in two of these it was perceived to be motivated by pregnancy-promoting intent; in survey data 45.7% perceived pregnancy-promoting intent.
Convergence Matrix of Qualitative and Quantitative Data.
Discussion
Latina women interviewed in this study described a range of perceived motivations behind their partners’ BCS behaviors, including pregnancy promotion but also including other motivations, and these perceptions varied by type of BCS behavior. Survey data were consistent with interviews; over one-third of participants who experienced BCS perceived pregnancy-promoting intent, while almost two-thirds did not. Interference with female-controlled methods (being told not to use birth control) was most likely to be perceived as pregnancy-promoting, while interference with male-controlled methods (condom refusal and removal) was least likely. This study contributes to the growing literature on RC by underscoring both the prevalence of BCS behaviors as well as situating these behaviors in a broader context of power and control beyond intent to promote pregnancy.
BCS has typically been included under the umbrella of pregnancy promotion in the study of RC. This mixed-methods study shows there is a combination of pregnancy promotion and other intentions (pleasure, dislike, machismo, violence, mistrust, and addiction) that appear to motivate BCS behaviors. Survey respondents who answered that their partners had prevented them from accessing birth control but were not trying to get them pregnant may have been simply unaware of their partner's intention, or it is possible that another rationale was explicitly stated, such as mistrust of the method. However, regardless of intent, in all cases, the risk for harm to sexual and reproductive health is the same and threats to reproductive autonomy and risk for unintended pregnancy are present.
What motivation a woman perceives for a partner's controlling behavior has implications for clinical response, for agency over risk of unintended pregnancy, and for how she will respond to research measures. Whether or not women perceive pregnancy-promoting motivation behind interference with male-controlled methods, the use of supplementary female-controlled methods can have significant benefit for women wishing to make autonomous decisions about when to get pregnant. Health disparities and health inequities experienced by Latina women have a complex and multifactorial etiology encompassing social and cultural injustice, discrimination, and oppression. Societal norms of patriarchy frame the “machismo” behaviors perceived by women, as well as the need to satisfy male pleasure. Historical abuses such as forced sterilization of immigrant women and the testing of oral contraceptive pills on women in Puerto Rico without their consent, justify mistrust of contraception and the healthcare system (Reilly, 2015; Ross & Solinger, 2017). Reproductive coercion is one factor that may impact disparate rates of unintended pregnancy and STIs among Latina women relative to white women, and this research contributes to an understanding of that construct and how to optimize the measurement of these experiences in this population.
Implications for Research
This study underscores that women perceive a variety of motivations for BCS behaviors, and thus BCS prevalence is lower when limited to pregnancy-promoting BCS specifically. Some, but not all, women who experience BCS perceive pregnancy-promoting intent. Limiting research measures to those who perceive pregnancy-promoting intent will underestimate the prevalence of the behavior by missing those who may have perceived a different intent. Overall, a more sensitive measure that does not rely on perceptions of pregnancy intention is a more useful instrument in the measurement of RC. And to capture the full spectrum of RC, a measure that includes interference with both male- and female-controlled methods in addition to pregnancy coercion and controlling the outcome of a pregnancy, is needed and is consistent with prior psychometric research on RC measures (McCauley et al., 2017). There is much current debate about the definition of RC and if a behavior can be considered to meet that definition if it is not explicitly intended to impact pregnancy (Tarzia & Hegarty, 2021). The line between RC and sexual violence continues to be defined and more research is needed on whether there are different associated health outcomes.
Implications for Clinical Practice
All the BCS behaviors have potential for harm to women's health (risk for unintended pregnancy and for sexually transmitted infection as well as potential physical and psychological trauma) and reproductive autonomy. Notably, women also made explicit connections between condom manipulation and refusal and experiences of sexual violence. Therefore, limiting clinical assessment to BCS with pregnancy-promoting intent may miss the opportunity to discuss these behaviors and ways to increase safety and reduce harm. Providers should be attuned to conducting deeper assessment of contraceptive nonuse to include both interference with female-controlled contraception as well as condom manipulation (Chamberlain & Levenson, 2012). A follow-up question about whether her partner may be trying to get her pregnant when she does not want to be can direct the provider to referral for relevant services and supports. Offering harm reduction strategies such as contraception that is more difficult to interfere with (women-controlled methods such as an injectable or intrauterine device) should not differ based on whether the patient perceives pregnancy-promoting intent on the part of her partner or not.
Removing intent from questions about BCS behaviors is beneficial for both research and clinical practice. Our recommendation for both research and clinical practice is to split BCS questions into two parts, first asking about the behavior and then asking “Is this person trying to get you pregnant” for any positive responses to more fully evaluate the impact of sexual and reproductive coercion on women's health.
Strengths and Limitations
Strengths of this study include the use of mixed methods and the robust community-based sample of Latina women. Recruitment was conducted at a single site; the resulting sample of Latina women accessing health services at a clinic serving predominantly low-income clients may not generalize to other settings. This study did not examine whether the coercive partner was also the father of the participant's children, which might further elucidate understanding of his motivations. Future research should examine men's experiences and behaviors and their stated intentions behind BCS behaviors. With consideration for safety and ethics, dyadic research with couples may also reveal additional nuances related to intentions and perceptions regarding sexual relationships and potential for pregnancy, and could elucidate the additive influence of other types of coercive behaviors. Questions remain about the construct and strict definition of RC (Tarzia & Hegarty, 2021); toward that end, separating intent from behavior in the measurement of BCS is likely to result in a more useful measure of RC. Psychometric testing across multiple populations is needed to further validate the wording proposed in this study.
Conclusion
Measures for research and clinical assessments for RC continue to be refined to enable the identification of potential harm and opportunity for intervention. How items are worded has critical implications for both research and clinical practice. Continuing to ask about partner pregnancy-promoting intent is essential to capture threats to reproductive autonomy that are distinct from other threats such as to the physical body, sexual health and decision-making, or to sexual consent, and to inform appropriate interventions. However, asking survivors to identify motivations behind specific behaviors may be unrealistic and irrelevant, and risks missing exposure to harmful behaviors when the survivor is unaware of the motivation or unwilling to consider what the motivation might be.
For comparison, in order to avoid minimization of or providing excuses for abusive behavior, standard screening for other IPV behaviors does not ask about intent but simply asks if the behavior has happened. In clinical practice, asking about the behavior is sometimes part of the intervention itself. If a woman can identify that a behavior is happening, the next step can be to consider the motivation and impact and whether it is harmful. Recognizing that a partner's intent is not always known, we recommend that RC clinical assessments and RC research measures ask BCS questions in two parts, asking first about behaviors and then about perceived intent.
It will be helpful to further explore the construct of RC by studying men and their stated motivations, but we argue that given the dearth of evidence on motivations and the difficulty in ascertaining another person's motivation, the most useful construct is one that focuses on behaviors with the potential to impact reproductive autonomy, rather than behaviors that are intended to impact reproductive autonomy. It is the RC behavior, not the intention, that has well-documented impact on health outcomes, and it is on behaviors that researchers and clinicians have the greatest potential to intervene.
Footnotes
Author's Note
This study would not have been possible without the contributions of all of the women who participated, and without the Mary's Center staff, providers and management who supported this study and helped recruit participants.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
Funding support for K. Grace: American College of Nurse Midwives Fellowship for Graduate Education, American Nurses Foundation (Anne Zimmerman, RN, FAAN Nursing Research Grant and Dorothy A. Cornelius Nursing Research Grant), the Melissa Institute for Violence Prevention and Treatment (Belfer-Aptman Scholars Award), the National League for Nursing (NLN Foundation Scholarship Award), the Council for the Advancement of Nursing Science/Southern Nursing Research Society Nursing Science Advancement (NSA) Dissertation Grant Award, and the Health Resources and Services Administration (HRSA) of the U.S. Department of Health and Human Services (HHS) (Number: T76MC00003 Title: Training Program in Maternal and Child Health).
Funding support for E. Miller: NICHD K24HD075862.
Funding support for K. Alexander and C. Holliday: NIH/NICHD/OWH K12HD085845.
Funding support for C. Holliday: NIMHD 1L60MD012089.
