Riehman, Kara S. PhD*; Wechsberg, Wendee M. PhD*; Francis, Shelley A. DrPH†; Moore, Melvin; Morgan-Lopez, Antonio PhD*
APPROXIMATELY HALF OF ALL NEW HIV infections in the United States occur among youths under age 25, the majority of whom are infected sexually.1 Rates of sexually transmitted infections (STIs), which are indicators of risk behaviors for HIV, are highest among those aged 15 to 24 and racial and ethnic minorities are disproportionately infected with HIV and other STIs, particularly in the South.2 Studies have found that HIV sexual risk behavior and HIV seropositivity are prevalent not only among injecting drug users (IDUs), but also among non-IDUs, including crack cocaine users and methamphetamine users.3–6 Furthermore, a significant proportion of HIV infections occur among drug users and their sexual partners, raising concern about the transmission of HIV from high-risk groups (e.g., drug users) to potentially lower-risk groups.
Research indicates that HIV risk may be higher among persons who report being in monogamous relationships as a result of lower condom use in longer, more committed partnerships.7–11 Individuals may be under the assumption that they are in monogamous relationships and do not need protection when, in fact, their partners may not have the same attitude about the relationship and may have other sexual partners. However, although researchers have begun to examine how certain partnership characteristics, including sexual concurrency and discordant reporting, influence STI risk, little research has been conducted on beliefs about monogamy and whether these beliefs are associated with greater HIV risk.
Research has found certain characteristics of partnerships are associated with HIV/STIs. Concurrent or overlapping partnerships—called sexual concurrency—may be as important as nonoverlapping multiple partners in the spread of HIV and STIs.12–15 Discordance in sexual behavior between partners also is associated with STI acquisition.12 Gorbach and colleagues examined the association between partnership-level STIs and discordance in various relationship characteristics, and found discordance in relationship commitment and number of lifetime partners were associated with current or recent partnership-level STI.12 Discordance in perceptions of partners' risk behavior and attitudes also has been found to increase HIV/STI risk. Drumwright et al found that lack of awareness of partner's sexual concurrency was associated with current STI status.16
Hence, the context and dynamics of committed partnerships may contribute to risk. Although consistent and correct condom use is associated with reduced HIV and STI transmission,17 individuals are less likely to use condoms in steady relationships than in casual or trading relationships. Individuals in close, longer-term relationships may assume their partners are monogamous and that partners do not have sex outside of the relationship.7–11,18 However, recent research indicates many individuals are unaware their partners engage in sex outside of the relationship.16,12,19,20
Researchers examining monogamy as a risk-reduction strategy have recently begun to focus on the meanings attached to individuals' perceptions of monogamy. Hearn and colleagues conducted in-depth interviews with women in methadone treatment and explored how women value monogamy in their relationships, and how this was related to their ability to protect themselves.21 Britton and colleagues found that women define monogamy in different ways, and some interpretations may lead to greater HIV risk.22 However, little is known about the extent to which individuals' perceptions of monogamy in their partnership are shared by their partners. In addition, although researchers have begun investigating concurrency, discordance in behavior, and discordance in perceptions of commitment and concurrency, less is known about discordance in perceptions of monogamy beliefs, whether discordance in beliefs about monogamy lead to greater HIV and STI risk, and whether other relationship dynamics are associated with monogamy beliefs and discordance in these beliefs in the partnership. The relationship characteristics of young adults are of particular interest because younger individuals are less likely to be in marital relationships and thus may have more sexual partners.
In the current article, we use data collected from both members of steady sexual partnerships of predominantly black young adult alcohol and/or drug (AOD) users in North Carolina to examine the association between individual and partnership characteristics with condom use, sexual concurrency, and discordance in monogamy perceptions to gain a better understanding of how discordance in monogamy beliefs may influence HIV and STI risk.
Materials and Methods
Between October 2003 and February 2004, 47 individuals aged 18 to 25 who reported drug and/or heavy alcohol use (index participants) and their adult steady sexual partners (total = 94 individuals) were recruited to participate in a cross-sectional pilot study to better understand risky partnership dynamics among young adult drug users. Eligibility criteria included the index participants being between the ages of 18 and 25, AOD users (defined as reporting alcohol use or illicit drug use on at least 13 of the past 90 days), residing in a noninstitutionalized setting; and involved in a steady heterosexual relationship with someone 18 years of age or older. Partners were eligible whether they reported drug or alcohol use. A steady relationship was defined as having had a minimum of 2 dating or sexual encounters over a period of at least 2 weeks with someone of the opposite sex with the expectation that the relationship would continue in at least the near future. Index participants were recruited from street outreach in Durham County and from a methadone clinic in Wake County, North Carolina. Recruitment through street outreach involved targeting drug-using hot spots in Durham identified through previous research studies with similar populations. Index and partner participants were administered the Couples' Risk Interview, which collected data on partnership characteristics (e.g., sexual concurrency, perceptions of monogamy, relationship duration, commitment, communication/trust, safe-sex negotiation, relationship power, and conflict); HIV/STI risk behavior (e.g., condom use, knowledge of partner HIV/STI status); drug use (e.g., current and history); selective psychologic measures (e.g., depression, self-esteem); and some background information (e.g., demographics, sexual and physical abuse history).23
All study participants who gave informed consent were assessed through self-report. The data collection session took approximately 60 minutes to complete and was conducted face-to-face with pen-and-paper interviews at the time of recruitment. Index and partner participants were interviewed simultaneously but separately. Informed consent and data collection procedures were approved by RTI International's Office of Research Protection and Ethics. Each participant received $15 for completion of the interview.
In this article, we examined 3 variables of interest and their association with several individual risk behaviors and partnership characteristics. Consistent condom use is a protective factor for HIV and STI transmission and, as such, was the primary dependent variable of interest. Sexual concurrency also is associated with HIV/STI transmission, although less research has been conducted identifying factors associated with concurrency, including condom use. Finally, although it is known that condoms are less likely to be used in monogamous relationships, very little quantitative empiric work has been conducted examining individuals' assessment of their partners' monogamy perceptions and how this relates to actual condom use.
Consistent condom use with current partner was measured as the proportion of protected vaginal sex episodes with the current partner in the past 30 days reported by each individual. This proportion was dichotomized into “never or inconsistent use” (0–99% of the time) and consistent use (100% of the time). Sexual concurrency was measured using reported dates of last intercourse with previous partner and first intercourse with current partner. Overlap of these dates was categorized as a concurrent partnership. Discordance in monogamy perceptions was measured using 2 questions. Individuals were asked, “Do you think of yourself as being in a monogamous relationship with your partner?” and “Do you think your partner views him- or herself as being in a monogamous relationship with you?” Individual and partner responses to these items were compared. Discrepant responses of the individual's response to the second question with the partner's response to the first question were classified as “discordant” and matching responses were classified as “concordant.”*
We examined associations between individual and relationship characteristics with these 3 variables of interest. Individual demographic variables included age, race/ethnicity, education, marital status, having any children, current involvement in the criminal justice system, employment, and income. Individual sexual risk variables included number of lifetime sexual partners, number of lifetime casual partners, number of lifetime steady sexual partners, and number sexual partners in past year, all of which were measured as separate continuous variables. Number of steady and casual sexual partners was assessed with 2 questions: “Of these [total number of] lifetime partners, how many would you consider to be steady or main partners? How many would you consider to be casual partners or partners that you did not consider steady or main partners or commercial partners?” Participants also were asked whether in the past year they had sex while high or drunk, had sex when their partner was high or drunk, had sex with an injecting drug user, traded sex for money or drugs, and whether they ever had a diagnosed STI, all of which were measured as separate yes–no categorical variables. We also examined individual substance use, because this is associated with greater HIV/STI risk behavior. We examined past 90-day use (yes–no) of alcohol to intoxication, marijuana, cocaine, crack, heroin, and the total number of substances reported ever used (continuous variable).
We examined a variety of partnership characteristics that we hypothesized might be related to condom use, concurrency and monogamy perceptions. Relationship duration was measured in total number of months individuals reported being in their current relationship. Relationship violence in the past month was measured as a positive response to any of 3 items: whether the current partner pushed, grabbed, or punched/hit the individual in the past month (reported by either member of the couple). Relationship conflict was measured as how frequently (never, less than once a month, once a month, 2–3 times a month, once a week, 2–3 times a week or almost daily) individuals and their partners argued or got in a verbal fight (reported by either member of the couple and collapsed into weekly or more vs. less than weekly). We also measured whether individuals told their partner they had a concurrent relationship (yes–no). Relationship commitment was measured using a 4-item Likert scale asking individuals to report how attached/committed they feel toward their partner (coefficient α = 0.88). Communication about partner risk and behavior was measured using a 7-item scale with items asking whether individuals ever asked their current partner about a variety of their partner's risk behavior, including whether he or she has HIV, whether the partner had ever taken an HIV test, whether the partner had ever had an STI, whether the partner had sex with someone who injects drugs, the partner's number of lifetime sexual partners, and whether the individual asked their partner to get tested for HIV or STIs (coefficient α = 0.83). Communication confidence was measured with a 10-item Likert scale with items asking how confident individuals felt (not at all, a little, somewhat, extremely) that they could ask their partner about their partner's risk behavior, to request condom use, and to request that the partner get tested for HIV and STIs (coefficient α = 0.87).
We present descriptive frequencies and means for all variables of interest. We conducted univariate analysis examining the association between independent variables and the 3 dependent variables of interest. Statistical significance of categorical variables was assessed using the Pearson chi-square statistic and one-way analyses of variance were used for continuous variables. We then conducted multiple logistic regression to assess the associations between individual and relationship factors and condom use. Generalized hierarchical linear modeling (GHLM) was performed using the GLIMMIX procedure in SAS 9.1 to account for 1) binary outcomes and 2) nonindependence of observations within couples. We included only those variables that were significantly related to condom use in univariate analyses at the P <0.05 level.
Table 1 presents sample characteristics and individual risk behaviors. The sample consisted predominantly of low-income blacks (83.9%) with low education levels and employment. The average age of index participants was 21.8, with steady partners being slightly older (24.7). Very few individuals were married (12.8%), and the majority (54.8%) was cohabiting with their current partner. Relatively few individuals were currently involved in the criminal justice system, although only noninstitutionalized individuals were eligible for study participation. Significant individual sexual risk factors were reported in this sample. By their early to mid-20s, these young adults reported an average of 13 lifetime sexual partners with an average of almost 3 partners in the previous year. Most previous partners were casual rather than steady partners (7.7 vs. 4.0). Almost three fourths of the sample reported AOD use before sex in the previous year, and over one third reported having sex while their partner was high or drunk. Only 4.3% reported having sex with an injecting drug user. Almost 10% of this sample reported trading sex for money or drugs in the past year. In addition, 28.1% reported having had a diagnosed STI in their lifetime.
Eighty-three percent of index participants reported use of an illicit drug in the previous 90 days; thus, the majority of individuals were eligible for the study through illicit drug use (analysis not shown). The most common substance reported in this sample was marijuana, with almost three fourths of the sample reporting use in the previous 90 days. Fewer individuals reported alcohol use or alcohol use to intoxication in the past 90 days (53.2% and 35.1%, respectively). Crack use was the next most reported drug (13.8%) followed by powder cocaine (7.4%) and heroin (6.4%). On average, individuals reported using almost 4 illegal drugs over their lifetime.
Table 2 presents characteristics of current partnerships. The average duration of the current partnership was 26 months, with a range of zero to 104 months. Only 37.4% of the sample reported consistent condom use with their current partner in the previous month. Over one third of the sample (37.5%) reported having a concurrent partnership, yet only 24% told their partner about the overlapping partnership. In addition, only 48.5% of those with concurrent partners used condoms consistently with the concurrent partner. The partnerships in this sample were characterized by significant levels of violence and conflict. Thirty-seven percent of the sample reported some type of relationship violence in the past month, and just over 50% reported weekly or more frequent conflict with their partner. In addition, although 83% of the sample considered the relationship to be monogamous and 90% thought their partners considered the relationship monogamous, just over one third were discordant in their monogamy perceptions (i.e., were wrong about whether their partner considered the relationship monogamous). Relationship commitment was relatively high (average of 15.8 on a scale of 6–20). Participants reported relatively high levels of communication and communication confidence, with these scores being slightly above the scale midpoint.
Table 3 shows the univariate association between consistent condom use, concurrency, and discordant monogamy with individual and partnership characteristics and risk behavior. Only those associations that were significant at the P <0.05 level are presented. Several factors were associated with consistent condom use in this sample at the univariate level. Fewer lifetime substances, not having sex while a partner was high or drunk, and shorter relationship duration were associated with consistent condom use. Individuals who reported greater relationship commitment were less likely to be consistent condom users. Relationship conflict was also associated with condom use, with those reporting weekly or more frequent conflict less likely to be consistent condom users. Discordant monogamy perceptions also were associated with condom use, with those who were discordant being more likely to consistently use condoms.
Very few individual or relationship characteristics were associated with either concurrency or discordant monogamy. Men were more likely to report having had a concurrent sexual relationship. Those who perceived their partners as being monogamous were less likely to have had a concurrent relationship. Monogamy discordance was associated with fewer lifetime numbers of substances. Those reporting lower communication confidence and lower relationship commitment were more likely to misperceive their partner's monogamy beliefs.
Table 4 presents the results of the GHLM predicting consistent condom use. Most factors associated with consistent use at the univariate level remained significant in the regression model. Greater lifetime number of substances, longer relationship duration, and at least weekly relationship conflict were associated with a lower likelihood of consistent condom use. Those with discordant monogamy beliefs were significantly more likely to consistently use condoms than those who accurately perceived their partners' monogamy beliefs. Sex while a partner was high or drunk and relationship commitment became insignificant in the full model when controlling for these other variables.
We examined a variety of individual risk and partnership characteristics in a sample of predominantly black, young adult, AOD-involved couples recruited from street outreach and a methadone clinic. Results indicate that many of these young adult drug users exhibit significant HIV/STI risk behavior. The sample had an average of 13 lifetime sexual partners and approximately one third reported consistent condom use with their current partner. Over one third of the sample had concurrent partnerships, very few of whom told their current partner about the overlapping relationship. Although we were not able to conduct biologic testing for STIs, our findings suggest that given the high rate of concurrency in this population, these young adults are at substantial risk for contracting STIs, including HIV. Indeed, 28% of the sample reported having had a diagnosed STI in their lifetime.
We also examined perceptions of monogamy in the current partnership and whether being wrong about partners' monogamy beliefs was associated with greater risk (i.e., less condom use). To date, this is the first quantitative study to examine discordance in monogamy perceptions and its association with risk behavior. In the present sample, over one third of the young adults were wrong about whether their partners considered the relationship monogamous. However, contrary to our hypotheses, findings indicate that having discordant monogamy perceptions was associated with less risk as measured by condom use with the current partner: those who were wrong about their partners' monogamy beliefs were more likely to use condoms. One possible explanation for this finding is that discordant monogamy perceptions may be a proxy for other partnership characteristics that would be associated with condom use, including relationship duration, communication, and commitment. For example, those who are discordant may not know their partners well, may be less committed to the relationship, or may communicate less with them compared with those who were not discordant. However, partner communication was not associated with either discordant monogamy or condom use, although confidence in communication was associated with discordant monogamy. Also, although discordant monogamy perception was associated with lower relationship commitment, and lower commitment was associated with consistent condom use, the association between monogamy perceptions and condom use remained even when controlling for relationship commitment in the multiple regression model. Finally, the effect of monogamy discordance remained when controlling for relationship duration.
Interestingly, neither individuals' own beliefs about monogamy in the relationship nor their beliefs that their partners were not monogamous influenced condom use. Only the couple-derived measure of discordance had an influence on condom use. This suggests that perhaps in some partnerships, there is a level of unstated skepticism about their partner's monogamy by one member of the couple that may lead to the outcome of condom use. An individual may not consciously believe that their partner is not monogamous (or may be unwilling to admit this to an interviewer) and may answer that they believe their partner considers him- or herself in a monogamous relationship. However, there may be something about the partner's attitude or behavior, perhaps subtle verbal or nonverbal cues, indicating that the partner does not consider him- or herself in a monogamous relationship that leads to a greater willingness (or even insistence) to use condoms.
The lack of association between sexual concurrency and condom use indicates that although a significant number of individuals have sex outside their steady partnerships, they may not be acting to protect their current partners from contracting STIs. However, our data do not provide information on whether sexually concurrent individuals discuss or practice “negotiated safety”; that is, an agreement between partners that although condoms are not used in the steady partnership, partners who engage in sex outside the partnership will use condoms with extradyadic partners. Further research should examine factors associated with negotiated safety agreements in drug users' steady partnerships.
Only one of the individual or partnership characteristics examined here was associated with concurrency. Individuals who believed their partners did not consider the relationship to be monogamous were more likely to have had a concurrent relationship. Interestingly, individuals' endorsement of their own monogamy beliefs was not associated with concurrency. We might expect that those who had a concurrent partner would be less likely to consider themselves to be in a monogamous relationship; however, this was not the case. Perhaps individuals may justify having a concurrent partner because they think their partner is not monogamous. In addition, individuals in longer relationships may have engaged in concurrency earlier in the relationship but currently think of their relationship as monogamous. Additional research should be conducted to better understand factors associated with concurrency to address this risk behavior.
Greater relationship conflict was associated with inconsistent condom use in this study. However, the current study does not provide details about the context or nature of conflict that may lead to less condom use. Future research should explore aspects of relationship conflict that may be important for risk behavior as well as examine potential gender differences in conflict and relationship violence that may lead to greater risk behavior.
This study is limited in that it includes a small, nonrepresentative sample of young adult couples recruited through street outreach and from a treatment facility. The high levels of risk behavior identified in this population may not be found in other subpopulations of young adult, AOD-involved couples, including those in college or working full-time. However, this sample enabled us to begin to identify specific partnership dynamics among AOD-involved couples that may increase their risk for HIV/STIs.
This study provides an initial attempt to untangle young adults' perceptions about monogamy in their relationships and how these perceptions may increase or decrease their HIV/STI risk. Sexual monogamy is strongly valued in this society; yet despite explicit discussions or implicit assumptions of monogamy, many individuals secretly engage in sex outside their primary relationships. Although individuals may suspect their partners are engaging in extradyadic sexual activity, they may deny this behavior or be hesitant to confront their partners for fear of relationship loss, relationship discord, or partner violence. Because many couples stop condom use after establishing monogamy, trust, and commitment in the relationship, some individuals may be at risk from their partner's sexual concurrency. However, it is unrealistic to expect that individuals in long-term, established relationships will initiate condom use for HIV/STI prevention given assumptions and societal expectations of monogamy. The challenge becomes to increase individuals' awareness of potential partner risk behavior in a context that is safe and does not unnecessarily introduce relationship discord. Thus, interventions for couples are needed that focus not on condom use as the primary prevention strategy, but on other risk-reduction strategies appropriate for committed long-term relationships such as increasing awareness of partner risk behavior, negotiating safety, and conflict resolution skills.
1. HIV Prevention Strategic Plan Through 2005. Atlanta: Centers for Disease Control and Prevention, 2001.
2. North Carolina Department of Health and Human Services (NCDHHS). Epidemiologic profile for 2006 HIV/STD prevention and care planning. 2005. Raleigh, NC: North Carolina Department of Health and Human Services, 2005. Available at: http://www.epi.state.nc.us/epi/hiv
. Accessed November 14, 2005.
3. Hawke JM, Jainchill N, DeLeon G. Adolescent amphetamine users in treatment: Client profiles and treatment outcomes. J Psychoactive Drugs 2000; 32:95–105.
4. Molitor F, Traux SR, Ruiz JD, et al. Association of methamphetamine use during sex with risky sexual behaviors and HIV infection among non-injection drug users. West J Med 1998; 168:93–97.
5. Tapert SR, Aarons GA, Sedlar GR, et al. Adolescent substance use and sexual risk-taking behavior. J Adolesc Health 2001; 28:181–189.
6. Wechsberg WM, Dennis ML, Stevens SJ. Cluster analysis of HIV intervention outcomes among substance-abusing women. Am J Drug Alcohol Abuse 1998; 24:239–57.
7. Bazargan M, Kelly EM, Stein JA, et al. Correlates of HIV risk-taking behaviors among African American college students: The effect of HIV knowledge, motivation, and behavioral skills. J Natl Med Assoc 2000; 92:391–404.
8. Eldridge GD, Reitman D, Little CE, et al. Factors influencing condom use among African American women: Implications for risk reduction interventions. Am J Commun Psychol 1998; 26:7–28.
9. Friedman SR, Flom PL, Kottiri BJ, et al. Consistent condom use in the heterosexual relationships of young adults who live in a high-HIV-risk neighborhood and do not use ‘hard drugs.' AIDS Care 2002; 14:493–507.
10. Jadack RA, Hyde JS, Keller ML. Gender risk and knowledge about HIV, risky sexual behavior, and safer sex practices. Res Nurs Health 1995; 18:313–324.
11. Macaluso M, Demand MJ, Artz LM, et al. Partner type and condom use. AIDS 2000; 14:537–546.
12. Gorbach PM, Drumright LN, Holmes KK. Discord, discordance, and concurrency: Comparing individual and partnership-level analyses of new partnerships of young adults at risk of sexually transmitted infections. Sex Transm Dis 2005; 32:7–12.
13. Manhart LE, Aral SO, Holmes KK, et al. Sex partner concurrency: Measurement, prevalence, and correlates among urban 18–39-year-olds. Sex Transm Dis 2002; 29:133–143.
14. Morris M, Kretzschmar M. Concurrent partnerships and the spread of HIV. AIDS 1997; 11:641–648.
15. Potterat J, Zimmerman-Rogers H, Muth S, et al. Chlamydia transmission: Concurrency, reproduction number, and the epidemic trajectory. Am J Epidemiol 1999; 150:1331–1339.
16. Drumright LN, Gorbach PM, Holmes KK. Do people really know their sex partners? Concurrency, knowledge of partner behavior, and sexually transmitted infections within partnerships. Sex Transm Dis 2004; 31:437–442.
17. National Institutes of Health. Workshop Summary: Scientific Evidence on Condom Effectiveness for Sexually Transmitted Disease (STD) Prevention. Summary Report, June 12–13, 2001, Herndon, VA.
18. Misovich ST, Fisher JD, Fisher WA. Close relationships and elevated HIV risk behavior: Evidence and possible underlying psychological processes. J Gen Psychol 1997; 1:72–107.
19. Fals-Stewart W, Birchler GR, Hoebbel C, et al. An examination of indirect risk of exposure to HIV among wives of substance-abusing men. Drug Alcohol Depend 2003; 70:65–76.
20. Riehman KS, Wechsberg WM, Zule WM, et al. Risky Relationship Dynamics Among Young Adult Drug Users. Presented at the American Public Health Association Annual Meetings, Washington, DC, November 2004.
21. Hearn KD, O'Sullivan LF, El-Bassel N, et al. Intimate partner violence and monogamy among women in methadone treatment. AIDS Behav 2005; 9:177–186.
22. Britton PJ, Levine OH, Jackson AP, et al. Ambiguity of monogamy as a safer-sex goal among single, pregnant, inner-city women: Monogamy by whose definition? J Health Psychol 1998; 3:227–232.
23. Riehman KS. Couples' Risk Interview. Research Triangle Park, NC: RTI International, 2004.