*Department of Epidemiology and Biostatistics, The George Washington University, Washington, DC
Departments of †Sociology
‡Statistics, University of Washington, Seattle, WA
§Department of Epidemiology, University of Michigan, Ann Arbor, MI
‖U.S. Centers for Disease Control and Prevention, Atlanta, GA
Departments of ¶Epidemiology
**Medicine, University of Washington, Seattle, WA
††HIV/STD Program, Public Health—Seattle & King County, Seattle, WA.
Correspondence to: Sara N. Glick, PhD, MPH, Department of Epidemiology and Biostatistics, The George Washington University, 2100-W Pennsylvania Avenue, NW, 8th Floor, Washington, DC 20037 (e-mail: firstname.lastname@example.org).
The authors have no conflicts of interest to disclose.
Supported by University of Washington STD/AIDS Research Training Program (T32 AI007140 to S. N. Glick) from the National Institutes of Health, U.S. Public Health Service. The Seattle Sex Survey RDD was funded by Public Health—Seattle & King County, the Center for Molecular and Clinical Epidemiology of Infectious Diseases, University of Michigan School of Public Health, and the University of Washington Center for AIDS and STD. The 2003 Seattle MSM RDD was supported by a Comprehensive STD Prevention System Syphilis Elimination grant from the Centers for Disease Control and Prevention and the 2006 RDD was supported by a grant to M.R. Golden (NIH K23 AI01846).
Portions of these data were presented at the 17th Conference on Retroviruses and Opportunistic Infections (February 2010) in San Francisco, CA, and at the 4th Annual National Graduate Student Research Festival (November 2009) in Bethesda, MD.
S. N. Glick and M. R. Golden, conceptualized the present study. S. N. Glick conducted the analyses and wrote the first draft of the article. B. Foxman, S. O. Aral, L. E. Manhart, and K. K. Holmes designed the Seattle Sex Survey, while M. R. Golden designed the Seattle MSM surveys. All authors contributed to the present study design, reviewed, and edited the article. All authors have read and approved the final version of this article.
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Web site (www.jaids.com).
Received September 21, 2011
Accepted December 20, 2011
Sexually transmitted infections (STI) disproportionately affect men who have sex with men (MSM). MSM comprise approximately 2% of the US population,1 but accounted for 59% of new HIV infections and 62% of cases of early syphilis in 2009.2,3 The Centers for Disease Control and Prevention estimates that HIV and early syphilis rates among MSM are >40 times higher than those among heterosexuals.4 In part, these differences reflect the fact that an individual MSM can engage in both insertive and receptive sexual roles (ie, versatility), whereas exclusively heterosexual men and women each engage in only 1 of these roles. Goodreau et al5 demonstrated that an MSM population with a very high level of versatility would have a higher HIV prevalence than one with less versatility. Furthermore, the transmission probability of HIV associated with anal sex is higher than that associated with vaginal sex.6,7 These factors alone would result in significant disparities in HIV rates between MSM and heterosexuals even if both populations had similar numbers of sex partners, frequency of sex, and condom use levels.8
However, the sexual behaviors of MSM and of male and female heterosexuals are substantially different in ways that are not explained by biology alone. Prior research has found that MSM tend to have higher numbers of sex partners than heterosexuals,9 but the dynamics of partnership formation,10 concurrency,11 and age mixing patterns12–14 have not been extensively characterized. These factors probably play important roles in the epidemiology of HIV/STI and in explaining observed disparities. Although some of these behaviors have been assessed to better understand observed racial disparities in HIV/STI rates,15–18 much less has been done to compare the behaviors of MSM and heterosexuals.9,19 In this article, we used population-based studies to compare the sexual behaviors of MSM and heterosexual men and women in the United States, paying particular attention to how patterns of behavior differed across age groups.
Study Populations, Study Designs, and Survey Instruments
We used data from 4 random digit dialing (RDD) surveys (Table 1). To calculate sexual behavior estimates for MSM, we used the Urban Men's Health Study (UMHS)20–22 and 2 Seattle surveys (SEA).23 For similar estimates among heterosexual men and women, we used data from the Seattle Sex Survey 2 (SSS).24
The SSS and the 2 SEA RDDs were conducted between 2003 and 2006, whereas the UMHS, which enrolled MSM from 4 US cities, was conducted from 1996 to 1998. For this analysis, we excluded SSS participants who had never had sex or reported partnerships that were not exclusively heterosexual. For UMHS and SEA, telephone numbers were sampled from zip codes with high proportions of MSM. Men who reported same-sex sexual behavior since 14 years were eligible. UMHS MSM were also eligible if they self-identified as homosexual, gay, or bisexual. Oral informed consent was obtained from all participants. To increase comparability across all surveys, we restricted analyses to participants aged 18–39 years.
Only SSS and SEA provided response rates, which were <50% and computed by dividing the number of interviews by an estimate of the number of eligible individuals from the RDD sampling frame. This denominator was the sum of the number of interviews, refusals, and an estimate of the number of eligible individuals among those for whom eligibility could not be determined.25 The cooperation rates were based on the proportion of eligible individuals who opted to participate and were higher for UMHS and SEA (78%–97%) than the SSS (46%). The UMHS and SEA surveys gathered partner-specific data only for partners in the past year. Thus, analyses regarding a participant's most recent partner were restricted to heterosexual participants reporting at least 1 opposite-sex partner in the past year, and to MSM reporting at least 1 male partner in the past year. For partner-specific measures among MSM, UMHS data reflected the participants' most recent male partners, whereas the SEA data included only male anal sex partners.
We chose sexual behavior measures that we hypothesized a priori might help explain differences in HIV/STI rates between MSM and heterosexual men and women. Although the exact wording of questions varied across surveys, most measures were similar. Supplemental Digital Content 1 (see Appendix, http://links.lww.com/QAI/A259) includes specific survey questions and more detailed descriptions of each measure.
Age at Sexual Debut and Number of Partners
We calculated age at sexual debut with any partner among heterosexuals and UMHS MSM. (SEA MSM did not provide these data.) Among MSM only, we also compared age at sexual debut with a male partner and the age at anal sex debut. The SSS and UMHS provided data on lifetime number of partners, although all surveys included a question about the number of partners in the past year.
Partnership Formation and the Cessation of New Partner Recruitment
We used partner-specific start and end dates (month and year) to estimate the time since the start of each participant's most recently formed partnership. Using these data, we calculated the proportion of participants who had formed any new partnership in the past year. We defined the cessation of new partner recruitment as having last formed a new partnership >5 years ago. To estimate an individual's total time of active partner seeking to date, we calculated the number of years between a participant's age at sexual debut and age at the start of the most recently formed partnership. Finally, we used the variable regarding number of partners in the past year to indicate if an individual had engaged in any sex in the past year.
Mixing by Age
We used age data about each participant's most recent partner to calculate the proportion who had partners with greater than a 5- or 10-year age difference.
Across all surveys, we assessed whether or not participants always used condoms with their most recent partner. This was based upon partner-specific questions about how often condoms were used, or by comparing the total number of sex acts with the number of protected sex acts. As a proxy for partner type, we stratified these findings by partnership duration (≤3 or >3 months).
Sex Partner Concurrency
Using available partnership start and end dates (month and year), we determined the proportion of respondents who reported any overlapping partnerships in the past year. We classified participants with no partners in the past year as having no partner concurrency.
Partner Meeting Place
Using a question about how respondents met their most recent partner, we created 3 categories26: through formal social venues (eg, friends, family, work, and private party), through less formal social venues (eg, bar, Internet, bath house, park), or by other means (ie, participant chose the “other” option).
We compared each measure among heterosexual men and women and MSM using descriptive statistics. As appropriate, we used t tests, χ2 tests, or Wilcoxon rank sum tests to assess for group differences, and linear and logistic regression to test for age trends. Although UMHS did include sampling weights, the SSS and SEA did not. Because we found no meaningful differences between the weighted and unweighted estimates in our analysis of UMHS data, we elected to apply a uniform analytic approach across data sets, and all presented estimates are unweighted. Therefore, in the context of these study designs, P values are best understood as a metric for comparing the strength of the observed associations within each study. We conducted all analyses using Intercooled Stata 12 (Stata Corporation, College Station, TX). Study procedures were either exempt from the review of institutional review board (UMHS and SEA) or approved (SSS) by the Human Subjects Division, University of Washington.
The mean age of participants included in this analysis was similar across surveys, ranging from 30.0–32.1 years. In all surveys, approximately three quarters of participants were non-Hispanic white and two thirds had obtained a 4-year college degree. Among MSM, 15.7% of UMHS participants and 10.1% of SEA participants reported being HIV infected; no heterosexual participants reported HIV infection.
Age at Sexual Debut and Number of Partners
The mean age at same-sex sexual debut was 16.5 years among UMHS MSM and 17.0 years among SEA MSM (Table 2). Approximately one quarter of UMHS MSM first had sex with a woman; when this was considered, the overall age at sexual debut among UMHS MSM was 15.4 years. Heterosexual debut was older at 17.4 years for men and 17.8 years for women. Anal sex debut among MSM was several years later than overall age of sexual debut, at 19.6 years in UMHS and 20.2 years in SEA.
At all ages, heterosexual men and women reported a median of 1 partner in the past year, whereas MSM reported a median of 4 partners in UMHS and 2–3 male partners in SEA. MSM reported significantly more lifetime partners than heterosexual men and women at all ages (P < 0.01 for each age group). The median lifetime number of sex partners among those aged 18–24 years was 4 in heterosexuals and 15 in UMHS MSM, and among persons aged 35–39 years, was 10 and 67, respectively. (The SEA did not ask about lifetime number of partners.)
Partnership Formation and the Cessation of New Partner Recruitment
Across all age groups, MSM were more likely than heterosexuals to report having a new partner in the past year (Fig. 1A). Among UMHS MSM, 86.4% of the youngest MSM (aged 18–24 years) and 71.7% of the oldest MSM (aged 35–39 years) formed a new same-sex partnership in the previous year. These proportions were lower among SEA MSM—in part, because of the fact that only male anal sex partners were counted—with 78.1% of the youngest MSM and 41.7% of the oldest MSM reporting a new male anal sex partnership in the past year. Among heterosexuals, the proportions were lower than in any of the MSM surveys, particularly among participants in their 30s. For example, 55.9% of the youngest men and 20.8% of the oldest men formed a new partnership in the previous year compared with 34.1% of the youngest women and 9.6% of the oldest women.
These trends are clearly seen when viewed over a 5-year period: a much higher proportion of heterosexuals than MSM reported that their last new partnership began >5 years ago (Figure 1b). For example, among participants aged 35–39 years, 50.0% of heterosexual men and 68.2% of heterosexual women reported that at least 5 years had passed since the start of their most recent partnership. Only 11.2% of UMHS MSM and 14.2% of SEA MSM in the same age range reported this. Consequently, MSM in each age group reported longer periods of new partner acquisition than heterosexuals, defined as the amount of time from sexual debut to the start of the most recently formed partnership. Because the age-specific means for heterosexual men and women, and MSM in both surveys, were so similar (within 1–3 years), we combined the data for these respective groups. Among participants aged 35–39 years, MSM reported an average of 20.2 years of new partner acquisition compared with 11.9 years among heterosexuals. This difference between heterosexuals and MSM increased linearly with age (P < 0.01 for interaction).
At the same time, the proportion of individuals who reported no sexual activity may be higher among older MSM than among their heterosexual counterparts, although these findings were not consistent across surveys. Unlike UMHS MSM and heterosexuals, the proportion of SEA MSM who reported no male partner in the past year increased linearly by age, including nearly one quarter (23.0%) of those aged 35–39 years (P < 0.01 for trend). This proportion was approximately 4 times higher than that observed among Seattle heterosexual men (4.8%) and women (6.7%) in the same age group (SEA MSM vs. heterosexual men and women, P < 0.01).
Mixing by Age
To accurately capture partnerships formed at a given age, we restricted these analyses to participants reporting a new partnership in the past year. Among all 18- to 24-year-olds, there were no recent partners who were >5 years younger. More than one half (52.0%) of SEA MSM aged 18–24 years reported a recent male anal sex partner who was >5 years older, whereas 42.5% of UMHS MSM reported this age difference with a recent male partner (Fig. 2A). By contrast, only 7.9% of heterosexual men and 10.0% of heterosexual women in this age group reported a recent partner who was >5 years older (MSM vs. heterosexual men and women, P < 0.01). No heterosexuals aged 18–24 years reported a recent partner with a 10-year age difference, whereas 16.7% of UMHS MSM and 28.0% of SEA MSM reported this (MSM vs. heterosexual men and women, P < 0.001). At older ages, the difference in the proportion of MSM and heterosexuals reporting a 5-year age difference with their most recent partner was less apparent (Table 2), although MSM aged 35–39 years were more likely than heterosexuals to report a 5-year age difference with their most recent partner (P = 0.018).
Among recent partnerships lasting ≤3 months, approximately 80% of MSM in UMHS reported always using condoms during insertive anal sex. A similar proportion reported this during receptive anal sex. By contrast, only 57.4% of heterosexual men and 44.4% of heterosexual women reported consistent condom use during vaginal sex in partnerships of ≤3 months. For partnerships lasting >3 months, consistent condom use was reported by 48.0% and 46.8% of MSM during insertive and receptive anal sex, respectively, and by less than one quarter of heterosexual men (23.8%) and women (19.7%). [Because of small cell sizes, we could not estimate position-specific (insertive and receptive) condom use among SEA MSM.]
As shown in Figure 2B, the prevalence of sex partner concurrency during the year was much lower among heterosexual men and women (9.7% and 7.5%, respectively) than among UMHS MSM (31.3%) and SEA MSM (17.8%) (MSM vs. heterosexual men and women, P < 0.01). The prevalence of partner concurrency declined linearly with age among heterosexual men (P = 0.03) and women (P = 0.01). The prevalence of concurrency increased with age among MSM, although this trend was not statistically significant in either UMHS (P = 0.11) or SEA (P = 0.14; Table 2).
Meeting New Partners
More than three quarters of heterosexual men (75.3%) and women (76.5%) met their most recent partner through a formal social venue (eg, friends, family, work, and school) compared with 32.9% of UMHS MSM and 30.3% of SEA RDD MSM (P < 0.001). MSM more often reported meeting partners through less formal social venues (eg, bars, Internet, and street). Among women and UMHS MSM, the proportion who met their most recent partner through formal social venues was lower among older age groups than younger age groups.
We used data from several large population-based surveys to estimate the dynamics of sexual partnership formation, concurrency, and age mixing among MSM and heterosexuals. MSM initiated sexual activity at slightly younger ages than heterosexuals, reported larger numbers of recent partners, continued to form new partnerships later into adulthood, and displayed more age disassortative mixing and sex partner concurrency. Along with biological factors, these sexual behavior patterns likely help explain the high HIV/STI rates among MSM. However, compared with heterosexuals, MSM used condoms more frequently and a larger proportion of MSM may become sexually abstinent in their 30s.
Our findings are most notable for demonstrating how partnership formation patterns differed between MSM and heterosexuals from time of sexual debut through their 30s. Approximately half of sexually active heterosexuals surveyed were effectively out of the risk pool for HIV/STI by 30 years because they reported no new partners during the previous 5 years. By contrast, a much higher proportion of MSM continued to form new partnerships well into their 30s. This is consistent with the Australian HIV data that suggested that the formation rate among MSM peaks in the mid-30s.27 Our study suggests that the average period of new partnership acquisition among MSM is at least twice as long as among heterosexuals. Indeed, mathematical models estimated that the mean age of HIV acquisition among Australian MSM is in the mid-30s,27 and researchers in the UK found that HIV incidence was highest among MSM in this age group,28 an age when most heterosexuals are no longer forming new partnerships.
The epidemiologic implications of sustained new partnership formation among MSM are probably magnified by the relatively high frequencies of age disassortative mixing observed among MSM compared with heterosexuals. For example, because HIV prevalence rises with age, younger MSM who have older partners are at increased risk of exposure to HIV infection.12,29,30
Mathematical models suggest that sex partner concurrency amplifies the spread of HIV through sexual networks.11 We found that the prevalence of concurrent partnerships in the past year was several fold higher among MSM than heterosexuals, and that, unlike among heterosexuals, the prevalence of concurrency did not decline with age. Data from other population-based studies have consistently found a higher prevalence of concurrency among heterosexual men than women.31–33 We also observed some gender asymmetry in the prevalence of concurrency among heterosexuals—slightly higher rates of concurrency for heterosexual men than women—which leads to small fragmented components (eg, 1 man with 2 female partners) and reduces the impact that concurrency has on the connectivity of the network.34 Among MSM practicing both insertive and receptive anal sex, however, concurrency may be more likely to be symmetric, which could lead to the rapid growth of larger connected network components and more efficient HIV/STI transmission within the network.
MSM overall reported both higher numbers of partners and more concurrency than heterosexuals. In a descriptive study, such as the present one, it is not possible to tease apart the relative impact of these two factors. Both likely contribute to greater transmission. It is possible that the high rates of partner acquisition among MSM are sufficient to produce and sustain the observed disparities in HIV/STI prevalence. But, it is also possible that concurrency produces a qualitative difference in transmission dynamics even in this context—creating more robust connectivity in the network, compounded by the window of high infectivity during acute infection. This qualitative effect has been shown in 2 recent modeling studies of heterosexual spread of HIV in Zimbabwe,35,36 and a similar modeling study would be needed here to identify the independent and joint impacts of concurrency and rapid partner acquisition.
This study had several limitations. First, we used 3 different surveys that limited the comparability of measures between groups. To our knowledge, however, no single survey includes large numbers of MSM and heterosexuals from rigorously sampled representative populations as well as the parameters we sought to study. The Seattle-based heterosexual and MSM RDDs were all conducted between 2003 and 2006, and our findings using the SEA survey were similar to what Levin et al9 reported regarding age at first sex and condom use using the small number of MSM participating in the SSS (n = 72). The UMHS, however, was conducted in 1996–1998, which was 5–10 years earlier than the Seattle surveys. Since the mid-1990s, antiretroviral therapy for HIV was introduced, and some evidence suggests that serosorting and HIV testing frequency increased among MSM.37,38 Despite these changes, we did not observe large consistent differences in the sexual behavior patterns we evaluated in MSM between the UMHS and SEA. For example, although there were drastic changes in Internet use in the decade following the mid-1990s, the proportion of MSM who met their partners through less formal social venues in general did not change. Another limitation of using UMHS data for this analysis was that it was conducted in different cities than the Seattle MSM surveys. Although not presented here, there were very few differences in behavior patterns between the UMHS cities suggesting that sexual behavior patterns among urban MSM were relatively similar.
A second limitation is that the low survey response rates may have affected the representativeness of our findings if there were differential participation rates associated with sexual behavior. Third, the cross-sectional data in these studies were prone to potential recall bias and the confounding of cohort effects, which can affect the estimation of longitudinal patterns. Finally, we did not directly measure the incidence of HIV or other STI. Because of this, we were unable to specifically evaluate the behaviors of persons with STI, which are clearly key with respect to transmission risk.
Our finding that MSM continued to form new partnerships later into adulthood than heterosexuals, coupled with much higher levels of concurrency and age disassortative mixing, demonstrates important ways in which the sexual behavior and sexual network patterns of MSM and heterosexuals differ. Although these differences may explain part of the observed disparity in HIV/STI rates by sexual orientation, our data provide relatively little insight into why these patterns of sexual behavior vary. Relatively few MSM partnerships in our analysis were formed in the context of personal social networks. Perhaps the separation of social and sexual networks observed among many MSM—a phenomenon that is at least partially conditioned by factors such as laws, culture, and attitudes toward homosexuality—plays a critical role in fostering the sustained HIV/STI epidemics in this population. Attitudes toward sexual minorities and laws affecting their relationships (eg, gay marriage) are changing in many parts of the world. Thus, it is plausible that the long-standing cultural forces that promote relatively safe patterns of sexual behavior among heterosexuals have exerted relatively little influence on MSM. Insofar as social norms pertaining to same-sex relationships, marriage, and parenting continue to change, MSM sexual behavior patterns may also change, ultimately reducing risk for HIV/STI among MSM.
The Urban Men's Health Study was conducted under the direction of Joe Catania and Ron Stall with support from NIMH (MH54320).
1. Glick SN, Golden MR. Persistence of racial differences in attitudes toward homosexuality in the United States. J Acquir Immune Defic Syndr. 2010;55:516–523.
2. Centers for Disease Control and Prevention. Diagnoses of HIV Infection and AIDS in the United States and Dependent Areas, 2009. Atlanta, Georgia: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention; 2009.
3. Centers for Disease Control and Prevention. Sexually Transmitted Disease Surveillance, 2009. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention; 2010.
4. Purcell DW, Johnson C, Lansky A, et al. Calculating HIV and syphilis rates for risk groups: Estimating the national population size of men who have sex with men (abstract). Paper presented at: National STD Prevention Conference; March 10, 2010; Atlanta, GA.
5. Goodreau SM, Goicochea LP, Sanchez J. Sexual role and transmission of HIV type 1 among men who have sex with men, in Peru. J Infect Dis. 2005;191(suppl 1):S147–S158.
6. McGowan I, Taylor DJ. Heterosexual anal intercourse has the potential to cause a significant loss of power in vaginal microbicide effectiveness studies. Sex Transm Dis. 2010;37:361–364.
7. Varghese B, Maher JE, Peterman TA, et al.. Reducing the risk of sexual HIV transmission: quantifying the per-act risk for HIV on the basis of choice of partner, sex act, and condom use. Sex Transm Dis. 2002;29:38–43.
8. Goodreau SM, Golden MR. Biological and demographic causes of high HIV and sexually transmitted disease prevalence in men who have sex with men. Sex Transm Infect. 2007;83:458–462.
9. Levin EM, Koopman JS, Aral SO, et al.. Characteristics of men who have sex with men and women and women who have sex with women and men: results from the 2003 Seattle sex survey. Sex Transm Dis. 2009;36:541–546.
10. Nelson SJ, Hughes JP, Foxman B, et al.. Age- and gender-specific estimates of partnership formation and dissolution rates in the Seattle sex survey. Ann Epidemiol. 2010;20:308–317.
11. Morris M, Kretzschmar M. Concurrent partnerships and the spread of HIV. AIDS. 1997;11:641–648.
12. Hurt CB, Matthews DD, Calabria MS, et al.. Sex with older partners is associated with primary HIV infection among men who have sex with men in North Carolina. J Acquir Immune Defic Syndr. 2010;54:185–190.
13. Berry M, Raymond HF, McFarland W. Same race and older partner selection may explain higher HIV prevalence among black men who have sex with men. AIDS. 2007;21:2349–2350.
14. Kraut-Becher JR, Aral SO. Patterns of age mixing and sexually transmitted infections. Int J STD AIDS. 2006;17:378–383.
15. Hallfors DD, Iritani BJ, Miller WC, Bauer DJ. Sexual and drug behavior patterns and HIV and STD racial disparities: the need for new directions. Am J Public Health. 2007;97:125–132.
16. 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.
17. Millett GA, Peterson JL, Wolitski RJ, Stall R. Greater risk for HIV infection of black men who have sex with men: a critical literature review. Am J Public Health. 2006;96:1007–1019.
18. Millett GA, Flores SA, Peterson JL, Bakeman R. Explaining disparities in HIV infection among black and white men who have sex with men: a meta-analysis of HIV risk behaviors. AIDS. 2007;21:2083–2091.
19. Brooks RA, Lee SJ, Newman PA, Leibowitz AA. Sexual risk behavior has decreased among men who have sex with men in Los Angeles but remains greater than that among heterosexual men and women. AIDS Educ Prev. 2008;20:312–324.
20. Blair J. A probability sample of gay urban males: The use of two-phase adaptive sampling. J Sex Res. 1999;36:39–44.
21. Mills TC, Stall R, Pollack L, et al.. Health-related characteristics of men who have sex with men: a comparison of those living in gay ghettos with those living elsewhere. Am J Public Health. 2001;91:980–983.
22. Catania JA, Osmond D, Stall RD, et al.. The continuing HIV epidemic among men who have sex with men. Am J Public Health. 2001;91:907–914.
23. Menza TW, Kerani RP, Handsfield HH, Golden MR. Stable sexual risk behavior in a rapidly changing risk environment: findings from population-based surveys of men who have sex with men in Seattle, Washington, 2003-2006. AIDS Behav. 2011;15:319–329.
24. Aral SO, Patel DA, Holmes KK, Foxman B. Temporal trends in sexual behaviors and sexually transmitted disease history among 18- to 39-year-old Seattle, Washington, residents: results of random digit-dial surveys. Sex Transm Dis. 2005;32:710–717.
25. American Association for Public Opinion Research. Standard Definitions: Final Disposition of Case Codes and Outcome Rates for Surveys. Deerfield, IL: American Association for Public Opinion Research; 2006.
26. Laumann EO, Gagnon JH, Michael RT, Michaels S. The Social Organization of Sexuality: Sexual Practices in the United States. Chicago, IL: University of Chicago Press; 1994.
27. Wand H, Wilson D, Yan P, et al.. Characterizing trends in HIV infection among men who have sex with men in Australia by birth cohorts: results from a modified back-projection method. J Int AIDS Soc. 2009;12:19.
28. Murphy G, Charlett A, Jordan LF, et al.. HIV incidence appears constant in men who have sex with men despite widespread use of effective antiretroviral therapy. AIDS. 2004;18:265–272.
29. Morris M, Zavisca J, Dean L. Social and sexual networks: their role in the spread of HIV/AIDS among young gay men. AIDS Educ Prev. 1995;7:24–35.
30. Coburn BJ, Blower S. A major HIV risk factor for young men who have sex with men is sex with older partners. J Acquir Immune Defic Syndr. 2010;54:113–114.
31. Morris M, Kurth AE, Hamilton DT, et al.. Concurrent partnerships and HIV prevalence disparities by race: linking science and public health practice. Am J Public Health. 2009;99:1023–1031.
32. Adimora AA, Schoenbach VJ, Taylor EM, et al.. Concurrent partnerships, nonmonogamous partners, and substance use among women in the United States. Am J Public Health. 2011;101:128–136.
33. Adimora AA, Schoenbach VJ, Doherty IA. Concurrent sexual partnerships among men in the United States. Am J Public Health. 2007;97:2230–2237.
34. Santhakumaran S, O'Brien K, Bakker R, et al.. Polygyny and symmetric concurrency: comparing long-duration sexually transmitted infection prevalence using simulated sexual networks. Sex Transm Infect. 2010;86:553–558.
35. Eaton JW, Hallett TB, Garnett GP. Concurrent sexual partnerships and primary HIV infection: a critical interaction. AIDS Behav. 2011;15:687–692.
36. Goodreau SM, Cassels S, Kasprzyk D, et al.. Concurrent partnerships, acute infection and HIV epidemic dynamics among young adults in Zimbabwe. AIDS Behav. 2012;16:312–322.
37. Golden MR, Stekler J, Hughes JP, Wood RW. HIV serosorting in men who have sex with men: is it safe? J Acquir Immune Defic Syndr. 2008;49:212–218.
38. Helms DJ, Weinstock HS, et al.. HIV testing frequency among men who have sex with men attending sexually transmitted disease clinics: implications for HIV prevention and surveillance. J Acquir Immune Defic Syndr. 2009;50:320–326.
© 2012 Lippincott Williams & Wilkins, Inc.