KRAUT-BECHER, JULIE R. PhD*; ARAL, SEVGI O. PhD†
From *Northern Illinois University, DeKalb, Illinois; and †Centers for Disease Control and Prevention, Atlanta, Georgia
The authors thank Charles Akers, PhD, Stuart Berman, MD, Sherilynn Spear, PhD, and Maya Sternberg, PhD, for their helpful comments and suggestions and Patricia Jackson for her outstanding support in the preparation of the manuscript.
Reprint requests: Julie R. Kraut-Becher, PhD, School of Allied Health Professions, DeKalb, Illinois 60115. E-mail: firstname.lastname@example.org
Received May 14, 2002,
revised August 12, 2002, and accepted August 20, 2002.
RECENT WORK in the behavioral epidemiology of sexually transmitted diseases (STDs) and the population transmission dynamics of STDs has highlighted the importance of sexual mixing patterns, sexual networks, and timing of sexual partnerships in the spread of STDs. 1–16 The sequential order among sex partners and the duration of time between sex partnerships may affect the rate of spread of STDs in populations and the risk of acquiring and transmitting STDs for particular individuals. 17–19
The average duration of time between consecutive partners of serially monogamous individuals and the prevalence of nonmonogamous (concurrent) partnerships are two distinct parameters that may influence the incidence and prevalence of STDs within a population. Theoretical models show that the proportion of the population in mutually nonmonogamous partnerships is an important parameter describing the establishment of infection. 17 Once infection is established within a population, the sexual mixing pattern of the group of highly sexually active persons predicts the prevalence of infection. 17 Empirical studies have shown substantial prevalence of concurrency in high-risk groups and in local and national populations. 1,5,20–21 Results from a nationally representative sample of women of childbearing age in the United States show that concurrency was associated with age, age at first intercourse, and marital status. These three factors are commonly found to be significant predictors of STD diagnoses. 1
The majority of individuals do not report concurrent partnerships; STDs spread across populations even in the absence of this type of partnership. 1,13 The duration of time serially monogamous individuals spend between partnerships may be an important factor that affects the rate at which sexually transmitted infections spread through populations. In this article we refer to this factor as the gap. The gap may have important implications for the spread of STDs because the mean duration of time members of a population spend between partners, together with the timeliness of diagnosis and treatment, may affect the probability of exposure between infected and uninfected individuals. The duration of infectivity varies across sexually transmitted infections. For example, the mean duration of infectivity has been estimated to be as long as approximately 15 months (456 days) for Chlamydia trachomatis and as short as almost 2 months for Neisseria gonorrhoeae if prevention and treatment programs are in place. 22 Thus, individuals who have gaps shorter than 15 months may, if infected with chlamydia in the context of an earlier relationship, infect their later partner with chlamydia in the absence of diagnosis and treatment. Therefore, different gaps between partnerships may fuel the spread of different STDs.
Although several empirical studies have demonstrated the relationship between concurrency and heightened risk of STDs, we are not aware of any empirical examinations focusing on the relationship between duration of time serially monogamous individuals spend between partnerships and STD risk. In this article we focus on gaps reported by a nationally representative sample of women. Gap is defined as the duration of time spent between first sex with the current or most recent partner and last sex with the previous partner. We measured the frequency of short and long gaps and examined differential patterns of gap length by demographic and socioeconomic characteristics and history of STD. We also computed the percentage of serially monogamous women who reported gaps that may be considered risky in terms of STD transmission.
We used data from the 1995 wave of the National Survey of Family Growth (NSFG) to examine temporal dynamics of women's sex partnerships. 23 The NSFG is a cross-sectional survey of a representative sample of civilian, noninstitutionalized women 15 to 44 years of age in the United States. The primary focus of the survey is women's reproductive health and health-related behaviors. The data set includes detailed information on women's sex partners and STD history as well as demographic information. The 1995 NSFG sample consisted of 10,847 women. All data were collected during face-to-face interviews, with the exception of highly sensitive information such as STD diagnoses, which were entered into a personal computer directly by the respondents.
The NSFG includes detailed timing and demographic information on a woman's first sex partner, current and former husbands, current and former cohabiting partners, and any other new sex partner(s) during the 5 years before interview. Therefore, specific information about sex partnerships other than spouses and cohabiting partners before 1991 is unknown. Thus, measures of all sex partnerships for each respondent would be closest to lifetime measures for only the youngest women.
Our analyses were restricted to sexually active women who report having had at least two sex partners (N = 7450). Given that information about sex partnerships other than spouses and cohabiting partners was restricted to only the 5 years before interview and that information on the continuity of those relationships was limited, we considered only two partnerships per woman. Women for whom information about date of first sex with their current/most recent partner or date of last sex with their previous partner was missing were excluded (N = 4682), resulting in a total sample of 2768 women. All of the data were weighted to make the sample nationally representative. Data management was conducted with SAS 8.0 software (SAS Institute, Cary, NC), and SUDAAN software (Research Triangle Institute, Research Triangle Park, NC) was used for all analyses to ensure that standard errors were adjusted for the complex survey design, comprising the 1995 NSFG.
We defined the gap to be the difference between the date of first sex with a current or most recent partner and the date of last sex with a previous partner. Date information in the NSFG is recorded in century months (i.e., month and year since 1900; for example, January 1900 is coded 1, whereas June 1995 is coded 1146). Negative gaps mean women have had overlapping, or concurrent, sex partnerships. Gaps of zero indicate that women have ended one sex partnership and started another within the same month, referred to here as “instantaneous partner change.” Positive gaps signify serially monogamous individuals (i.e., women who had nonoverlapping partnerships).
First, we examined the proportion of women reporting concurrency, serial monogamy, or instantaneous partner change by demographic and socioeconomic characteristics and self-reports of STD diagnoses. Chi-square tests were used to determine associations between the gap and various demographic and socioeconomic characteristics and STD history. Second, we examined the mean gap among serially monogamous individuals by subpopulation. To test for statistical significance of differences between subpopulations, we used a t test for difference in means. Finally, we measured the percentage of serially monogamous women who reported gaps that were sufficiently shorter than the mean duration of infectivity of some bacterial STDs. Relatively short gaps may allow the transmission of STDs. We examined the distribution of “risky” gaps by demographic characteristics and STD history.
Approximately 70% of women who had had at least two sex partnerships during their lifetime reported serially monogamous partnerships, and about one quarter reported overlapping or concurrent partnerships (Table 1). The proportion who reported instantaneous partner change was very low for all subpopulations. The proportion of women who reported concurrent partnerships increased slightly with age and decreased with income.
Women with a history of STD diagnoses were more likely to report concurrent partnerships than women without a history of STD (35% versus 24%;P < 0.01). Most women report nonoverlapping partnerships regardless of their race/ethnicity. However, non-Hispanic blacks were more likely to report concurrent partnerships than whites and others (35% versus 23% and 25%, respectively;P < 0.01). Women with higher levels of education and family income were less likely to report concurrency and more likely to report partnerships that did not overlap in time than were women with lower levels of education and lower income. Reports of concurrent partnerships did not vary by geographic area of the country or by rural/urban status but did vary by residence in Standard Metropolitan Statistical Areas (SMSAs). Women living in SMSAs reported concurrency more and serial monogamy less than women who did not live in SMSAs (P = 0.05).
The mean gap for serially monogamous women was significantly shorter among adolescents than among young adults and older women (Table 2). Women with a history of STD diagnoses reported significantly shorter gaps than women without a history of STD (9 and 14 months, respectively). Reported gaps were significantly shorter among non-Hispanic white women than among non-Hispanic black and other women. The mean gap was not associated with education, income, or residence, whether measured by urban/rural, SMSA/non-SMSA, or geographic region of the country.
A large proportion of serially monogamous women reported gaps that were sufficiently shorter than the mean duration of infectivity of some bacterial STDs (Table 3). Approximately half of the women reported gaps that were sufficiently short to allow transmission of syphilis and gonorrhea, and more than half of the women reported gaps that would allow transmission of chlamydia and chancroid (Table 3).
Compared with other women, a higher percentage of adolescents reported gaps that were shorter than the mean duration of infectivity for gonorrhea and syphilis. This was the case even in the presence of STD prevention and treatment programs (Figure 1). Gaps that were sufficiently short to allow transmission of several bacterial STDs, even in the presence of STD prevention and treatment programs, were reported more frequently by non-Hispanic black women than by non-Hispanic white women and women of other race or ethnicity groups (Figure 2). Compared with women with no STD history, women who reported a history of bacterial or viral STD were more likely to report time intervals between partnerships that were short enough to allow the transmission of gonorrhea and syphilis (Figure 3).
We examined data from a nationally representative sample of women of reproductive age in the United States to measure the length of time women spend between partnerships. More than two thirds of women with at least two sex partners during their lifetime reported positive gaps, indicating that their most recent partnership did not overlap in time with their previous partnership. Although most women reported serially monogamous partnerships, slightly more than one quarter of women reported negative gaps, indicating concurrent or simultaneous partnerships.
Among serially monogamous women, the shorter the duration between consecutive partners, the higher is the risk of STD transmission. We found that a high proportion of all serially monogamous women switch partners in time periods shorter than the mean infectivity periods of many bacterial STDs, thus increasing their risks of STD transmission.
High incidence of STDs among adolescents is often attributed to high-risk sexual practices and sexual mixing patterns associated with this population. 3,4,20–21 In particular, having multiple partners, especially at a specific point in time (concurrency), is frequently recognized as a risk factor for STDs in this group. 5,12 Our findings suggest that short gaps rather than concurrent partnerships contribute to the risk of STDs among young people. We found that adolescent women reported concurrency less frequently than young adults and older women. Although teenage women reported the greatest proportion of instantaneous partner changes, overall the reported proportions were small. Without information on specific dates of sexual intercourse, it is difficult to determine the proportion of reported instantaneous partner changes that indicated concurrent partnerships. Moreover, adolescent women reported the shortest mean gap of any subgroup, even shorter than that of women who reported a history of STD (8 months versus 9 months) (Table 2).
There are several limitations of our analyses. First, because of the limited sexual history information in the NSFG, we considered only two partnerships per woman. Women who report short gaps between their current/most recent partner and the previous partner may or may not represent people who consistently have short gaps between their sex partnerships. It is unknown whether they are reporting a once-in-a-lifetime experience or a pattern of behavior. Women who start and end partnerships often in relatively short periods of time may be at higher risk of STDs than women who rarely experience quick transitions between relationships. Second, we measured the gap by using retrospective data on dates of sexual partnerships. Individuals asked to recall past events may inadvertently or intentionally forget or confuse the details, thereby introducing bias to the data. Third, the STD diagnosis information in the NSFG is not temporally defined. Therefore, we cannot associate a STD diagnosis with a particular partnership. We examined proportions of concurrency, instantaneous partner change, and serial monogamy by demographic characteristics and STD incidence or prevalence. These results are descriptive; additional data would be needed to determine whether the associations we report are causal.
Despite the limitations, our data are consistent with those of other studies. Recent studies focused on measuring prevalence of concurrency have found that the majority of women report serially monogamous partnerships, but a substantial, nonnegligible proportion of women report concurrent partnerships. 1,5,13 For example, Manhart et al 5 found that 22% of men and women aged 18 to 39 years in a local urban population reported that they have had concurrent partnerships and 28% reported that they or their partners had concurrent partnerships. In a recent analysis of the 1995 NSFG data, Adimora et al 1 found a 12% overall prevalence of concurrency by dividing the weighted number of respondents with overlapping sexual partnership dates by the weighted total number of respondents.
We have reported gaps between current/most recent partner and previous partner for a nationally representative sample of women. Future research should also measure gaps for men. Risk of STD depends not only on an individual's own risk factors but also on the risk factors of their partner(s). 9 The degree of an individual's protective precautions may differ depending on whether the individual is knowingly involved in concurrent partnerships or unknowingly involved through his/her partner's unsuspected concurrency. 6 A partner's gap may also be relevant in determining risk of STD for an individual. We were not able to assess the effect of partners’ risk factors because information is not available in the NSFG for other partners of women's partners.
STD prevention messages typically focus on reducing the number of partners and encouraging serial monogamy. However, sexual network studies have shown that the timing of sexual partnerships may be a more important risk factor than the number of partners. We have found that adolescents and women with a history of STD, groups typically found to be at high risk of acquiring or reacquiring STDs, tend to report intervals of time between partnerships, gaps, shorter than the mean duration of infectivity of some bacterial STDs. Given that many STDs are often asymptomatic, short gaps may present a problem if women and their partners are not routinely screened for STDs. It is imperative that clinicians increase their efforts to take sexual histories and screen for infections that would otherwise go undetected.
1. Adimora AA, Schoenbach VJ, Bonas DM, Martinson FEA, Donaldson KH, Stancil TR. Concurrent sexual partnerships among women in the United States. Epidemiology 2002; 13: 320–327.
2. Aral SO, Blanchard JF. Phase-specific approaches to the epidemiology and prevention of sexually transmitted disease: an overview. Sex Transm Infect 2002; 78 (Suppl 1): i1–i2.
3. Ford K, Sohn W, Lepkowski J. American adolescents: sexual mixing patterns, bridge partners, and concurrency. Sex Transm Dis 2002; 29: 13–19.
4. DiClemente RJ, Wingood GM, Crosby RA, et al. Sexual risk behaviors associated with having older sex partners: a study of black adolescent females. Sex Transm Dis 2002; 29: 20–24.
5. Manhart LE, Aral SO, Holmes KK, Foxman B. Sex partner concurrency: measurement, prevalence and correlates among urban 18–39 year olds. Sex Transm Dis 2002; 29: 133–143.
6. Friedman SR, Aral S. Social networks, risk-potential networks, health, and disease. J Urban Health 2001; 78: 411–418.
7. Gorbach PM, Sopheab H, Phalla T, et al. Sexual bridging by Cambodian men: potential importance for general population spread of STD/HIV epidemics. Sex Transm Dis 2000; 27: 320–326.
8. Aral SO, Hughes JP, Stoner B, et al. Sexual mixing patterns in the spread of gonococcal and chlamydial infections. Am J Public Health 1999; 89: 825–833.
9. Laumann EO, Youm Y. Racial/ethnic group differences in the prevalence of sexually transmitted diseases in the United States: a network explanation. Sex Transm Dis 1999; 26: 250–261.
10. Potterat JJ, Zimmerman-Rogers H, Muth SQ, et al. Chlamydia transmission: concurrency, reproduction number, and the epidemic trajectory. Am J Epidemiol 1999; 150: 1331–1339.
11. Potterat JJ, Rothenberg RB, Muth SQ. Network structural dynamics and infectious disease propagation. Int J STD AIDS 1999; 10: 182–185.
12. Rosenberg MD, Gurvey JE, Adler N, Dunlop MBV, Ellen JM. Concurrent sex partners and risk for sexually transmitted diseases among adolescents. Sex Transm Dis 1999; 26: 208–212.
13. Aral SO. Patterns of sex partner recruitment and types of mixing as determinants of STD transmission: limits to the spread of sexually transmitted infections. Venereology 1995; 8: 240–242.
14. Morris M, Kretzschmar M. Concurrent partnerships and transmission dynamics in networks. Soc Networks 1995; 17: 299–318.
15. Watts CH, May RM. The influence of concurrent partnerships on the dynamics of HIV/AIDS. Math Biosci 1992; 108: 89–104.
16. Anderson RM, May RM. Epidemiological parameters of HIV transmission. Nature 1988; 333: 514–519.
17. Ghani AC, Swinton J, Garnett GP. The role of sexual partnership networks in the epidemiology of gonorrhea. Sex Transm Dis 1997; 24: 45–56.
18. Morris M, Kretzschmar M. Concurrent partnerships and the spread of HIV. AIDS 1997; 11: 641–648.
19. Garnett GP, Johnson AM. Coining a new term in epidemiology: concurrency and HIV. AIDS 1997; 11: 681–683.
20. Howard MM, Fortenberry JD, Blythe MJ, Zimet GD, Orr DP. Patterns of sexual partnerships among adolescent females. J Adolesc Health 1999; 24: 300–303.
21. Lindblade K, Foxman B, Koopman JS. Heterosexual partnership characteristics of university women. Int J STD AIDS 1994; 5: 37–40.
22. Brunham RC, Plummer FA. A general model of sexually transmitted epidemiology and its implications for control. Med Clin North Am 1990; 74: 1339–1352.
23. Abma J, Chandra A, Mosher W, Peterson L, Piccinino L. Fertility, family planning, and women's health: new data from the 1995 National Survey of Family Growth. National Center for Health Statistics. Vital Health Stat 1997; 23: 19.