Sex Partner Selection, Social Networks, and Repeat Sexually Transmitted Infections in Young Men: A Preliminary Report : Sexually Transmitted Diseases

Secondary Logo

Journal Logo


Sex Partner Selection, Social Networks, and Repeat Sexually Transmitted Infections in Young Men: A Preliminary Report

Ellen, Jonathan M. MD*; Gaydos, Charlotte DrPH; Chung, Shang-en ScM*; Willard, Nancy BA*; Lloyd, Laura V. MPH; Rietmeijer, Cornelis A. MD, PhD

Author Information
Sexually Transmitted Diseases 33(1):p 18-21, January 2006. | DOI: 10.1097/01.olq.0000187213.07551.a6
  • Free

MEN WITH REPEAT SEXUALLY TRANSMITTED infections (STIs) may be more responsible for maintaining endemic rates of gonorrhea and chlamydia than men with single infections.1 Understanding factors associated with repeat infection in men is crucial to developing new strategies to reduce or eliminate endemic STIs.

Incident STIs have been shown to be associated with number of sex partners, inconsistent condom use, and new sex partners. Although no studies have demonstrated that these factors are predictors of repeat gonorrhea or chlamydia infection,2–12 research has demonstrated that reexposure to an untreated partner is a reason for repeat infection.13–15 One emerging perspective is that network level factors may be operating, i.e., that a high prevalence of STIs in a person’s social network may be associated with a high risk for incident as well as repeat infections.10 For example, a study of the social networks of individuals with chlamydia infection suggested that individuals who are more directly or indirectly connected to others are more likely to have repeat chlamydia infections.1

The objective of the current longitudinal study is to examine whether the risk for repeat STIs among male adolescents and young adults in Baltimore, Maryland, and Denver, Colorado, is associated with sex partners coming from social networks. Given that social networks may vary in their prevalence of gonorrhea and chlamydia, leaving a network in which one acquired an initial infection when choosing new sex partners might be protective against repeat infection. In the current study, we characterized sex partners on the basis of their overlap with social networks. We hypothesized that a decrease in the percentage of sex partners who are within the index’s social network is associated with a decreased risk for repeat gonorrhea and chlamydia infection over 4 months.



Asymptomatic male adolescents and young adults, 13 to 25 years old, diagnosed with chlamydia and/or gonorrhea as part of a chlamydia and gonorrhea male screening project in Baltimore, Maryland, and Denver, Colorado,18 were recruited to participate in the current study. The male screening project was conducted in multiple types of clinics and venues in both cities, but recruitment for the current study only occurred at noncorrectional sites with a high percentage of eligible males. The recruitment started in February 2001 in both cities and ended in September 2002 in Baltimore and March 2003 in Denver.

In Baltimore, eligible men were recruited from school-based health centers and a free teen clinic run by the health department. In Denver, eligible men were recruited from an STI clinic, a drug treatment facility, and nonclinical venues, including the streets and community-based organizations. There were 106 participants enrolled in Baltimore and the acceptance rate was 44% (106 of 239). There were 208 participants enrolled in Denver and the acceptance rate was 49% (208 of 424). We examined participants’ age, race, STI history, numbers of sex partners in lifetime, and numbers of sex partners in the past 2 and 12 months. We found no statistically significant difference between enrollees and refusers for Baltimore. In Denver, we found that white, older-age males with an STI history were more likely to enroll.

Participants provided 20 mL of first-catch urine (FCU) for testing for Neisseria gonorrhoeae and Chlamydia trachomatis by a nucleic acid amplification assays (NAAT). Initially, Baltimore used the ligase chain reaction test (LCR; Abbott, Abbott Park, IL), but beginning in June 2002, Baltimore switched to strand displacement amplification (SDA; ProbeTec; Becton Dickinson, Sparks, MD). In Baltimore, 86% (91 of 106) of samples were tested with LCR. Denver originally used polymerase chain reaction (PCR; Roche Molecular Diagnostics, Indianapolis, IN) and then switched to ProbeTec in July 2002. In Denver, 75% (156 of 208) of samples were tested with PCR. Collection, storage, and transport of the urine followed the manufacturer’s directions. Infected participants in both cities were treated with a 1-g oral dose of azithromycin.


As part of the screening program, trained staff was sent into the field to notify infected men, ensure their treatment, and notify sex partners. In Baltimore, disease investigation specialists (DIS) and, in Denver, experienced outreach workers conducted these activities. After notifying eligible men of their infection status and eliciting names of sex contacts, DIS and outreach workers described the current study to the eligible men. The DIS and outreach workers then enrolled and obtained written informed consent from men who chose to participate.

Once consent was obtained, participants were interviewed about their sexual behaviors and the social characteristics of each of their sex partners. In some cases, these interviews were conducted at the time of treatment. Participants were reimbursed $25 for their time and were asked to return 1 month and 4 months later. At the 1-month and 4-month visits, participants were reinterviewed and screened for a repeat infection by testing their urine for gonorrhea and chlamydia.

The current study was approved by the Colorado Multiple Institutions Review Board and the Institutional Review Boards of Johns Hopkins School of Medicine and the Centers for Disease Control and Prevention (CDC).


Participant Demographics and Sexually Transmitted Infection History.

As part of the baseline interview, participants provided information on their age, race/ethnicity, marital status, school status, living condition, STI history, number of sex partners during their lifetime and in the past 2 months, as well as condom use at last sexual intercourse with main and casual partners.

Sex Partner Characteristics.

At baseline interview, participants were asked the initials of all of their sex partners (up to a total of 6) in the last 2 months. At subsequent interviews, participants were asked about all sexual partners (new and continuing) since the last interview. Participants were reminded of the initials of the partners they had mentioned at previous interviews.

At each interview, for each sex partner, participants were asked whether the partner was a main or a casual partner, whether he used a condom at last sex with this partner, how and where he first met this partner, whether the partner was in school, whether any close friends had ever had sex with this partner, whether this partner knew any of his other sex partners, STI history of this partner, and whether the partner was a new partner.


We coded each sex partner as within a participant’s social network or not within his social network. A sex partner was considered within a participant’s social network if the participant reported any of the following: he met the sex partner through close friends, more than half of his friends knew the sex partner, any of his friends had sex with the sex partner, or the sex partner knew his other sex partners. For each interview, we calculated the percentage of sex partners who were in the participant’s social network (100 × number of interval sex partners in social network/number of total interview sex partners). This allowed us to compare the percentage of a participant’s sex partners belonging to his social network at the initial and at the follow-up interview. Based on this comparison, we dichotomized participants’ scores into decrease in overlap for participants who had a lower percentage of sex partners belonging to their social network from baseline to follow-up visit and no decrease in overlap for participants whose percentage of sex partners in their social network stayed the same or increased.

A participant was determined to have a repeat STI if either 1) they were diagnosed with gonorrhea at baseline and had a subsequent gonorrhea diagnosis or 2) if they were diagnosed with chlamydia at baseline and had a subsequent chlamydia diagnosis.

After transforming the variables, we conducted stratified analyses by city. The unit of analysis was first follow-up visit. If there was no 1-month follow-up visit, the 4-month follow-up visit was substituted. Each participant contributed one follow-up visit. First, we examined whether reporting a new partner at follow up predicted repeat STI overall in each city. Next, among those with a new partner, we examined whether a decrease in overlap was associated with repeat STI. These analyses were conducted using chi-squared and Fisher exact tests. Odds ratio and 95% confidence interval were also calculated.


There were 47 participants in Baltimore and 92 participants in Denver who had complete social network (defined previously) and STI information. Thirty-four percent (16 of 47) of follow-up visits were from the 4-month follow up for Baltimore and 40% (37 of 92) for Denver. As shown in Table 1, the mean age of participants in Baltimore was 18.5 years and in Denver 20.6 years. All of the Baltimore participants were black, whereas the race/ethnicity of participants in Denver was much more varied. Table 2 describes the recruitment venues for the participants in the 2 cities and Table 3 shows the dispositions on partners of participants in the 2 cities. At their follow-up visits, 78% of participants had one sex partner, 14% had 2 sex partners, 2% had 3 sex partners, and 6% had no sex partners in Baltimore. In Denver, 83% of participants had one sex partner, 8% had 2 sex partners, 5% had 3 sex partners, and 4% had 4 partners at their follow-up visit. At 6.4% (3 of 47) of the Baltimore follow-up visits, participants were diagnosed with a repeat STI and at 6.5% (6 of 92) of the Denver follow-up visits, participants were diagnosed with a repeat STI.

Baseline Characteristics of Participants in Baltimore, MD, and Denver, CO, 2001–2003
Participant Recruitment Venues in Baltimore, MD, and Denver, CO, 2001–2003
Disposition of Locatable Named Sex Partners of Participants in Baltimore, MD, and Denver, CO, 2001–2003

Baltimore and Denver participants with new sex partners over the follow-up interval appeared to be more likely to be diagnosed with a repeat STI compared with those without new sex partners, although the results were not statistically significant (see Table 4). Furthermore, a decrease in the percentage of partners belonging to a participant’s social network (overlap) was protective for repeat STI (see Table 5). For those with no decreased overlap in Baltimore, 2 of 12 had repeat infection, whereas zero of 4 of those with decreased overlap had a repeat infection. In Denver, 2 of 16 participants with no decreased overlap had a repeat infection, whereas one of 13 of those with decreased overlap had a repeat infection. In both cities, there were less repeat STIs among the participants with a reduced percentage of sex partners coming from their social network.

New Sex Partner Between Visits and Odds Ratio for Repeat Sexually Transmitted Infection (STI) Diagnosed at a Follow-Up Visit in Baltimore, MD, and Denver, CO, 2001–2003
Decrease in Overlap From Baseline to Follow Up for Repeat Sexually Transmitted Infection (STI) Among Those With New Sex Partners in Baltimore, MD, and Denver, CO, 2001–2003

To test whether having untreated sex partners confounded the findings in either city, we examined whether participants whose baseline sex partners were all notified and treated were less likely to get reinfected than those for whom only some or none of their sex partners were notified and treated. In Baltimore, one of 18 (5.6%) had a repeat STI among those whose baseline sex partners were all notified and treated, whereas 2 of 29 (6.9%) had a repeat STI among those for whom only some or none of sex partners were notified/treated. In Denver, the figures were zero of 8 (0%) and 6 of 84 (7%), respectively.


The objective of this study was to determine whether selecting new sex partners from outside social networks reduces risk for repeat infection among asymptomatic male adolescents and young adults. A common way for individuals to find sex partners is in their social networks. Therefore, if an individual acquires an STI, his or her social network could be considered a high-prevalence STI group, and that individual could benefit by leaving that group to find new sex partners. In this study, we characterized sex partners on the basis of their membership in men’s social networks. Men testing positive for gonorrhea or chlamydia at an initial visit were interviewed about their sex partners and asked to return for follow up, at which point the screening and interview were repeated.

We found preliminary evidence in both cities that suggests having new sex partners is a risk factor for repeat STI in the context of partner management. We also have preliminary evidence that selecting sex partners from outside social networks was associated with reduced risk for repeat infection. Although the study sample was small, the 2 cities showed a similar trend despite differences in the recruitment venues and racial/ethnic composition of participants in the 2 cities. In addition, the fact that adolescents and young adults with untreated sex partners were not more likely to have a repeat STI suggests that differences in partner notification did not account for the differences between cities in the main study findings.

Although this study is limited by its small sample size, which restricts power as well as the ability to do multivariate analyses, the study provides preliminary evidence that network-level factors may be associated with repeat STI. To the extent that other investigators are able to replicate our Baltimore findings in other populations, these data suggest that interventions must be designed to reduce the prevalence of infection in the social networks of infected men. This might be accomplished by more extensive contact tracing or by the implementation of innovative partner management interventions, including expedited partner therapy13 or treating social contacts including other men.16,17


1. 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.
2. Barnett SD, Brundage JF. Incidence of recurrent diagnoses of Chlamydia trachomatis genital infections among male and female soldiers of the US Army [erratum appears in Sex Transm Infect 2001; 77:464]. Sex Transm Infect 2001; 77:33–36.
3. Burstein GR, Zenilman JM, Gaydos CA, et al. Predictors of repeat Chlamydia trachomatis infections diagnosed by DNA amplification testing among inner city females. Sex Transm Infect 2001; 77: 26–32.
4. Ellen JM, Hessol NA, Kohn RP, Bolan GA. An investigation of geographic clustering of repeat cases of gonorrhea and chlamydial infection in San Francisco, 1989–1993: Evidence for core groups. J Infect Dis 1997; 175:1519–1522.
5. Hillis SD, Owens LM, Marchbanks PA, Amsterdam LF, Mac Kenzie WR. Recurrent chlamydial infections increase the risks of hospitalization for ectopic pregnancy and pelvic inflammatory disease. Am J Obstet Gynecol 1997; 176:103–107.
6. Oh MK, Cloud GA, Fleenor M, Sturdevant MS, Nesmith JD, Feinstein RA. Risk for gonococcal and chlamydial cervicitis in adolescent females: Incidence and recurrence in a prospective cohort study. J Adolesc Health 1996; 18:270–275.
7. Rietmeijer CA, Van Bemmelen R, Judson FN, Douglas JM Jr. Incidence and repeat infection rates of Chlamydia trachomatis among male and female patients in an STD clinic: implications for screening and rescreening. Sex Transm Dis 2002; 29:65–72.
8. Whittington WL, Kent C, Kissinger P, et al. Determinants of persistent and recurrent Chlamydia trachomatis infection in young women: Results of a multicenter cohort study. Sex Transm Dis 2001; 28:117–123.
9. Brooks GF, Darrow WW, Day JA. Repeated gonorrhea: an analysis of importance and risk factors. J Infect Dis 1978; 137:161–169.
10. Gunn RA, Fitzgerald S, Aral SO. Sexually transmitted disease clinic clients at risk for subsequent gonorrhea and chlamydia infections: Possible ‘core’ transmitters. Sex Transm Dis 2000; 27:343–349.
11. Kinghorn GR, Pryce D, Morton RS. Repeated gonorrhea in Sheffield: The size of the problem, epidemiologic significance, and personal characteristics of repeaters. Sex Transm Dis 1982; 9:165–169.
12. McEvoy BF, Le Furgy WG. A 13-year longitudinal analysis of risk factors and clinic visitation patterns of patients with repeated gonorrhea. Sex Transm Dis 1988; 15:40–44.
13. Golden MR, Whittington WL, Handsfield HH, et al. Effect of expedited treatment of sex partners on recurrent or persistent gonorrhea or chlamydial infection. N Engl J Med 2005; 352:676–685.
14. Schillinger JA, Kissinger P, Calvet H, et al. Patient-delivered partner treatment with azithromycin to prevent repeated Chlamydia trachomatis infection among women: A randomized, controlled trial. Sex Transm Dis 2003; 30:49–56.
15. Kissinger P, Brown R, Reed K, et al. Effectiveness of patient delivered partner medication for preventing recurrent Chlamydia trachomatis. Sex Transm Infect 1998; 74:331–333.
16. Rothenberg R. The transformation of partner notification. Clin Infect Dis 2002; 35(suppl 2):S138–S145.
17. Rothenberg RB, Sterk C, Toomey KE, et al. Using social network and ethnographic tools to evaluate syphilis transmission. Sex Transm Dis 1998; 25:154–160.
18. Schillinger JA, Dunne EF, Chapin JB, et al. Prevalence of Chlamydia trachomatis infection among men screened in 4 US cities. Sex Transm Dis 2005; 32:74–77.
© Copyright 2006 American Sexually Transmitted Diseases Association