Secondary Logo

Journal Logo


Do People Really Know Their Sex Partners?

Concurrency, Knowledge of Partner Behavior, and Sexually Transmitted Infections Within Partnerships

Drumright, Lydia N. MPH*; Gorbach, Pamina M. MHS, DrPH; Holmes, King K. MD, PhD

Author Information
Sexually Transmitted Diseases: July 2004 - Volume 31 - Issue 7 - p 437-442
doi: 10.1097/01.OLQ.0000129949.30114.37
  • Free

CONCURRENT SEXUAL PARTNERSHIPS have been defined as those in which there is partner overlap, for example, when 1 or both of the partners has other sexual partners while continuing sexual activity with the original partner.1 Prevalences of reported concurrent partnerships have ranged from 32% to 54% among adolescents2,3 and 12% to 40% among adults1,4,5 in different regions of the United States. An individual’s practice of having concurrent partners (individual’s concurrency) has been associated with acquisition of sexually transmitted infections (STIs), including Chlamydia trachomatis in the United States,6Neisseria gonorrhoeae in Trinidad,7 and bacterial STIs in adolescents,3 even after adjusting for the individual’s number of partners. These studies assessed individuals’ reports of having concurrent partnerships. Individuals who report having partners with concurrent sexual partners (partner’s concurrency) are also more likely to be transmitters of syphilis (defined as having a partner at an earlier stage of syphilis than oneself) than those who do not report their partner’s concurrency.8 To further understand the association between concurrency and STI acquisition and transmission, the risk behaviors of both members of a sexual dyad must be considered. Although individual and partner concurrency reported by the individual have been associated with STI,9 assessments of partners’ behaviors could be inaccurate10,11 because these assessments are rarely validated with data from the partners themselves. There is a need to examine the contribution of concurrency to the risk of specific STIs independent of other risk factors and to untangle the relative importance of an individual’s concurrency versus the partner’s concurrency on an individual’s risk for STIs. Adequate examination requires obtaining data from both members of a partnership.

Theoretical evidence from mathematical models of STI transmission suggests that an individual’s STI risk depends as much on the partner’s behavior as on the individual’s own behavior.12–17 Provision of actual data linking behaviors of both partners to STI prevalence in each partner and within the partnership can guide selection of parameters for use in models of STI transmission dynamics. Our study collected information from both members of sexual partnerships to examine whether young adults are aware of their partners’ concurrent involvement with other sexual partners; whether concurrent sexual partnerships reported by these individuals or by their partners independently predict STI in the individual; and for young adults whose partners have concurrent sexual partners, whether those who do not know of their partners’ concurrent sexual partnerships are at greater risk for STI than those who do know.

Materials and Methods

Between August 2000 and September 2001, 96 young heterosexual adults (ages 18–25 years) seeking care at sexually transmitted disease (STD) and family planning (FP) clinics in San Diego County were recruited together with their new sexual partners (ages 18–30 years) to participate in a 12-month longitudinal study of the natural history of sexual partnerships. A new sexual partner was defined as someone with whom the participant had had vaginal intercourse for the first time within the previous 3 months. Forty-three index partners were male and 53 were female. Eligible partners attended the initial visit together as a couple and were interviewed simultaneously in separate, private locations using a 240-item audio-computer-assisted interview of 60 minutes duration. Urine specimens collected from all participants at the initial visit were tested for C. trachomatis and N. gonorrhoeae by ligase chain reaction (LCx; Abbott Diagnostics, Abbott Park, IL); and urine specimens from men and self-collected vaginal swabs from women were tested for Trichomonas vaginalis by culture (InPouch; Biomed Diagnostics, San Jose, CA). All participants were deemed mentally fit at the time of enrollment, received a study code to protect confidentiality, received financial compensation for participation, and gave informed consent. The Institutional Review Boards of the University of California, Los Angeles and San Diego State University reviewed and approved the protocol.

Sexual partner concurrency after the partnership had been established was determined by asking both members of each partnership to report whether they had had sexual activity with anyone other than their enrollment partner after they first had sexual intercourse with that partner. Participants were also asked if they thought that their enrollment partner had had sexual activity with others after the partnership had been established. Frequency of condom use, number of sexual partners in the past 1 month, 12 months, and lifetime, and sexual history and demographic characteristics were also assessed.

Data used for these analyses were from the initial baseline visit interviews. Individuals’ knowledge of their partners’ concurrency behaviors was assessed by the kappa statistic to test the overall agreement of couples’ responses to questions about concurrency while taking into account agreement that would be expected by chance.18 The level of agreement was assessed by predefined categories for kappa as poor (≤0.40), fair (0.41–0.60), good (0.61–0.80), and excellent (0.81–1.00).19 Associations between each individual’s current STI status and that individual’s awareness of partner’s concurrency, the partner’s reported concurrency, or the individual’s own concurrency, and additional risk factors for STI, were examined using chi-squared analysis and t tests. Three different multivariate logistic regression models were used to examine associations between current STI and the different measures of concurrency or knowledge of the presence of other risk factors. Each of the 3 models included all of the same potential confounding and influencing variables in examining the associations between concurrency and STI. These variables included number of sexual contacts in the past month for the individual or for the partner, age at sexual debut, the individual’s education level, ethnicity, condom use, commitment to the partnership, residence in south San Diego, and time from meeting the partner until sexual contact. To prevent collinearity of data, the 3 models differed by the concurrency measure (individual’s concurrency, partner’s concurrency or knowledge of partner’s concurrency) but not by covariates. All models controlled for reported number of sexual contacts during the past 12 months to assess whether any associations between concurrency measures and STI were independent of the number of sexual contacts. Multivariate models that examined associations between individuals’ concurrency and STI included the individual’s number of sexual contacts during the past 12 months, whereas those that examined partner’s concurrency and awareness of partner’s concurrency included the partner’s number of sexual contacts as a potential confounder. All models were analyzed using survey estimation techniques to control for the presence of both members of a couple in the models. Data were entered by participant response into Ci3 (Sawtooth Technologies, Northbrook, IL) and analyzed using STATA version 7.0 SE (STATA Corp., College Station, TX).


Of the 192 individuals from 96 partnerships who completed the baseline questionnaire and provided specimens for STI tests, 18 (8 women, 10 men) tested positive for C. trachomatis, three (2 women, 1 man) for T. vaginalis, and 1 woman tested positive for both C. trachomatis and T. vaginalis. None tested positive for N. gonorrhoeae. Overall prevalence of either of these STIs was 11.5% at baseline, with equal numbers of infected men and women. The median age of participants was 22 years (mean, 21.8 years); men were significantly older than women (22.6 vs. 21.0 years, respectively; P = 0.001). The mean number of years of education for participants was equivalent to completion of high school (median, 13 years). Multiple ethnic groups participated, with less than half of the sample reporting white ethnicity (46%). Age of sexual debut averaged 16 years for both genders, and men and women reported similar numbers of sexual partners during the month (1.4) and 12 months (4.3) before the interview. The reported lifetime number of sexual partners was higher for men (20.2) than for women (13.9, P = 0.04); however, after controlling for age, lifetime numbers of partners did not differ by gender. Only 48 participants (25%) reported consistent condom use during the month before interview, whereas 56 (30%) reported never using condoms during that time period. Although 32% of individuals reported having concurrent sexual partners, only 16% reported thinking that their partner was having concurrent sexual relationships. Participants from STD and FP clinics reported no significant differences in demographic or background sexual characteristics.

Individuals’ Ability to Predict Sexual Partner’s Concurrency

Sixty-one (32%) of the participants reported having concurrent sexual partnerships. The overall ability for individuals to predict whether their partners were having concurrent sexual partnerships was poor (kappa = 0.17; Table 1). Of the 132 individuals whose partners reported not having concurrent partners, 116 (86%) did correctly report their partner’s nonconcurrency; however, of 61 individuals whose partners reported having concurrent partners, only 16 (26%) correctly reported the partner’s concurrency. Of those individuals whose partners did not report concurrent partners, 15 (14%) believed that their partner had concurrent partners.

Individuals’ Ability to Predict Partners’ Concurrency Using Kappa Analysis

Risk Factors for Current Sexually Transmitted Infection

Univariate Analysis.

In univariate analysis treating partnership as a cluster and clinic type as a stratum, individuals who did not know their partners’ concurrency status had significantly more STI (C. trachomatis or T. vaginalis infection) than those who knew that their partner had concurrent partners (Table 2). Similarly, individuals who had sexual contact within 1 week of being acquainted with their partner were more likely to have an STI than those who reported waiting longer before initiating sex (23.2% vs. 6.2%), and more of those living in south San Diego than those who lived elsewhere had an STI (22.7% vs. 8.1%). A partner’s practice of concurrency showed a trend toward an association with current STI in the individual (P = 0.056), whereas the individual’s own concurrency showed no such association (P = 0.6). Commitment to the partnership, condom use during the past month, ethnicity, education, age at sexual debut, and number of sexual contacts of the individual or partner during the past 12 months were not significantly associated with current STI (P >0.10).

Associations Between Current Sexually Transmitted Infection (STI) Status (Either Partner Having an STI) and Risk Factors for STI (n = 192): Univariate Analysis

Multivariate Models.

Three multivariate models were used to determine associations between an individual’s current STI and different measures of concurrency (individual’s concurrency, partner’s concurrency, and individual’s knowledge of partner’s concurrency). Partner’s concurrency (model 2), but not individual’s concurrency (model 1), was significantly associated with an individual’s current STI after controlling for all other potential risk factors and number of sexual partners during the past 12 months (Table 3). In model 3, poor knowledge of a partner’s concurrency was also associated with STI; individuals who were unaware of their partner’s concurrency status, either believing their partner was not having other partners when their partner was (OR, 4.5) or believing their partner was having other partners when their partner was not (OR, 4.7), were more likely to have a current STI than those who correctly reported their partners’ concurrency status. In all 3 of the models, living in South San Diego County and having sexual contact with the enrollment partner within 1 week after meeting were associated with the individual’s current STI. The individual’s report of being committed to continuing the partnership was also associated with increased prevalence of STI in multivariate models examining partner’s concurrency (model 2) and awareness of partner’s concurrency (model 3), but not in the model including the individual’s concurrency (model 1). Consistent condom use in the past month, gender, ethnicity, education, age at sexual debut, and number of sexual contacts in the past 12 months were not significantly associated with an individual’s current STI in any of the 3 models.

Three Multivariate Logistic Regression Models of Current Sexually Transmitted Infection (STI) Using 3 Different Concurrency Measures (n = 192)

Consistent condom use was reported in more of the high-risk partnerships (ie, those in which both partners were having concurrent partners) than in the partnerships in which only 1 partner or neither was having a concurrent partner (Fig. 1, P = 0.017). Individuals in partnerships in which only the partner reported concurrency were less likely to be aware of the partner’s behavior than individuals in partnerships in which both partners reported concurrency, neither reported concurrency, or when only the individual had a concurrent partner (P <0.001).

Fig. 1:
Condom use, current sexually transmitted infection, and awareness of partner’s concurrency broken down by whether neither member, 1, or both members of the partnership had concurrent sexual partners (n = 192).


Young adults are not able to accurately report on a partner’s behavior unless that behavior is something they practice together. High levels of agreements about coital frequency and condom use within partnerships have been found among heterosexual20–22 and same-sex couples23,24 and among the partnerships in this study (kappa = 0.69, data not shown). However, neither the behaviors that partners practice separately such as having concurrent sexual relationships10,11 nor the partners’ STI/HIV status25 are accurately perceived. Our findings indicate that this poor ability to assess a partner’s behavior is associated with increased risk of acquiring an STI.

Our previous qualitative research26 identified different types of concurrent partnerships with potential for differential risk of STIs. For example, “experimental concurrency” that occurs at the beginning of partnerships when individuals are exploring sexual activity with more than 1 new partner, or “reciprocal concurrency” in which both partners agree to have other partners, could be less risky than “separational concurrency” in which 1 or both partners have concurrent partners when they are physically separated, or “compensatory concurrency” when 1 or both partners have other partners to compensate for perceived deficiencies in their partnership. Individuals in the first 2 types of partnerships often know that their partner has concurrent partners and might be more likely to use condoms than individuals in the latter 2 categories who are often unaware of their partner’s concurrency. Among our participants, there was poor awareness of partner concurrency, particularly in not suspecting partners of having concurrent sexual relationships. This incorrect perception of partner’s concurrency was significantly associated with STIs.

There could be different explanations for why incorrect perceptions both of a partner’s concurrency and of a partner’s nonconcurrency status could be associated with increased risk of STI. First, individuals unaware of their partner’s concurrency could be less likely to use condoms because they perceive the partnership to be committed, whereas those aware of their partners’ concurrency or who themselves have concurrent partners could be more likely to use condoms27 because their partnerships are perceived as less stable and more at risk for introduction of an STI. We found that when both members of a partnership reported concurrent partners, condom use was significantly higher than when only 1 reported concurrency. Among individuals with concurrent partners, those also aware of a partner’s concurrency reported consistent condom use more often than those who were not aware (44% vs. 28%, respectively). Secondly, those who suspected their partner had concurrent sexual relationships when the partner did not were themselves significantly more likely than those who correctly reported their partners’ nonconcurrency status to have concurrent partners (46% vs. 26%, respectively); however, report of consistent condom use was low even among the former group. Those who knew their partner had concurrent sexual relationships were also more likely to themselves have concurrent partners, but they were also more likely to report consistent condom use, and they did not have increased prevalence of a current STI. Individuals in the first scenario could represent “reactive concurrency” in which 1 member believes the other has concurrent partners and therefore has concurrent partners themselves. The second scenario when both partners are concurrent could represent “experimental” or “reciprocal” concurrency. “Reactive” concurrency, when there is no agreement between partners, could be more risky than the open forms of concurrency such as “experimental” or “reciprocal” because individuals in the latter tend to practice condom use.

If an individual’s concurrency and a partner’s concurrency increased the probability of STI additively, we would expect those partnerships in which both members had concurrent sexual relationships to have a higher proportion of STIs than those in which only 1 member practiced concurrency. We did not observe this, although the total number of STIs identified was small. Finally, in partnerships in which both partners practiced concurrency, individuals were more aware of their partner’s concurrency than in those partnerships in which only 1 member had concurrent sexual relationships, consistent with the hypothesis that more open types of concurrency could allow individuals a greater ability to protect themselves.

Another factor significantly associated with STI in this study was residing in south San Diego. This area is composed of low-income communities with large migrant populations, dynamics that could contribute to higher individual risks. STI and partnership patterns of risk have been previously shown to be associated with geographic region.28

This study could be limited by reporting biases found in studies of socially sensitive behaviors and by its relatively small sample size, with small numbers of STI. Reporting bias could have been minimized through computer-based interviewing, which increases reporting of sensitive behaviors.29–34 Furthermore, even with our sample size, the magnitudes of the associations were substantial. In larger studies, condom use by type of concurrency such as reciprocal versus reactive as well as who is practicing it (individual, partner, or both) could be examined in more detail. Studies that include both members of partnerships can aid understanding of the different influences of individuals’ and partners’ concurrency on STI acquisition and transmission.


1. Adimora AA, Schoenbach VJ, Bonas DM, et al. Concurrent sexual partnerships among women in the United States. Epidemiology 2002; 13:320–327.
2. Ford K SW, Lepkowski J. American adolescents: Sexual mixing patterns, bridge partners, and concurrency. Sex Transm Dis 2002; 29:13–19.
3. Rosenberg MD GJ, Gurvey JE, Adler N, et al. Concurrent sex partners and risk for sexually transmitted diseases among adolescents. Sex Transm Dis 1999; 26:208–212.
4. Kraut-Becher JR, Aral SO. Gap length: An important factor in sexually transmitted disease transmission. Sex Transm Dis 2003; 30:221–225.
5. Adimora AA, Schoenbach VJ, Martinson FEA, et al. Concurrent partnerships among rural African Americans with recently reported heterosexually transmitted HIV infection. J Acquir Immune Defic Syndr 2003; 34:423–429.
6. 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.
7. Castor D, Jolly P, Furlonge C, et al. Determinants of gonorrhea infection among STI clinic attenders in Trinidad—II: Sexual behavioural factors. Int J STD AIDS 2002; 13:46–51.
8. Koumans EH FT, Gibson JJ, Langley C, et al. Characteristics of persons with syphilis in areas of persisting syphilis in the United States: Sustained transmission associated with concurrent partnerships. Sex Transm Dis 2001; 28:497–503.
9. Manhart L, Aral S, Holmes K, et al. Sex partner concurrency: Measurement, prevalence and correlates among urban 18–39 year olds. Sex Transm Dis 2002; 29:133–143.
10. Stoner BP, Whittington WLH, Aral SO, et al. Avoiding risky sex partners: Perception of partners’ risks vs partners’ self reported risks. Sex Transm Infect 2003; 79:197–201.
11. Ellen JM, Vittinghoff E, Bolan G, et al. Individuals’ perceptions about their sex partners’ risk behavior. J Sex Research 1998; 35:328–332.
12. Ferguson N, Garnett G. More realistic models of sexually transmitted disease dynamics: Sexual partnership networks, pair models and moment closure. Sex Transm Dis 2000; 27:600–608.
13. Morris M, Kretzschmar M. Concurrent partnerships and the spread of HIV. AIDS 1997; 11:641–648.
14. Wats C, May R. The influence of concurrent partnerships on the dynamics of HIV/AIDS. Math Biosci 1992; 108:89–104.
15. Chick S, Adams A, Koopman J. Analysis and simulation of a stochastic, discrete-individual model of STI transmission with partnership concurrency. Math Biosci 2000; 166:45–68.
16. Kretzschmar M, Morris M. Measures of concurrency in networks and the spread of infectious disease. Math Biosci 1996; 133:165–195.
17. Ghani A, Garnett G. Risks of acquiring and transmitting sexually transmitted diseases in sexual partner networks. Sex Transm Dis 2000; 27:579–587.
18. Armitage P, Colton T, eds. Encyclopedia of Biostatistics. New York: Wiley, 1998.
19. Fleiss J. Statistical Methods for Rates and Proportions. New York: Wiley, 1981.
20. Padian N. Sexual histories of heterosexual couples with one HIV-infected partner. Am J Public Health 1990; 80:990–991.
21. Upchurch D, Weisman C, Shepherd M, et al. Interpartner reliability of reporting of recent sexual behaviors. Am J Epidemiol 1991; 134:1159–1165.
22. Ellish N, Weisman C, Celentano D, et al. Reliability of partner reports of sexual history in a heterosexual population at a sexually transmitted diseases clinic. Sex Transm Dis 1996; 23:446–452.
23. Seage GR, Mayer K, Horsburgh CJ, et al. Corroboration of sexual histories among male homosexual couples. Am J Epidemiol 1992; 135:79–85.
24. Coates RA, Calzavara LM, Soskolne C, et al. Validity of sexual histories in a prospective study of male sexual contacts of men with AIDS or an AIDS-related condition. Am J Epidemiol 1988; 128:719–728.
25. Niccolai L, Farley T, Ayoub M, et al. HIV-infected persons’ knowledge of their sexual partners’ HIV status. AIDS Educ Prev 2002; 14:183–189.
26. Gorbach P, Stoner B, Aral S, et al. It takes a village: Understanding concurrent partnerships in Seattle, WA. Sex Transm Dis 2002; 29:453–462.
27. Macaluso M, Demand M, Artz L, et al. Partner type and condom use. AIDS 2000; 14:537–546.
28. Jennings JM, Ellen JM. Does sex partner concurrency and geographic context affect STI risk for female adolescents and their main sex partners? J Adolesc Health 2003; 32:127–128.
29. Boekeloo BO, Schiavo L, Rabin DL. Self-reports of HIV risk factors by patients at a sexually transmitted disease clinic: Audio vs. written questionnaires. Am J Public Health 1994; 84:754–760.
30. Murphy DA, Durako S, Muenz LR, et al. Marijuana use among HIV-positive and high-risk adolescents: A comparison of self-report through audio computer-assisted self-administered interviewing and urinalysis. Am J Epidemiol 2000; 152:805–813.
31. Des Jarlais DC, Paone D, Milliken J, et al. Audio-computer interviewing to measure risk behavior for HIV among injecting drug users: A quasi-randomised trial. Lancet 1999; 353:1657–1661.
32. Kissinger P, Rice J, Farley T, et al. Application of computer-assisted interviews to sexual behavior research. Am J Epidemiol 1999; 149:950–954.
33. Metzger DS, Koblin B, Turner C, et al. Randomized controlled trial of audio computer-assisted self-interviewing: Utility and acceptability in longitudinal studies. Am J Epidemiol 2000; 152:99–106.
34. Turner CF, Ku L, Rogers SM, et al. Adolescent sexual behavior, drug use, and violence: Increased reporting with computer survey technology. Science 1998; 280:867–873.
© Copyright 2004 American Sexually Transmitted Diseases Association