Introduction

Unprotected sexual intercourse accounts for the largest proportion of new HIV infections globally [1–2]. In 2016, adult and adolescent women comprised an estimated 19% of the 39,000 new HIV infections in the United States (U.S.), with 86% resulting from heterosexual contact [1]. Despite multiple studies showing that correct and consistent condom use can reduce STD and HIV transmission risk by 80%, actual rates of condom use in high risk and HIV-positive women remain suboptimal [3–5]. In a large, national sample of heterosexually active female and male adults ages 15–44, 21% of women reported using a condom with last sex compared to 29% of men [6]. Even more concerning, was the statistically significant finding that only 32% of single/unmarried women with HIV-related sexual risk behaviors reported using a condom, which included those having an HIV-positive partner [6]. It is well established that high-risk sexual behaviors increase the transmission of HIV but these risks are even greater in women who are already HIV positive, because they not only have the potential to infect sero-negative partners, but they themselves are also susceptible to other pathogens or resistant strains of HIV. As such, consistent condom use in HIV positive women is even more important because it serves as a measure of both primary and secondary prevention [3–5, 7].

Many of the previous studies that have examined condom use practices in HIV-positive adult women were done in sub-Saharan Africa where HIV rates are much higher compared to the U.S. [8–10]. However, these populations may not be generalizable due to socio-economic, cultural, and religious differences. Within the U.S., the Southern states accounted for more than half of new HIV diagnoses in women in 2017 [1]. Furthermore, within the Southern states lives the highest concentration of Black/African Americans, a population that is disproportionately affected by HIV (43% of new HIV diagnoses, 59% of new HIV diagnoses in women, and 42% of all people living with HIV) despite the fact that they comprise only 12% of the nation’s population [1,11]. Additionally, a smaller percentage of African Americans living with HIV received medical care in 2014 with even fewer achieving a suppressed viral load compared to other racial/ethnic groups [12–13]. Promoting condom use as an effective method to prevent HIV transmission and development of HIV resistance is especially important in populations that are less likely to receive HIV treatment. Structural level interventions like condom distribution programs have been shown to have greater success in reducing the risks of HIV when combined with individual-level interventions [14].

It is important to understand the attitudes, behaviors, relationship dynamics and barriers faced by HIV-positive women during sexual encounters to better educate and promote safe sexual practices. The objective of this study is to evaluate factors associated with self-reported condom use in a cohort of adult, predominantly African American women receiving HIV care in Atlanta, Georgia. We hypothesized that multiple factors influence condom use, including those related to HIV as well as relationship, communication, and behavioral factors.

Materials and methods

Results

Of 187 women enrolled in the study, 170 reported having vaginal intercourse in the past six months and answered the questions pertaining to primary outcomes of interest. The mean age was 35.3 (standard deviation 7.6). Most participants were African-American, had an educational level of high school or less, and were unemployed (88%, 58%, and 66%, respectively; Tables 1 and 2). Two-thirds were married or in a committed relationship; over half of those women had been in their current relationship for at least one year.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Demographics and associations with condom use at last vaginal intercourse among sexually active HIV-positive women (N = 170).

https://doi.org/10.1371/journal.pone.0225406.t001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Demographics and associations with consistent condom use during vaginal intercourse over last 6 months among sexually active HIV-positive women (N = 170).

https://doi.org/10.1371/journal.pone.0225406.t002

Heterosexual transmission was the principal transmission mode (83%). Forty-two percent had engaged in anal sex in the last six months. The majority were currently taking ART (73%), had undetectable viral loads based on self report (64%), and did not have an AIDS diagnosis (58%). Over half reported first intercourse by age 16. Seventy-four percent used condoms the last time they had vaginal intercourse and 53% always used condoms over the last six months.

Discussion

Rates of condom use at last vaginal intercourse (74%) and consistent condom use (always) during vaginal intercourse in the past six months (53%) were suboptimal in our cohort of adult women, similar to findings in prior studies among adolescent HIV-positive females as well as a large multicenter study of HIV-positive and high-risk HIV-negative adult women in the U.S. [17–22]. Although women who used condoms during last vaginal intercourse cited HIV prevention as their main reason for condom use, the decreased rate of consistent condom use during the previous six-month period demonstrates that other factors influenced their actual practices over multiple sexual episodes. This is also supported by the trend toward reduced condom use in women who reported greater sexual frequency, which has been seen in other studies [10, 23].

Women in the oldest age group were almost five times as likely in adjusted analyses to report consistent condom use in the past six months versus women in the youngest group. Older women in our cohort may have been better able to negotiate condom use with sexual partners because of maturity, previous experiences, willingness to refuse unprotected sex, and lack of desire for future fertility [8]. Their behaviors contradict the findings of two large national surveys where the percentage of women having unprotected sex increased with age, but HIV sero-status was not taken into account [6, 24]. Our study did not reveal a difference in condom use by race, however, the majority of HIV-positive women (over 80%) in our population demographic identified as African American, which is consistent with the Centers for Disease Control and Prevention data that African American women are disproportionately affected by HIV [12].

Women who were single or dating had three times the odds of always having protected vaginal intercourse compared to their married/committed counterparts. This finding is similar to studies of at-risk heterosexual women in the U.S., including one that assessed predictors of unsafe sex in a group of at-risk women in Atlanta [25–27]. Women in urban communities have shown less intention of using condoms with “main” or steady sexual partners compared to secondary partners and they often defer decision-making to the main male partner [25–27]. Reasons for condomless sex with a main or committed partner were related to trust, infidelity, and fear of arousing suspicion, rejection or anger leading to abuse [25]. The most common reasons why women in our study reported not using condoms the last time they had sex were that they did not think about it, that they only had sex with each other, and that they trusted each other. In order to overcome the perceived barriers to condom use by women in committed relationships or with known HIV-positive partners, couples’ counseling programs that address relationship dynamics and promote equal decision-making power between heterosexual partners should be implemented per World Health Organization guidelines [28].

Self-confidence to negotiate condom use, even when a male partner does not want to wear one, was significantly associated with consistent condom use. Several studies have found an association between self-esteem levels and involvement in risky sexual behaviors, including unprotected intercourse due to antagonistic feelings about condom use and failure to disclose HIV or STD status to sexual partners [17–18, 26, 29–30]. Self-confidence and self-esteem are not mutually exclusive concepts; the ability to assert oneself in specific situations like negotiating condom use may be enhanced by having positive feelings about oneself [31]. The majority of women in our cohort (81%) reported self-confidence to discuss condoms and had revealed their HIV status to some or all of their sexual partners (79%). These findings further support the importance of: developing interventions to bolster women’s self-esteem to increase their likelihood of engaging in protected sex; and developing female-controlled modalities that do not require partner acquiescence or approval, which is especially important for women who are still working to acquire self-confidence. A meta-analysis by Crepaz et al revealed that in African American women specifically, behavioral interventions that addressed empowerment issues and negotiating skills pertaining to safe sex and condom use were successful in HIV/STD prevention [32]. Together, individual and couple’s counseling programs offer multiple forums in which women can address important personal and relationship dynamics that affect their ability to achieve higher levels of condom efficacy.

Women with only HIV-negative partners were more likely to use condoms consistently than women in sero-concordant relationships. Condom use still plays an integral role in preventing other STDs or HIV super-infections in concordant positive couples; thus, it is imperative that HIV-positive people be counseled on the value of using condoms with every sexual partner, regardless of their sero-status or ART use. Additionally, we wanted to see whether perceived viral load and ART use at the time of intercourse had any impact on condom use practices because there were concerns in earlier studies, mainly in homosexual men that ART would promote riskier sexual behaviors because of decreased HIV transmission risk [33–34]. However, there was no significant difference in consistent or recent condom use between women currently taking ART and those who were not, nor in women who reported having a detectable viral load [35]. Furthermore, results of the secondary descriptive analysis showed that our study population believed that the risk of HIV transmission is over 50% at the time of sex without condoms even with ART use, but they still did not use condoms consistently.

Finally, our results demonstrate an association between consistent condom use during vaginal intercourse and during anal intercourse within the last six months. Women engaging in protected anal intercourse were more likely to have protected vaginal intercourse. Unfortunately, the total number of women who always used condoms with anal intercourse was a small percentage (29%) of those who admitted to having anal intercourse at all, despite it being one of the riskiest sexual behaviors for infection transmission. Factors associated with condom use during anal intercourse have been assessed in men who have sex with men; however, less information is available for heterosexual, HIV-positive women [33–34]. Nasrullah’s national survey of heterosexually active women whose HIV serostatus was not reported demonstrated that the overall prevalence of condom use with anal sex was less than with vaginal sex and were affected by gender and marital status [6]. Reasons for decreased condom use with anal intercourse may be due to the absence of pregnancy risk for women, partner dynamics and communication about anal intercourse, and lack of knowledge pertaining to the risks of infection transmission with anal intercourse. Efforts should be made to evaluate anal sex risk in all women, and especially HIV-positive women, during routine screening for behavioral risk factors and provide appropriate education.

Our outcomes of interest are limited by recall bias. We relied on subjective retrospective reporting, which may be inaccurate given the need for participants to remember condom use during every coital episode. Previous studies have demonstrated that self-reported condom use may be an unreliable indicator of true condom use, and condom-failures were not considered [36]. This study may also be limited by selection bias because we used a convenience sample and did not obtain detailed information on patients who chose not to participate in the study. Finally, for our secondary analysis only, we had a different set of questions for participants who did not use condoms with last intercourse, which decreased the sample size of responses in these analyses. Our findings are most generalizable to HIV-positive, lower socio-economic status African American women living in the Southeast U.S. A strength of our study is that it broadens our understanding of factors associated with condom use in a population that has disproportionately higher rates of HIV and lower rates of treatment.

Empowering HIV-positive women to discuss consistent and correct condom use with sexual partners is critically important to prevent onward transmission. Health care providers should strive to have an open dialogue with patients about whether they engage in high-risk behaviors like anal sex or illicit drug use, the specific risks of anal sex, whether they feel comfortable negotiating condom use, and counsel on risk reduction strategies. Furthermore, dispelling misconceptions about HIV/STD transmission in HIV positive individuals with repetitive educational messaging and routine condom distribution may improve consistent condom use.

Supporting information