Concurrent sexual partnerships among African American women in Philadelphia: results from a qualitative study

Abstract

Background

African American women are disproportionately affected by HIV/AIDS. Concurrent sexual partnerships may contribute to racial disparities in HIV infection. Little is known about attitudes and practices related to concurrency among African American women and the social, structural and behavioral factors that influence concurrency.

Methods

We recruited 19 heterosexual African American women engaging in concurrent sexual partnerships from a public health clinic in Philadelphia in 2009. We conducted in-depth interviews exploring social norms, attitudes and practices about concurrency, and the structural, social and behavioral factors influencing concurrent sexual partnerships. Grounded theory guided interview protocols and data analysis.

Results

Seventeen women reported one main and one or more non-main partners; two reported no main partners. Many women used condoms more frequently with non-main than main partners, noting they trust main partners more than non-main partners. Social factors influencing concurrency included social normalization of concurrency, inability to negotiate partners’ other concurrent partnerships, being unmarried, and not trusting main and non-main partners. Not trusting partners and the community at large were the most commonly cited reasons that women engaged in concurrent partnerships. Structural factors included economic dependence on partners, partners’ dependence on women for economic support and housing, and incarceration that interrupted partnerships. Behavioral factors including alcohol and cocaine use influenced concurrency.

Conclusions

Social, structural, and behavioral factors strongly influenced these African American women’s concurrent sexual partnerships. Many evidence-based interventions (EBIs) disseminated by the US Centers for Disease Control and Prevention (CDC) focus largely on behavioral factors and may fail to address the social and structural factors influencing African American women’s sexual networks. Novel HIV prevention interventions that address the social determinants of African American women’s HIV risks in addition to conventional HIV risk- taking behaviors are urgently needed.

Introduction

An estimated 1.1 million people in the United States are infected with HIV [1]. African Americans have HIV/AIDS rates eight times those of Caucasians [2]. Racial disparities in infection are particularly marked among African American women; African American women are diagnosed with HIV 19 times the rate of White women [3] and 64% of women living with HIV/AIDS in the United States are African American [4]. HIV/AIDS is the leading cause of death of African American women age 25 to 34 [4]. Recent research also finds that African American women living with HIV/AIDS have a lower life expectancy than HIV-positive women of other races [5] and are more likely to have HIV/AIDS-related adverse health events than their counterparts of other races [6]. Differences in individual behavioral risk factors such as drug use, condom use, and number of sexual partners do not fully account for these racial disparities in HIV infection [7].

Social determinants of HIV risks may be equally as important as individual behavioral factors in explaining some of these racial disparities in HIV infection among women [8] and other adverse HIV/AIDS health outcomes in the US. Social determinants of health are defined by the World Health Organization as “the conditions in which people are born, grow, live, work and age, including the health system” [9].

Sexual networks, or groups of people directly or indirectly linked through sexual contact, may help explain the disproportionate burden of HIV/AIDS and other sexually transmitted infections (STIs) among African Americans. Concurrent sexual partnerships, or sexual partnerships that overlap in time, have been correlated with HIV infection [10–12] and raise the risk of HIV infection more than a series of non-overlapping partnerships by eliminating the delay between sexual partnerships [13]. This delay may be particularly important during the first four to six weeks following HIV transmission, when individuals are most infectious and before the body mounts an immune response [11, 14–16].

The Centers for Disease Control and Prevention (CDC) attributes 86% of new HIV infections among African American women to heterosexual sex [17]. This may be explained in part by the role of concurrent sexual partnerships; recent research documents higher rates of concurrent partnerships among African Americans than individuals of other races [18–20]. Analyses of the National Survey of Family Growth found overall prevalence of concurrency among women to be 12%; rates of concurrency were highest among African American women (21%), followed by White women (11%), Hispanic women (8%) and Asian American and Pacific Islanders (6%) [18]. A subsequent analysis found that African American women are 1.78 times as likely to engage in concurrency as women of other races [20]. A recent study also found that urban African American women with larger social networks, particularly networks with members they depended on economically, are more likely to engage in concurrent partnerships [21].

Study Setting

Philadelphia has an HIV incidence rate of 114 per 100,000 people, approximately five times the national average [22]. African Americans in Philadelphia are disproportionately affected by HIV: although they represent 43% of the city’s population [23], African Americans accounted for 66% of new HIV infections in 2008 [22]. Philadelphia recently implemented a rapid HIV testing program in public health clinics to diagnose more people living with HIV/AIDS. Approximately 15,000 people have undergone rapid testing for HIV in the program since mid-2007; 88% are African American; HIV prevalence among this population is approximately 1.1% [24].

Study Design

This article examines the role of social, structural and behavioral factors in contributing to concurrent partnerships among urban African American women. We unpack social and individual determinants of health into distinct categories for the purposes of analysis. We define social factors as “factors related to how humans interact and relate to others,” structural factors as “physical, environmental or economic factors” [9] and behavioral factors as “individual behaviors that influence HIV risk.” Concurrency among women has also been statistically associated with social factors such as having an incarcerated male partner [10, 25], single marital status [18, 20], structural factors such as poverty [18–19, 21], and drug and alcohol use [19–21]. While these statistical associations suggest that complex social, structural and behavioral factors contribute to concurrent partnerships, the pathways through which these social determinants interact to impact concurrency and HIV infection are not yet understood. This qualitative analysis among African American women helps fill that gap.

To understand individual attitudes and practices related to concurrency and the social, structural and behavioral factors contributing to concurrency among urban African American women, we conducted in-depth qualitative interviews with heterosexual African American women engaged in concurrent sexual partnerships presenting for rapid HIV testing in a public clinic in Philadelphia. This article examines how social, structural and behavioral factors influence concurrent partnerships among urban African American women.

Data and Methods

We conducted qualitative interviews with 19 heterosexual African American women engaged in concurrent partnerships in a zip code with high HIV incidence in Philadelphia. All participants were recruited when presenting for HIV testing at a public health clinic. Participants were undergoing rapid HIV testing as part of Philadelphia’s rapid HIV testing program, which includes behavioral risk assessments prior to testing. Recruitment took place one afternoon per month for six consecutive months in 2010. Trained research assistants offered eligible women an opportunity to participate in the study immediately after their behavioral risk assessment. Women were eligible to participate in the study if they self-identified as African American and heterosexual, reported engaging in one or more concurrent sexual partnerships within the last six months in their behavioral risk assessment, reported only ever having had sex with men, were English speaking, were at least 18 years old, and were able to provide written informed consent. Both HIV-negative and HIV-positive women were recruited. Patients were able to decline to participate, and declining did not affect patients’ clinical care. Data were not collected on individuals who declined to participate in the study. Interviews took place within one week of testing. Participants provided verbal and written informed consent, and received a $20 public transportation voucher and a $30 Wal-mart™ gift card. Study protocols were approved by the Institutional Review Boards of the Philadelphia Department of Public Health and The Miriam Hospital, a Brown University affiliate.

We employed the United Nations Working Group on Concurrent Partnerships’ definition of concurrency: “Overlapping sexual partnerships in which sexual intercourse with one partner occurs between two acts of intercourse with another partner within the last six months ” [26].

We used the “grounded theory” qualitative interviewing approach in which data collection informs the development of theory and subsequent data collection and analysis [27–28]. A semi-structured interview guide was informed by literature on social [10, 12, 29], structural [10, 12, 18–20, 30–31], and behavioral factors influencing concurrent partnerships [19–20] and data on Philadelphia’s HIV/AIDS epidemic [32–33].

Interview guides included questions about participants’ sexual practices attitudes and social norms related to concurrency and condom use practices with main and non-main partners. A main partner was defined as someone with whom the respondent had an emotional bond and regular sexual intercourse, such as boyfriend, spouse, significant other, or life partner. A non-main partner was defined as someone with whom the respondent had sexual intercourse but was not considered a main partner. Guides also included questions about HIV risk behaviors, access to medical and social support services, and perceived sexual risks with partners. Guides inquired about how social factors such as marriage, parenting and availability of partners and support systems, structural factors such as incarceration, employment and poverty, and individual behavioral factors such as drug and alcohol use impacted concurrency. Interviews lasted 45 to 75 minutes and were loosely structured to allow the interviewer and respondent to introduce topics freely [34–35]. Interviews were digitally recorded and professionally transcribed; identifying information was removed from transcripts.

We coded transcripts immediately after interviews to identify the most common and salient emerging themes, and to revise interview guides to include important new themes as new topics emerged. In accordance with grounded theory, we used a purposive sampling strategy [36–37]. We stopped recruiting when we reached saturation; that is, when no new information was emerging from interviews. This determined our sample size. We employed an open coding process, which permits the researchers to analyze themes according to topics arising during the interview, rather than by predetermined topics [36–37]. Interviews were reviewed and coded by two data analysts to help ensure the reliability and validity of the study findings. Discrepancies in interpreting themes were discussed and resolved with the assistance of a third analyst to develop a final coding and analytical scheme [36–37]. Key findings were summarized in analytic memos and final results are presented here.

Results

Demographic information of study participants is presented in Table 1. The average age of participants was 35. Fourteen women were unemployed, six lacked stable housing, and eighteen had never been married. Five women reported current cocaine use, four reported current marijuana use, three reported current alcohol use, and two reported current heroin use. Two women reported they were HIV positive. All respondents reported engaging in concurrent sexual partnerships with at least two partners; seventeen women reported currently having one main and one or more non-main partners, and two women reported having only non-main partners.

Table 1.

Demographic Information of Participants

Demographic	N
Average Age (Standard Deviation)	35 (8.2)
Employment
Unemployed	14
Employed	4
Student/job training program	1
HIV Status
Negative	17
Positive	2
Incarceration History
Yes	12
No	7
Current Drug Use
Cocaine	5
Marijuana	4
Heroin	2
Alcohol Use	3
Housing*
Unstable	6
Stable	11
Transitional/in drug treatment program	2
Marital status
Never married	18
Divorced	0
Married	1

^*Unstable housing is defined as currently being homeless, or living with a friend or family member on a temporary basis. Stable housing was defined as permanently living with a spouse, steady partner or family member, or renting or owning one’s own home or apartment at the time of the interview.

We report on the most common themes that emerged from interviews, including attitudes and practices about concurrency as well as the social, structural and behavioral factors influencing concurrency among this group of low-income African American women.

Attitudes and Practices Related to Concurrent Partnerships

Although all respondents were engaged in concurrent partnerships at the time of the interview, most indicated a strong preference for monogamous relationships over multiple concurrent partnerships. Two women described their preferences and desires for monogamous relationships:

Most of it is just for an orgasm. But it still makes me feel empty; I would rather be in a relationship with somebody.

You should have just one partner. You should just find somebody that you really love and care about.

While most women preferred monogamous relationships, participants had wide-ranging opinions about what constituted an “ideal” monogamous partnership. Some preferred to be married and live with their partners, while others expressed desires to have monogamous relationships in which their partners didn’t cohabitate. For example, one woman who had previously lived with one of her partners described:

I really want one partner, just one. And I don’t want to live with him and I don’t want him living with me, like that. I just want to see just that one person.

Participants commented frequently about the stigma associated with women engaged in concurrent partnerships, and often explained they never imagined they would be engaged in concurrent partnerships. One woman recounted:

I think it comes from where you were brought up or what kind of mentality you have, how much you care about the person that you’re with, or maybe the way you carry yourself. I always have thought badly about people that slept around, but then when you’re in that situation, you try to make it seem like it’s better than what it really is. So I don’t think people really expect to be with other people. I don’t just be out looking for people; they just appear and it just happens.

Main and Non-Main Partners Fulfill Different Needs

Responses from most women also indicated that main and non-main partners fulfill different sexual, emotional, or other practical roles in their lives. One woman engaged in three concurrent partnerships reflected the common trend of finding a non-main partner to fulfill sexual needs unmet by a main partner:

I went out and had sex with another man because my boyfriend at home wasn’t paying me any attention… he wasn’t giving me sex when I wanted it. A couple days after that I had sex with another guy that I met on the bus. I had sex with him for like three or four months.

Similarly, women frequently maintained ongoing emotional, sexual and financial relationships with their children’s fathers, even if they and their children’s fathers had other main or non-main partners. One participant described how her child’s father didn’t fulfill her desire for a trusting relationship with strong emotional support, but explained that this need was fulfilled by another non-main partner:

And the other guy, I love him, too, as a friend. He’s a good person and, I mean, we can talk about anything. But with my son’s father, I couldn’t really talk to him because I really didn’t trust him and I wasn’t really trying to open my heart too much.

Condom Use with Main and Non-Main Partners

Half of respondents reported using condoms more frequently with non-main partners than main partners, whom they generally reported trusting less than their main partners. One woman who identified her child’s father as her main partner commented:

With the [other] guy, I started off having protected sex with him. But I never had protected sex with my baby’s father.

Other women explained their occasional condom use with main partners. Although respondents generally understood that concurrency raised their HIV and STD risks (including having unprotected sex with main partners), they generally used condoms more with non-main than main partners. One woman explained:

Sometimes [I use condoms]. I’m not going to say all the time… If it was a regular [partner], probably not. That’s one of the times I caught something, was when it was a regular guy that I was with.

Women who reported more consistent condom use with non-main partners frequently commented that how much they trusted their partners impacted their condom use. One woman’s comments about her non-main partner reflect this trend:

I didn’t trust him and I’m glad that I didn’t, because if I did, and hadn’t used condoms, I might be sick [infected with HIV].

Two other women’s comments echoed this sentiment:

I try to stay away from the people who I know don’t want to use condoms.

Because I wanted to protect myself from getting [HIV] by using a condom.

Social Factors Affecting Concurrency

Social Norms

Many women commented that concurrency is common in their communities, particularly among men. Most women expected their non-main partners to have other concurrent sexual partners. Two women’s comments exemplify this common theme:

I think people expect men to run around. They say that normally it’s women that do it more, but men just don’t find out about it. But I think men are expected to just go and do what they do.

You know men-- they can always find time to cheat.

In spite of most women’s desires for monogamous relationships, many women felt they could not request their partners, particularly non-main partners, be monogamous, and explained that they accepted their partners’ concurrent partnerships. Two women explained their inability to negotiate their partners’ behaviors:

I can’t tell him not to because he’s not my boyfriend. I can’t give him no ultimatums or anything; I can’t do any of that with him. I want to, but I can’t. I don’t care if he has somebody else; just don’t bring me anything home [STDs].

My other partner, I know he has sex with other girls. There isn’t anything I can do about that; I can’t have him on a leash. That’s not my boyfriend. If he were my boyfriend then I could come at his neck about it, but right now I can’t.

partnerships. One woman’s comments reflected this common theme:

I gave my boyfriend permission to get oral sex from other women. I did it because he’s a man and is not going to be satisfied with having sex with me every three weeks.

Trust of Partners

Fifteen women commented that they did not trust their partners to be monogamous. One woman engaged in an ongoing relationship with her children’s father explained:

Well, my son’s father, he always claims not to lie, saying he ain’t messing with other women. But he has other kids, and he says he’s not messing with his other children’s mother, but I get the feeling he’s with her, and it stinks. I’m not stupid; I don’t trust him.

Another woman explained why she did not trust her main partner:

It’s hard to be with someone that knows that they have something but will still have unprotected sex with you. It’s hard to look at them the same way. After having sex with him, I have an STD. I felt violated, disrespected, like he didn’t care enough to tell me or something like that.

Most women explained that suspecting their partners had other partners also influenced their own decisions to engage in concurrent partnerships. One woman explaining her concurrent partnerships recounted:

What made me do what I was doing [have more than one partner] was that he always wanted go out and be with other women.

In addition to not trusting their sexual partners, women also reported general distrust of people in their communities, and this often impacted trust of their partners. One woman’s comment exemplifies this common theme:

I don’t want anybody knowing where my family lives. That’s why I don’t let my boyfriend know where my house is. He drops me off a couple blocks away from my house. I don’t trust him. I don’t trust anybody enough to let them know where I live.

Marital Status

Only one respondent had ever been married. Women often explained that being unmarried contributed to concurrency. Many mentioned a strong preference for marrying and explained that if they were married, they would not engage in concurrent partnerships. For example, one woman remarked:

My mindset would have to change in terms of other partners. If we got married, I obviously wouldn’t cheat on my husband.

Another woman expanded on this common theme by explaining how being married and engaged in a monogamous relationship might decrease her risks for acquiring sexually transmitted infections:

If I were married, I wouldn’t have to have sex with other people. I’d be having sex with my husband. I’d be true to my husband and I know he’d be true to me. It does change things---when you are married to somebody and that’s the only man you’re having sex with, you go to the clinic together and get an HIV test and get physicals. You won’t have anything [STDs]…And you know your husband doesn’t have anything either.

Structural Factors Impacting Concurrency

Economic Dependence on Partners

In many cases, economic factors strongly impacted concurrent partnerships. Fourteen of the women were unemployed, and all were from poor inner-cities communities. Many women mentioned their financial dependence on one or more partners:

He has a catering company; any catering job that he goes on I’m a part of, so I get paid. He literally pays me. Whatever the other employees get, I get paid for.

He’ll go out and he’ll spend $100 or $200 and come in and say, babe, look what I got you. He makes sure I’ve got everything that I need, so I don’t have to go outside [the relationship]…

He—that-- was like my extra money, my get high money. I had my own money when I was working for my bills, paying my mother or paying for babysitting, helping buy food. But I needed extra money to buy clothes, to get my drugs, my drinks, go out, put gas in my car, I needed the extra money for that. And he said he would help me out.

In contrast, some women explained how their partners depended on them economically, even as they simultaneously depended on other partners:

My son’s father… he lost his job and then I was paying the bills and all of that. But my other homeboy, he would let me drive his car… I’d say I needed some money and he’d just give me money.

Another described how men depended on her for other needs:

They didn’t wanna be man enough to say, ‘okay, we’re going to get a house together.’ They always depended on me to do stuff and I did it. I was weak-minded…but I can’t do it no more.

Incarceration Disrupts Partnerships

Numerous women reported histories of incarceration, and many had partners who had been incarcerated. Women frequently explained how their main partners’ incarceration interrupted their sexual partnerships, often prompting them to undertake concurrent sexual partnerships. One woman dependent on her main partner for economic support described how his incarceration prompted her to undertake four new concurrent partnerships:

He was incarcerated for a year and a half. During that year and a half I was out there with different parties, doing sexual activities.

Another woman described how incarceration of her main partner impacted her other sexual partnerships and STD risks:

He went away to prison for ten months. Then I slept with someone else and I came back with an STD, which was trichomonas.

When her main partner was released, the respondent continued sexual relationships with both men.

Behavioral Factors Impacting Concurrency

Drug and Alcohol Use

Alcohol and drug use were often associated with concurrency. Two main themes related to this topic emerged. First, several respondents recounted how they depended on concurrent partners to supply them with money to buy drugs. One respondent’s comments reflected this common theme:

In the past year my son’s father would give me money, so that’s why I say I didn’t have to be out there prostituting. I did ask [another partner] for money because I wanted to get high. But even if I didn’t ask him he would just call me and say “What’s up, you need some money?” or “I’m getting ready to bring it up there,” and I’d accept.

Another woman explained how crack cocaine use impacted her concurrent partnerships:

[Before I was incarcerated], I was always with somebody different. It was about the drugs and having a place to stay. My mother would put me out or sometimes I would just leave because I didn’t want my daughters to see me, all crazy. So I’d hook up with somebody just to have somewhere to lay my head or supply me with some drugs.

Secondly, many women recounted how drug and alcohol used also contributed to sexual risk behavior. Seven respondents commented that cocaine use contributed to concurrent partnerships, and five explained that alcohol use contributed to concurrent partnerships. When asked how alcohol and drug use impacted their concurrent relationships, two women explained:

Half the time I was probably drunk or high, with liquor, beer, and marijuana. Drug use takes you out of yourself --I was more crazy, making the wrong decisions.

I was drinking. I don’t normally drink, but I went out with a couple friends. It was excessive and when you drink, you’re off a little.

Another woman explained how her drug addiction prompted her to undertake commercial sex work:

I was arrested for prostitution. Which is a direct result of trying to flip a drug.

Discussion

Although concurrency is nearly twice as common among African American women as white women [20] and has been associated with structural factors such as incarceration [10, 12, 25, 38] and poverty [10–12, 18, 20, 38], few studies explain attitudes and practices related to concurrency among African American women in detail. This qualitative study explains the attitudes and practices as well as social norms about concurrency among a group of low-income African American women at high risk for contracting HIV. This study also enhances understanding of how structural, social and behavioral factors interact to contribute to women’s concurrent sexual partnerships. Most women reported having a main partner in addition to one or more non-main partners. Social factors such as marriage and family ties had strong impacts on concurrency; most women indicated they would not engage in concurrency if they were married, and many women explained they continued to have sexual partnerships with their children’s fathers after they had broken up. The impact of co-parenting on concurrency has been documented elsewhere among other similar populations [39–40].

Importantly, although most women expressed strong preferences for having monogamous partnerships, and also understood that engaging in concurrency raised their risks for contracting HIV and other sexually transmitted infections (STIs), most women believed they had little power to negotiate their partners’ decisions to engage in concurrent partnerships. Many women also reported granting their main partners tacit or explicit permission to engage in other sexual partnerships, particularly non-main partners. Most reported that they expected their male partners to engage in concurrent partnerships and that this, in turn, influenced their decisions to engage in concurrency. These findings highlight important social norms about concurrency and how many social factors beyond women’s individual behaviors impact their HIV risks. These women’s desires to have monogamous partnerships also sharply contrast with other studies among urban African American men (and even urban African American men from Philadelphia) in which men expressed positive attitudes towards concurrency, even associating concurrency with masculinity [41–43]. These findings suggest important differences in attitudes and social norms about concurrency between African American men and women and are a novel contribution of this study.

Structural factors, including poverty and economic dependence on partners, impacted these women’s concurrent partnerships. Many respondents depended on one or more sexual partners economically, while reporting that one or more partners simultaneously depended on them. This finding contributes to a growing body of literature linking poverty line with concurrent partnerships among both men and women [18–20]. Philadelphia has the highest poverty rate among the nation’s ten largest cities [44–45]; high poverty rates likely contribute to this phenomenon.

Incarceration, particularly of male partners that women depended on economically, was an important structural factor affecting these women’s concurrent partnerships. Many women noted that their partners’ incarceration interrupted sexual partnerships and contributed to concurrency. Philadelphia’s incarceration rate is 4^th highest in the nation; 5.7 of every 1,000 residents are behind bars [46]. Our findings linking incarceration to interruption in partnerships and concurrency supports other similar findings elsewhere [10, 25, 30].

Most women in this group felt they were unable to negotiate monogamous partnerships and that this contributed to concurrency. This phenomenon may be attributable to the sex ratio, or ratio of eligible women to men in a community. Other research highlights how the sex ratio impacts women’s health risks through complex social networking patterns [38]. These women’s narratives highlight how the sex ratio may impact women’s concurrent partnerships and HIV risks. This important phenomenon underscores how social factors such as low marriage rates, social norms related to concurrency and structural factors interact to impact women’s attitudes and practices related to concurrency. These phenomena, coupled with the fact that these women also explained their non-main partners often fulfilled emotional, financial, or sexual roles not satisfied by main partners, underscore the need for novel interventions that address some of the social and structural factors underlying sexual networking patterns and HIV risk-taking behaviors of both men and women. These findings also suggest that many of the CDC’s evidence-based interventions that focus largely on behavior change models for HIV prevention fail to address many of the social and structural factors that put African American women at increased risk for contracting HIV.

Perhaps most notably, trust of partners played a critical role in women’s concurrent partnerships. Most women in our study indicated they trusted neither main nor non-main partners to remain monogamous. Additionally, several women indicated that they do not trust other people in the community. Many women explained not trusting partners to remain monogamous prompted them to undertake concurrent partnerships; this phenomenon has been described elsewhere as “reactive concurrency” [40]. Interestingly, even though women often believed their main partners had other partners, they frequently did not use condoms with their main partners because they reported trusting them more than non-main partners. This important phenomenon suggests that low levels of trust of sexual partners and the community at large impacts sexual behaviors, including condom use, and likely raises HIV risks for these women and their concurrent partners. Reported low levels of trust may reflect low levels of social capital in North Philadelphia. Social capital can be defined as “levels of community trust, community participation and civicness” [47–48]. Low levels of social capital in this inner-city community seems to have important impacts on concurrent sexual partnerships and condom use among these low-income African American women; similar findings are reported in another study among African American men in Philadelphia [41]. Taken together, these findings add to a growing body of evidence underscoring the importance of couching HIV prevention within a social determinants framework [8, 49].

Finally, individual behavioral factors such as drug use influenced concurrent partnerships through several means. First, women frequently reported depending economically on multiple partners to support their drug use. Secondly, women reported that drug or alcohol use directly affected their HIV risk-taking behaviors with concurrent partners. This supports recent research linking concurrency to drug and alcohol use among American women [20].

This qualitative study has several limitations. Our findings are based on a small sample of African American women undergoing rapid HIV testing at an inner city clinic in Philadelphia and may not be generalizable to other populations. This sample of women are at higher than average risk for HIV and may not be representative of the experiences of all women engaged in concurrent partnerships, or even African American women engaged in concurrent partnerships. Because respondents were selected based on reporting concurrent partnerships, our findings may be subject to some social desirability bias. Trends related to condom use with main and non-main partners may also be subject to some recall bias. Lastly, because of the qualitative nature and small size of the study, we were unable to assess whether concurrent partnerships are statistically associated with testing HIV-positive.

This study nevertheless highlights a number of important social, structural and behavioral factors that contribute to concurrent sexual partnerships among African American women of low socioeconomic status. Social factors including norms about concurrent partnerships, trust of partners, and being unmarried, as well as structural factors such as poverty and incarceration, strongly impacted women’s concurrent partnerships. The overwhelming majority of new HIV diagnoses among African American women are attributed to heterosexual transmission; these social phenomena therefore have important implications for HIV prevention as well as clinical providers of healthcare to low-income African American women. Most HIV interventions disseminated by the Centers for Disease Control and Prevention (CDC 2008) to reduce HIV infection among African American women focus on individual behavioral factors. Our findings suggest that the social determinants of concurrent partnerships are equally, if not more important, than individual behavioral factors; and that many strong social forces beyond the control of women impact their concurrent partnerships, and therefore their HIV risks. A more holistic approach to HIV prevention among African Americans should include novel, interdisciplinary interventions designed to address the social and structural factors that place African American women at disproportionately higher risk for contracting HIV.