Where Sex Ends and Emotions Begin: Love and HIV Risk among Female Sex Workers and their Intimate, Non-Commercial Partners along the Mexico-U.S. Border

Abstract

This study explores the affective dimensions of female sex workers’ relationships with their intimate, non-commercial partners and assesses how emotions shape each partner’s sexual and drug-related risk within their relationship. We draw on qualitative data from a study of HIV, sexually transmitted infections, and high risk behaviours among female sex workers and their non-commercial partners in Tijuana and Ciudad Juárez, Mexico, to illustrate that these couples share relationships based on love, trust, respect, and emotional and material support. These relationships ranged in emotional intensity, which shaped partners’ decisions not to use condoms with each other. Drugs were important in most couples’ relationships. Among injectors, syringe sharing was common and represented both a sign of care and a pragmatic reaction to conditions of material scarcity. Our findings suggest that couple-based HIV interventions to address dual sexual and drug-related risks should be tailored to the emotional dynamics of sex workers’ intimate relationships.

…in a relationship like mine [with a sex worker] you need to define where sex ends and emotions begin. I think it is the same for my partner because it is more of an emotional need that I need to give her because of her work and in that sense I don’t think there is someone out there that loves her like I do. I also don’t think that there is another woman that will love me like she does.

(Luis, 2-year-old heroin injector on his six year relationship with Paz, a 40-year-old sex worker)

Introduction

Intimate relationships are important sites of sexual and drug-related HIV risk, but female sex workers’ intimate, non-commercial relationships are rarely addressed in prevention strategies. Research suggests that emotional approaches to understanding HIV risk behaviours may hold promise for designing couple-based interventions (Bluthenhal and Fehringer 2011). This study explores how love and other emotions shape HIV risk among female sex workers and their intimate partners in the Mexico-U.S.A. border region.

Researchers suggest that disadvantaged women often engage in unprotected sex with steady partners for emotional and material security, as not using condoms is symbolic of trust and fidelity (Rhodes and Cusick 2000; Sobo 1995). Studies show that sex workers are less likely to use condoms with intimate partners than with clients (Stoebenau et al. 2009; Sanders 2002) and often reserve special acts (e.g., kissing) for their non-commercial partners (Allen et al. 2003). Drug use complicates relationships by introducing physical, social, and legal risks, but such concerns are often secondary to maintaining the relationship (Rhodes and Quirk 1998; Simmons and Singer 2006). Injection and non-injection drug use is associated with unprotected sex (Strathdee and Sherman 2003). Women may also engage in sex work to care for and support their partner’s drug use (Lam 2008). Syringe sharing is common among injecting couples (Gyarmathy et al. 2010; Cleland et al. 2007), and researchers have speculated that emotional meanings (e.g., love and trust) may be important in shaping couples’ injection practices (MacRae and Aalto 2000; Barnard 1993).

Since the era of U.S. Prohibition, the growth of a leisure-based economy in the largest Mexican border cities has reinforced socio-economic inequalities and contributed to linked epidemics of sex work, drug abuse, and HIV (Ramos et al. 2009). Tijuana, south of San Diego, California, has a population of 1.6 million residents. Sex is sold in clubs, bars, and on the streets of the Red Light District (Bucardo et al. 2004). While sex workers are required to obtain permits, studies suggest that only about half do (Sirotin et al. 2010). Ciudad Juárez, in the state of Chihuahua, has a population of 1.3 million residents and forms a port of entry with El Paso, Texas. The red light districts of Juárez have undergone gentrification and permits are not required (Wright 2004).

Studies in Tijuana and Juárez have documented a rise in HIV prevalence among sex workers from less than one percent in the late 1990s to six percent in 2006 (Strathdee and Magis-Rodriguez 2008). About half of sex workers in these cities have intimate partners with whom they were less likely to report condom use than with clients (Ulibarri et al. 2012). Sex workers who injected drugs were also more likely to have intimate partners who injected (Strathdee et al. 2008). This work suggests that male partners may significantly contribute to sex workers’ risk.

Within this context, we examined how love and other emotions shaped sexual and drug-related HIV risk among sex workers and their intimate partners. Our analyses are guided by critical phenomenology (Desjarlais 1997), which combines concern with lived experience and subjectivity with political economy perspectives that highlight the historical and structural production of risk. As a holistic construct, “love” is a useful lens to connect emotional, social, and structural dimensions of experience (Padilla et al. 2007). When viewed within a political economic context, analyses of love illuminate “the subjectivity, emotion, and agency of people who struggle within these constraints to establish emotional connections and intimacy with others” (Padilla et al. 2007: xii).

Our critical phenomenology framework places couples’ subjective emotional experiences at the center of our analysis; emanating from this understanding, we then trace the ways affect shapes their health behaviours while considering the broader constraints on couples’ options. For sex workers along the border who endure poverty, marginality, addiction, and increased HIV risk, we suggest that maintaining emotional bonds with intimate partners may be of paramount importance. As illustrated by Luis’s quote in the opening of this paper, our analysis gives voice to the emotional lived experiences of sex worker couples. Our objective is to draw on couples’ subjective experiences as a means to inform couple-based interventions.

Methods

This study draws on qualitative data collected as part of Proyecto Parejas (Couples Project), a mixed methods study of HIV/STI risk among female sex workers and their noncommercial male partners in Tijuana and Juárez, Mexico. The aims of the study were to document HIV/STI prevalence and incidence, examine how relationship factors shape disease risk, and assess the feasibility of developing a couple-based intervention. Our study protocol is described elsewhere (Syvertsen et al. 2012). Briefly, we recruited through the female partner. Eligible women were at least 18 years old, reported lifetime drug use, and were active sex workers in a non-commercial relationship with a male partner for at least six months. Women who passed the primary screener brought their partner to the study offices for a couple-based screening to verify their relationship.

Between February 2010 and September 2011, we recruited 214 couples. Each partner provided written consent for quantitative surveys and HIV/STI testing at baseline and follow-up every six months for 24 months. A subset of couples also provided written consent to participate in qualitative interviews. All protocolswere approved by institutional review boards of the University of California, San Diego, the Hospital General and El Colegio de la Frontera Norte in Tijuana, and the Universidad Autónoma de Ciudad Juárez.

At enrollment, a sub-set of couples at each site participated in qualitative interviews exploring the relationship context of HIV risk. We used purposive sampling (Johnson 1990)to obtain a sample with maximum variation in characteristics relevant to relationship quality and HIV risk: partner ages, length of the relationship, male employment, and drug use. Between February 2010 and March 2011, we interviewed 18 couples in Tijuana (18 joint and 36 individual interviews) and 23 couples in Ciudad Juárez (23 joint and 45 individual interviews). In the 122 total interviews, we repeatedly heard similar information about our topics of interest and determined that we had reached theoretical saturation, or empirical confidence that the sample size was sufficient to adequately explore the themes of interest (Guest, Bunce, and Johnson 2006).

Interviews explored the context of the relationships (e.g. how the couple met), finances, sexual behaviours, sex work, drug use, and drug treatment. Questions relevant to this analysis included, “Do you consider yourself to be in love with your partner?” We used the same guide for the joint and individual interviews. Interviews were conducted in Spanish or English by trained interviewers in private rooms at the project offices, and lasted from 30–90 minutes. Interviews were audio recorded and transcribed verbatim, and interviewers wrote notes on key findings and behavioural observations.

Data analysis involved a multi-step process. First, the research team read through identical interviews and independently generated codes based on emergent themes (Ryan and Bernard 2003). The team discussed and refined these codes, constructed a codebook, and independently coded a second set of transcripts, including an identical round of transcripts to check consistency in code application. Analysts met regularly to discuss coding progress and resolve issues (MacQueen et al. 1998).

The current analysis is based on a phenomenological approach, which focuses on meaning and subjective lived experience (Creswell 2007). Typically, analyses require the researcher to set aside their personal viewpoint to gain a deeper understanding of the data (Moustakas 1994). The first level of reading for this analysis facilitated an understanding of couples’ perspectives in their own words. Subsequent readings required a reflexive process to construct linkages between emotions and behaviours. As public health researchers concerned with HIV risk, we acknowledge that couples often view and prioritise their risks in ways that contradict official public health discourse. Our analyses attempt to critically reconcile issues of couples’ subjectivity with real physical risk for disease.

By repeatedly reviewing interview transcripts to understand how participants defined the meaning of their relationships (e.g., love, trust, support), we created a typology of emotional closeness among couples. Emotional data from individual and joint interviews were aggregated to place each couple into one of three categories of emotional closeness that emerged from the texts. We then compared these emotional categories with sexual behaviour within the relationship (e.g. condom use) and drug use practices (e.g. syringe sharing) to synthesize and elucidate patterns of HIV risk at the couple level. We acknowledge that a typology imperfectly embodies the complexity of these relationships and does not represent categories with which the research participants would necessarily identify. However, we use the typology as a heuristic device to better understand and draw linkages between emotional intimacy and risk.

We assigned pseudonyms to all participants to protect confidentiality. Exemplary quotes were drawn from individual interviews unless otherwise indicated. The original Spanish is provided when necessary to preserve nuanced meanings.

Results

On average, among 41 couples (n=82 individuals), partners were 36 years old (range: 20 – 61) and had been together for nearly five years (range: 7 months – 25 years). Partners had an average of seven years of schooling and the majority earned less than 2500 pesos per month (less than U.S. $200). Although two thirds of all couples had children, only about one quarter had young children living with them. The majority actively used drugs: heroin was common to both sites, while methamphetamine was popular in Tijuana and powdered cocaine and crack were used in Juárez. More than half of all participants injected drugs. Sex work in this context was driven by financial and drug-related need and lack of economic opportunities. Men were typically under-employed in informal jobs (e.g. street vending, washing cars) or resorted to crime.

The results of our study first outline how partners construct and experience the meaning and significance of love within their intimate relationships. We then describe sexual behaviour, condom use, and drug use practices within the context of a typology built on these couples’ emotional experiences. Our findings suggest that the subjective meanings of couples’ relationships have important implications for couple-based interventions. These key findings are described below.

Significance and meaning of love

Couples were involved in relationships that spanned a range of love and emotional intensity. Juan Carlos, a 53-year-old heroin injector from Juárez who had been with his partner for one year, suggested that “there are many forms of love (amor)” and partners can sense when their love is “real.” Partners utilised the richness of the Spanish language to describe their relationships, and at times offered complex explanations of their personal feelings. In Spanish, love is expressed in multiple words that imply different strengths of emotion: amar and enamorarse imply a strong, passionate love, while querer typically signifies a warm, friendly love.

In addition to “love,” couples discussed other forms of emotional intimacy, or close feelings for each other. Partners commonly invoked the importance of trust (confianza), respect (respeto), understanding (comprension), support (apoyo), gratitude (agradecimiento), affection (cariño), friendship (companerismo), protection (protection), and happiness (felicidad). Trust, respect, and understanding were particularly important, as was gratitude for the relationship through good and bad times (en lasbuenas y lasmalas), particularly during times of illness. Support and care were important for couples with limited access to healthcare or who felt discriminated against because of their drug-involved identity.

Relationships varied in intensity and expression of these feelings. Based on how the couples talked about their relationships, the words they used to define their feelings, and the sense of companionship and commitment they shared, we developed three categories to describe the progressively intense emotional closeness of these relationships: 1) los involucrados, or relationships based primarily on mutual forms of support rather than affect; 2) los queridos, or “typical” couples who loved each other; and 3) los enamorados, or couples with intense emotional attachment who were “in love.”

One quarter of couples said their relationships were not based on love or strong emotions. Among these involucrado couples, one or both partners said they were unsure if they loved the other, their feelings had faded, or they remained together for material forms of support (e.g., children, housing). Love was infrequently discussed in these relationships, but rather these couples felt “comfortable” or “accustomed” to their relationships. Nevertheless, partners provided vital material and emotional support, and they considered themselves to be a couple because of how they prioritized their time and resources for each other.

Several involucrado couples admitted that their relationships were based on drugs. Katrina, a 28-year-old heroin injector in Tijuana, met her partner in a shooting gallery and “can’t imagine it working without drugs.” Alda, a 32-year-old heroin injector in Juárez, also held a fatalistic attitude about her relationship. While their relationship has lasted over two years, she wondered if it would endure:

This is not forever, any relationship ends and well, that’s it. And I think that’s fine, I mean, that time comes. I understand that all relationships end, especially one like ours because we’re addicts; maybe we are together because of drugs, maybe it’s not so much because the affection [cariño] or anything, maybe it’s because of drugs …

Los queridos comprised the majority of our sample. These couples said they were not “in love” (enamorado) and did not feel strong love (amor) for each other, but they frequently expressed a warm and caring love for each other (querer).

In love (enamorado), I can’t say that, but I do love him a lot (lo quiero mucho). (Martina, age 34, 1.7 years together, Tijuana)

I mean because a feeling like love (amor), well I don’t feel it… I do love her (lo quiero), I love her very much and also respect her, but love (amor), no, it’s not.(Esteban, age 22, 4.8 years together, Tijuana)

Los querido srelationships were often long-term and committed, and endured based on a constellation of shared experiences and affect. Partners reported that trust, respect, and understanding were particularly important, as partners frequently referenced the trauma, poverty, and discrimination that they had experienced throughout their lives, and felt that their relationship helped them through adversity. During their couple interview, Rafael, a 42-year-old recovering drug injector, admired his partner’s perseverance:

People judge people like her [sex workers] like trash, you know? Like someone who has no worth, we don’t see if they have feelings or if it’s someone who is worthy. Because she is making an effort, there are men as well as women that don’t fight to make it, to survive, and she has done whatever she has to do as a way of [survival]. I think that it is difficult for a woman to make a decision like that [sex work] and it is difficult to find someone who will support them, someone who values them, you know? (1.7 years together, Tijuana)

Finally, almost one quarter of couples said they were “in love” (enamorado). These couples expressed strong feelings for each other and their relationships were central in their lives. Los enamorados relationships also involved feelings of mutual care, respect, trust, understanding, support, and long-term commitment. Partners were secure in their feelings and unequivocal regarding the importance of their partner in their lives.

Marta, a 45-year-old heroin injector, declared: “We are a couple; we are friends, lovers, everything, complete” (1.2 years together, Tijuana). Ricardo, a 37-year-old non-drug user in Tijuana, said that even though a previous partner was educated and his family approved of her, he never loved her like Annie, his current partner of three and a half years, a 24-year-old sex worker who smokes methamphetamine. According to Ricardo, in his past, “there wasn’t a place for feelings and now even if my partner is a prostitute there is more love, more feeling (hay mas amor, hay mas sentimiento).”

Often, these relationships were viewed as transformative forces in couples’ lives. Several partners spoke about changing their behaviours (e.g., attempting to reduce their drug use). Others said their relationships rescued them from loneliness and depression:

When I decided to be with him it was because I was feeling very lonely and stressed and I felt I needed someone by my side, but slowly I came to realize that I love him [lo quiero y lo amo] as a partner, husband, father of my son, and as a person. He has respected me as a person and I respect him.(Mona, age 28, 2 years together, Juárez)

I think that if I hadn’t met her maybe I would even be dead now… for me she is the most beautiful thing I’ve had in my life…she is my light, she is my star, she is the path to my happiness.(Francisco, age 42, 6.5 years together, Juárez)

While drug use was a central part of many of these relationships, emotionally close couples claimed that their love for each other transcended their addiction. Although they both injected heroin and cocaine, Guillermo said his relationship did not “just revolve around a syringe” and he did not view his partner as an “addict”, but rather as a woman whom he loved:

…my relationship with her is not based on drugs, it is based on feelings that I have for her, I am in love with her (estoy enamorado de ella), and I love her (la quiero) the same way with drugs or without drugs…I need her more than I need drugs. (Guillermo, age 44, 6 years together, Juárez)

Sexual behaviour

All couples had sexual relationships but the frequency of sex intersected with social factors such as living with children and drug use. Frequently, both partners acknowledged that women were “too tired” to have sex after work. One male partner made the analogy of being a cook at a restaurant who does not want to come home and make dinner after working all day.

For several women in involucrado relationships, their HIV risk began with incongruent sex drives that rendered sex a source of conflict. In one extreme case, Imelda, a 34-year-old methamphetamine user, said her partner’s meth use turned him into an aggressive and “sexually sick” person. She reported the single case of partner rape in our sample and compared him to a client in saying that “he feels like money.” Several other women said they sometimes had to “give in” and felt coerced into having sex with partners.

In contrast, querido and enamorado couples clearly separated the physical act of sex within their relationship from commercial sex. Women maintained that sex with clients was not for personal satisfaction and there was no emotion involved. They distinguished the “feeling” in sex with their intimate partner from “pretending to be blank” when servicing clients. Male partners also discussed the pleasure in their sex as distinct from sex work. Juan Carlos insisted that he fulfilled a different sexual role:

I satisfy her, let’s call it that. Simply to make her feel pleasure because many times people think that because of being a sex worker the spouse or the sentimental partner doesn’t satisfy her anymore, you understand me? But it’s very different, at work compared to the relationship she has with her partner, you don’t have to fake it with your partner…when I’m with her, it’s only me and her.

Condom use patterns

Only one couple reported consistently using condoms due to her recent HIV diagnosis (he tested negative). Five couples reported ever using condoms, typically when one partner had an STI, and three couples had tried condoms early in their relationship but quickly ceased using them. One couple described having a fight after which the woman threatened her male partner with condom use as a demonstration of her anger. As condoms were clearly viewed as inappropriate within intimate relationship contexts, we examined partners’ perspectives on condoms and found that reasons varied according to the emotional content of the relationships.

Involucrado couples often did not use condoms simply because they “do not like” them. Non-condom use was viewed as normative in intimate relationships. Several others justified not using condoms because they “haven’t been sick,” or they did not perceive themselves to be at high risk through unprotected sex with intimate partners. For some, not liking condoms and not being sick overlapped to contribute to fatalism about the futility of using condoms in their relationships.

Querido couples often gave multiple reasons for non-condom use. In addition to the reasons above, the subjective meaning of sex began to drive their risk behaviours. Most querido couples and all enamorado couples justified condom non-use for reasons relating to subjective meaning: partners shared the sentiments that “we are a couple,” “we love each other,” and the male partner “is not a client.” Some couples established formal rules that condoms were not to be used within their relationship and should only (and always) be used with clients. Couples trusted each other: women were supposed to “protect themselves” with clients and men were supposed to be monogamous. To Cindy, a 29-year-old heroin injector, using a condom with her partner would change the experience, physically and emotionally:

Well, to me, it doesn’t feel the same, it really doesn’t, one. And two, I mean, you feel closer to the person without a condom, and you feel more like you’re trusting each other; you really, truly, trust him by not using a condom with him. By mentioning a condom, or wanting to use a condom with him, you’re kind of telling him that you don’t trust him much. (1.5 years together, Tijuana)

In contrast to involucrado couples, among couples for which subjective meaning overlapped with “not being sick,” not using condoms was a strategy for establishing trust rather than a form of fatalism. Suzy, a 50-year-old cocaine user, has never used condoms with her partner, but she indicated in their couple interview that she would protect him if she were to get sick:

I mean since he’s my partner, obviously I trust him completely, right? Because I know that he’s not going to give me a disease, because he only gets involved with me and with the people that I get involved with I use condoms… he trusts me and the day when I tell him that I’m ill or whatever, I won’t even[have sex] with him. (6.5 years together, Juárez)

Drug-related risks

In addition to sexual risk, almost all individuals were active drug users, which introduced additional HIV risk into their relationships. More than half of all couples shared the same drug use patterns (e.g., both injected), while the remainder had different preferences (e.g., one partner injected and the other smoked), and a minority were not currently using. Drug addiction, in part, drove women’s sex work.

Couples in which both partners injected were common in both study sites and represented all levels of emotional closeness. All injecting couples pooled resources to purchase and use drugs together several times per day, and nearly all couples shared syringes within their relationship. Although partners preferred to use their own syringes, sharing occurred in times of scarcity, when one partner’s syringe became clogged, or partners confused their syringes in pooled storage (e.g., their syringes were kept together). Syringes were frequently stored for multiple uses until the tip became too dull to inject, and such reuse was viewed as a prudent conservation of limited resources.

Nevertheless, emotions and structural issues often converged to shape sharing practices among injecting couples. Among involucrado injecting couples, affect was less important in their drug use and partners shared syringes within and outside of the relationship. Couples whose relationships were based on drugs often injected in shooting galleries or other public places, which provided opportunities to share drugs and syringes outside of the relationship. Scarce resources, policing practices, and malilla (drug withdrawal) often prompted sharing in public.

In contrast, close querido couples and all enamorado couples relied on their relationships in their drug use. These couples injected together at home and confined their syringe sharing within their relationships. They trusted each other and viewed sharing with their partner as a form of care for each. Sharing was “safe” if it occurred “only between us, not with other people.”

Cindy, who is quoted above on the importance of trust in her non-condom use with her partner, also indicated that syringe sharing was part of their relationship:

I usually have a hard time with it [injecting], I will be using mine [syringe] and it will get clogged up and then I wash his so I can use it. He will wash mine because it gets clogged and then we wash them and put them away and buy two more, and sometimes we use them again and get them confused so we end up using each other’s. We do use each other’s needles and we don’t use condoms so I guess it doesn’t matter…

Taken together, these data suggest the importance of structural reasons in driving couples’ syringe sharing, but that emotional closeness also shapes sharing practices.

Discussion

Female sex workers and their non-commercial partners along the Mexico-U.S.A. border experienced love and emotional fulfillment in their relationships in grades of intensity. Viewed through the lens of critical phenomenology, our results suggest that couples’ vulnerability within broader contexts of poverty, social marginalization, and discrimination often placed the emotional refuge they found in each other at the center of their lives. The subjective meanings that these intimate partners ascribed to their relationships carry important implications for couple-based HIV interventions.

This study adds to a growing body of literature emphasizing the importance of relationships based on love and emotional intimacy (Hirsch and Wardlow 2006; Padilla et al. 2007). Scholars suggest that economic, social, and cultural shifts have transformed personal relationships to privilege emotional intimacy (Smith 2001; Giddens 1992). Our work extends the literature to encompass the importance of emotions among female sex workers and their intimate male partners, a relationship context that previous research has rarely addressed in depth.

Our data support Ratliff’s (1999:87) assertion that “a sex worker’s occupation does not negate her ability to love and experience the same emotions as others,” and we further assert that the same holds true for their intimate male partners. Men’s emotional perspectives have largely been excluded from HIV prevention efforts (Higgins, Hoffman, and Dworkin 2010). Our work challenges conventional depictions of men as simply unemotional beings driven by their sexual needs, but rather suggests that men can be just as emotionally invested in their intimate relationships as women. One key contribution of this research is how it highlights the emotional needs of female sex workers’ male partners as an important factor in shaping HIV risk.

Like other couples studies, trust was a central feature in sex workers’ relationships (Rhodes and Cusick 2000; Corbett et al. 2009). In contrast to Hirsch and colleagues’ findings from rural Mexico, where respect was more commonly idealized by older couples (Hirsch et al. 2002), we found respect to be key among couples of all ages due in part to the extraordinary intersecting stigmas that both partners face. Couples emphasized the importance of understanding each other’s hardship and adversity, and they found respect for each other that was not available elsewhere.

We interpret our findings within the broader literature on couples who face multiple HIV risks. In the United Kingdom, for example, early studies documented frequent syringe sharing and unprotected sex among intimate partners (Rhodes et al. 1994; Rhodes, Bloor, et al. 1993). While harm reduction strategies targeted drug-related HIV risk by increasing syringe access (Stimson 1998, 1991), sexual risk has remained more difficult to address. Many drug-using couples forgo condom use with regular partners (Rhodes, Donoghoe, et al. 1993; Donoghoe, Stimson, and Dolan 1989), suggesting the importance of considering the social context and meaning of sexual relationships (Rhodes, Stimson, and Quirk 1996). Moreover, the international literature consistently finds that sex workers are less likely to use condoms with intimate partners than with clients (Hoffman et al. 2011; Wang et al. 2007). Taken together with our finding that virtually no sex worker couples use condoms, this evidence suggests a need to reconsider sexual risk reduction strategies in contexts of sex work and drug use.

Considerations for couple-based interventions

Based on our results, we offer concrete intervention suggestions to reduce couples’ HIV risk that are sensitive to their experiences. In our study, couples’ non-condom use was shaped by subjective understandings of risk. Involucrado couples often reported “not liking” condoms (Stevens and Galvao 2007) and “not being sick” (McQuiston and Gordon 2000) as reasons for non-condom use. Programmes should evaluate condom use norms and cultural scripts dictating that condoms are not pleasurable (Philpott, Knerr, and Boydell 2006). The logic of “not being sick” may be an embodied sense of fatalism that mere condoms will not provide salvation from the broader health threats looming in these couples’ lives (Meyer-Weitz 2005). This fatalism suggests that interventions should help partners realistically evaluate their risks (El-Bassel et al. 2010).

For more than half of couples in our study, including all emotionally close couples, subjective notions of love, trust, and emotional intimacy motivated unprotected sex. Demarcating the physical aspect of women’s sex work as separate from the emotional aspect of their intimate sexual relationships is “where sex ends and emotions begin.” Unprotected sex defined the very essence of many sex workers’ intimate relationships beyond the physical act itself. Simply instructing partners to start using condoms without considering the emotional context surrounding this behaviour could fundamentally alter the meaning of these relationships by casting suspicions or creating barriers between partners. Such messages may also lead couples to ignore public health interventions altogether if they are not perceived as relevant to couples’ experiences.

As such, new ways to reframe emotions as an important part of HIV interventions are needed (Warr 2001; Race 2008). Interventions for close couples should capitalise on emotions and frame safer sex behaviours as ways to protect, care, and respect their partner’s wellbeing. Strategies of “negotiated safety,” whereby couples agree to have protected sex with outside partners and engage in regular HIV/STI testing may also be a realistic alternative to reduce couples’ risk (Corbett et al. 2009).

When emotional meanings overlapped with “not being sick” in justifying condom non-use, partners perceived each other’s individual health status (e.g., free of HIV/STIs) as biological evidence of the emotional integrity of the relationship. However, studies suggest that sex workers have difficulty consistently using condoms with clients (Choi and Holroyd 2007) and infidelity is a cross-cultural phenomenon (Hirsch et al. 2007). Interventions have focused on changing risks within relationships while generally failing to address sensitive topics including outside risks. Approaches to help couples disclose outside sexual risks and HIV/STI results in safe ways are urgently needed (El-Bassel et al. 2010).

Importantly, interventions should also address dual sexual and drug-related risks from a relationship perspective, including injection drug use, syringe sharing, and sexual risk in the context of drug use (El-Bassel et al. 2011). Consistent with other studies of drug using couples (Go et al. 2006; Gyarmathy et al. 2010), nearly all injecting couples shared drugs, syringes, and injection equipment within their relationships. Less emotionally close couples also shared outside of their main relationships, suggesting the importance of addressing extra-dyadic drug risks in interventions (El-Bassel et al. 2010). For couples who injected in shooting galleries, sharing occurred within contexts of vulnerability and resource scarcity, highlighting the importance of social and structural factors in shaping risk (Finlinson et al. 2005; Page 1990). Harm reduction initiatives such as outreach and increased access to syringes should be translated to this context (Stimson 1998).

Emotionally close injector couples were less likely than others to report sharing injection equipment outside of the relationship. Love and trust play a role, but it may also signify that close couples are more socially isolated, rely on each other more generally, and do not have the same social opportunities to share apart from their partner. Research suggests that couples who spend most of their time together are more likely to share (Bryant et al. 2010). Nevertheless, while drug use practices demonstrate care, they are also shaped by the biological imperatives of addiction and pragmatics of material scarcity. These findings speak to the need to address affective as well as structural factors in reducing couples’ HIV risk that go beyond the social cognitive and individual-level behavioural approaches that traditionally have informed interventions (El-Bassel et al. 2010).

Limitations

There are of course several limitations to our study. First, we excluded women concerned about serious intimate partner violence as a result of participating; their relationship experiences may dramatically differ. The couples who self-selected into our study may be emotionally closer than other couples in this setting. Moreover, a typology imperfectly represents the complexity of these relationships. Nevertheless, this approach was intended as a heuristic to understand the linkages between emotions and risk. Our data suggested a patterned variation of emotional closeness across these relationships and couples fit into one category over another, which ultimately corresponded to patterned HIV risk behaviours.

Conclusion

Female sex workers and their intimate, non-commercial partners along the Mexico-U.S.A. border shared relationships based on love and emotional intimacy. While these relationships ranged in emotional intensity, couples separated the meaning of their primary relationships from sex work and relied on each other for emotional and material support. Not using condoms within primary relationships constituted an intentional act of agency among partners that demarcated relationship boundaries. Drug use and syringe sharing among injectors represented acts of care within broader contexts of scare material resources, particularly for emotionally close couples who tended to confine syringe sharing within their intimate relationship. Our results underscore the importance that sex workers’ intimate relationships hold in each partner’s life and suggest the need for couple-based HIV prevention interventions that address the affective dimensions of interrelated sexual and drug-related HIV risks among this vulnerable population.