Secondary Logo

Journal Logo

Epidemiology and Social Science

Transmission Behaviors of Recently HIV-Infected Men Who Have Sex With Men

Gorbach, Pamina M. MHS, DrPH*; Drumright, Lydia N. MPH; Daar, Eric S. MD; Little, Susan J. MD§

Author Information
JAIDS Journal of Acquired Immune Deficiency Syndromes: May 2006 - Volume 42 - Issue 1 - p 80-85
doi: 10.1097/01.qai.0000196665.78497.f1
  • Free


Each specific sexual practice carries a different risk for transmitting HIV;1 therefore, the probability of an infected individual transmitting HIV to his or her sexual partner is dependent on the sexual behaviors practiced with that partner as well as on the magnitude of the viral load in the HIV-infected partner.2 Some but not all HIV-infected individuals may intentionally modify their sexual behavior to reduce this risk of transmission to others,3 and these choices may be affected by perceptions about transmission risk. It has been suggested that lessened concerns about the seriousness of HIV infection because of the availability of highly active antiretroviral therapy (HAART) have resulted in an increase of risk behavior among HIV-positive individuals and, consequently, a higher incidence of sexually transmitted diseases (STDs), including HIV.4,5 Whereas HIV-positive men who have sex with men (MSM) may continue risk behavior after their HIV diagnoses,6-9 studies comparing reported risk behaviors over time do not show increased risk taking among HIV-positive MSM.10,11 The rising rates of STDs such as gonorrhea among HIV-positive MSM challenge these data, however, and suggest the need for more focused research on the sexual behavior of HIV-positive MSM.

It has become clear that individuals, regardless of HIV status, vary their practice of different sexual behaviors across the types of partnerships in which they engage.12 For example, more HIV-positive MSM report practicing unprotected anal intercourse (UAI) in stable or main partnerships than in shorter term or less intimate partnerships.13-15 Some HIV-positive MSM also may try to minimize transmission to others through serostatus mixing16 (ie, when HIV-positive individuals partner with those who are also HIV-positive). It is not clear if this practice pervades across all types of partnerships, however. Behavior within partnerships can also be modified to change the amount of transmission risk. For example, MSM may opt for sexual behaviors that may have less transmission risk (eg, using condoms or taking a receptive versus insertive role during UAI).17 Little is known about the timing or types of partnerships in which such decisions are made, however.

Most studies of sexual behavior among HIV-positive individuals involve those who have been infected with HIV for many years. Persons with early HIV infection may represent an especially high transmission risk to their partners because of increased infectiousness associated with having high viral loads.2,18 Understanding which sexual behaviors change after diagnosis may be important in interrupting HIV transmission at early stages of infection when individuals may be most infectious.19


Between May 2002 and October 2004, 164 men and women from southern California who were recently infected with HIV (within the past 12 months) were enrolled in the Acute Infection and Early Disease Research Program (AIEDRP) cohort and completed questionnaires using computer-assisted self-administered interviews (CASIs). Of these participants, 160 were men and 153 (96%) reported sex with men within the 12 months before diagnosis. This study focuses on newly infected MSM who reported sexual activity within the 12 months before diagnosis and completed the follow-up interview. Baseline interviews were completed, on average, 6 weeks after diagnosis (median = 4 weeks) and 13 weeks after the estimated date of infection (median = 14 weeks). One hundred sixteen MSM completed at least 1 follow-up interview an average of 17 weeks after baseline (median = 14 weeks). At follow-up, 106 of the 116 MSM reported sexual contact with other people; however, 3 reported that their last partner was a woman and were therefore dropped from these analyses.

At each interview, participants were asked to report on the numbers and types of sexual partners that they had contact with in the past 3 months. Partner types were derived from our previous qualitative research on sexual behavior of MSM20 and further confirmed in our more recent qualitative work.21 These included the following partnership types (definitions of each type in Table 2): unknown partner, 1-time partner, acquaintance partner, friend partner, regular partner, main partner, and trade partner. Additionally, at baseline and follow-up interviews, MSM were asked to provide detailed information about the last person with whom they had sexual contact, including partner characteristics, partner type, types of sexual activities, drug use, knowledge of partner's HIV status, and disclosure of the participant's HIV status to the partner. At follow-up, participants were also asked to identify if a partner was the same partner reported in the baseline interview and what happened in the preceding period with that partnership. Participants provided first names or initials for partners reported, but the interview specified that these could be invented or nicknames.

Change in Number of Partners by Type From Baseline to Follow-Up (n = 113)

Changes in numbers of partners from baseline to follow-up were examined using paired t tests. Predictors of UAI within each period were determined using χ2 analysis and the Fisher exact test. Generalized estimating equations (GEEs) were used to determine changes in UAI by partner type and which partner types were more likely to be involved in UAI across data points. Multivariate logistic regression models were used to examine baseline or follow-up characteristics that predicted having UAI with the last partner at follow-up or having more than 1 sexual partner at follow-up.


The mean age of the 113 MSM who completed baseline and follow-up interviews was 35 years; most were white (71%), most were employed (68%), and most had completed at least some college (86%) (see Table 1). Greater than 50% of subjects reported 95 or more lifetime partners at baseline. The mean numbers of partners reported were 27.6 in the 12 months before their baseline interview, 7.9 in the 3 months before their baseline interview, and 3.5 in the month before their baseline interview. Most reported ever seeking sex on the Internet (71%) and having group sex in the past year (82%), but fewer reported ever selling sex (16%) or buying sex (23%) in their lifetime or injection drug use in the past year (7%) (see Table 1). There were no significant differences in the demographic or behavioral characteristics for the MSM who were recruited in San Diego (n = 90) and Los Angeles (n = 23); therefore, they were combined in the analyses.

Baseline Characteristics and Sexual Behaviors of Newly Infected Cohort by HIV Transmission Behavior (n = 113)

At follow-up, 91% of the participants reported sexual activity since their baseline interview. Almost half of those sexually active MSM (46.9%, n = 53) reported a decrease in number of partners in the 3 months before the interview from baseline to follow-up, a third (33.6%, n = 38) reported similar numbers of partners, and a fifth (19.5%, n = 22) reported an increase in the number of partners. This corresponded to an overall decrease in the mean number of partners reported by this cohort from baseline (mean = 7.9) to follow-up (mean = 5.2; Table 2). Participants also reported significantly fewer mean numbers of 1-time partners (1.9 vs. 0.8) and acquaintance partners (1.1 vs. 0.5) and nearly significantly fewer unknown partners (3.7 vs. 2.6). Numbers of friend, regular, main, and trade partners were not different between baseline and follow-up. At baseline, 46.0% (n = 52) of participants reported that they had a main partner; at follow-up, 59% (31 of 52 participants) of those with a main partner at baseline still had their same main partner, 19% (10 of 52 participants) had a new main partner, and 26% (14 of 52 participants) had separated from their main partner and were single. Of the remaining 58 participants who did not have main partners at baseline, 19% (11 of 58 participants) reported a main partner at follow-up and 81% (47 of 58 participants) did not report a main partner at any point between baseline and follow-up.

There was no difference in the percentage reporting UAI (insertive or receptive) with their last partner from baseline to follow-up (59% at both). At baseline, only those whose last partner was a main partner were significantly more likely to report having UAI than not having UAI with that partner, with an odds ratio (OR) of 3.29 (95% confidence interval [CI]: 1.46 to 7.43); this was also true at follow-up (OR = 3.79, 95% CI: 1.55 to 9.24). Additionally, at baseline only those whose last partner was a friend were significantly less likely to report UAI than no UAI (OR = 0.17, 95% CI: 0.03 to 0.87). At follow-up (but not at baseline), those participants whose last partners were unknown or 1-time partners were less likely to report UAI than to have UAI with these types of partners (OR = 0.29, 95% CI: 0.10 to 0.81; OR = 0.20, 95% CI: 0.39 to 1.06, respectively). The only partner type in which there was a significant change in the percentage reporting UAI was unknown partners; fewer men reported UAI with these partners at follow-up than at baseline (OR = 0.24, 95% CI: 0.06 to 0.93). When data were pooled across baseline and follow-up, MSM were more likely to report UAI with a main partner than with any other partner type (Fig. 1), although the difference was not significant for regular and acquaintance partners.

Generalized estimating equations of UAI comparing probability of UAI with main partner vs. all other partners: Data Pooled Across Baseline (n = 152) and Follow-up (n = 113)

Although UAI was associated with having a main partner at baseline, partner serostatus was not associated with UAI regardless of the type of partner (Fig. 2). Moreover, the serostatus of a main partner was not significantly associated with the practice of UAI at baseline in that partnership (19.4% reported UAI with an HIV-positive main partner, 63.9% reported UAI with an HIV-negative main partner, and 20.8% reported UAI with a serostatus unknown main partner; χ2 test, P = 0.40). At follow-up, however, UAI was associated with partner serostatus; those with HIV-positive partners were more likely to report UAI than those with negative or unknown serostatus partners (see Fig. 2). This was also true within main partnerships (61.3%, 32.2%, and 6.5% reported UAI with HIV-positive, HIV-negative, and serostatus unknown partners, respectively; χ2 test, P = 0.02). Among main partnerships that continued from baseline to follow-up (n = 31), 16% versus 42% reported an HIV-positive main partner at baseline and follow-up, respectively, and 55% versus 42% reported an HIV-negative main partner at baseline and follow-up, respectively. Finally, 29% of those with main partnerships that continued from baseline to follow-up reported that they did not know their main partner's serostatus at baseline, and 16.1% (n = 5) reported that they did not know this at follow-up.

Among those reporting UAI with last partner, sero-status of that partner at baseline (n = 113) and follow up (n = 103). a, Follow-up analyses include only the 103 MSM who reported sexual contact with another man at follow-up. b, This figure depicts two separate chi-square tables, and differences shown are significant at P < 0.05

Multivariate analyses were used to examine variables that were associated with reported practice at follow-up of behaviors risky for transmission: UAI and more than 1 recent partner (Table 3). After controlling for age and ethnicity, UAI with the last partner at follow-up was associated with reporting that the last partner was the main partner (OR = 2.94, 95% CI: 1.04 to 8.33), reporting an HIV-positive partner at baseline (OR = 3.36, 95% CI: 1.27 to 8.88), having more than 1 partner (OR = 0.28, 95% CI: 0.09 to 0.86), reporting that the last partner at follow-up was HIV-negative (OR = 0.28, 95% CI: 0.08 to 1.00) or of unknown HIV status (OR = 0.23, 95% CI: 0.08 to 0.71), and having experienced nonconsensual sex (OR = 0.25, 95% CI: 0.09 to 0.72). Reporting more than 1 partner at follow-up was associated with not having a main partner at baseline or follow-up (OR = 2.76, 95% CI: 1.12 to 6.78), reporting more partners in the last 12 months (OR = 1.02, 95% CI: 1.01 to 1.04), and reporting UAI with the last partner (OR = 0.36, 95% CI: 0.14 to 0.90).

Multivariate Logistic Regression Models for (1) More Than 1 Partner at Follow-Up and (2) UAI With Last Partner at Follow-Up (n = 113)

Comparisons between baseline data on the 113 MSM who completed a follow-up interview and the 40 MSM who did not complete a follow-up interview or were excluded revealed that there were no differences by age, ethnicity, number of sexual partners reported in the 3 months or 1 month before the baseline interview, or drug use with the last 3 partners. MSM not followed up did report more sexual partners over their lifetime and the 12 months before baseline than those who did complete the follow-up interview, however.


In this cohort, MSM diagnosed with recent HIV infection reported a modest but significant decrease in risk behaviors soon after diagnosis. Nearly one half of these MSM reported fewer sexual partners in the 3 months after their baseline interview (which occurred approximately 1 month after diagnosis) than in the 3 months before their baseline interview. This suggests that the HIV diagnosis and/or symptomatic illness associated with HIV seroconversion precipitated a drop in, although not elimination of, transmission risk. In addition to fewer total numbers of partners, significantly fewer anonymous and 1-time partners were reported. Some maintained their main partnerships, and some formed new main partnerships. Our findings illustrate that UAI occurred frequently within main partnerships and more often than in other types of partnerships, raising concerns about transmission within main partnerships. Additionally, of those main partnerships that persisted from baseline to follow-up, 4 of the main partners were reported by the participant as HIV-negative at baseline but were reported as HIV-positive at follow-up. Although it is not possible to determine if these partners were infected by the participant without sequencing data from both partners and such data are not currently available, the change in the reported serostatus of partners of these recently infected MSM is foreboding.

Many of the newly infected HIV-positive MSM in this study reported main partners at baseline and at the 3-month follow-up. In multivariate analyses, the existence of a main partner at follow-up was significantly associated with 2 different behaviors risky for transmission: having more than 1 partner and having UAI. Those who did not have a main partner at either interview reported more partners at follow-up. Those who reported that their last partner was a main partner were more likely to have had UAI with this partner at baseline and follow-up, however. Although data on sexual positioning at baseline was only available for some participants in the study (102 of 153 participants) because of the late addition of this variable, participants were more likely to report insertive UAI with main partners than with other types of partners. This suggests a high risk of transmission within main partnerships, whereas for those without a main partner, a risk of transmission to other partners only if condoms fail. Therefore, it is important to determine partnership status of MSM recently infected with HIV, because different approaches to prevention are needed based on their types of partners. For example, for MSM with main partners, prevention strategies may be more helpful if focused on prevention for the partner and adopting harm reduction within the relationship. New approaches to counseling may be necessary to reduce risky behaviors in established and intimate partnerships. For example, the negotiation around condom use may be different with main partners than with other types of partners.

Our findings also suggest the need for other types of counseling at the time of HIV diagnosis for recently infected individuals. Men report UAI with HIV-negative partners at the time of seroconversion, presumably because they believe that they are still HIV-negative. Therefore, identifying MSM with HIV infection as close to seroconversion as possible and counseling them on reducing their numbers of sexual partners and decreasing UAI could help to reduce the incidence of HIV infection. One well-tested theory, the Stages of Change Theory,22 illustrates that behavior change follows a continuum in which rather than undergoing complete and permanent changes in behavior, individuals are expected to "fail" along the path to change (ie, recidivism is expected to be part of the process of change). Viewed in this light, any decrease in the number of partners or increase in condom use might be considered a success or as part of the process of behavior change. Therefore, these findings might offer encouragement that individuals who are recently infected with HIV make progress in reducing their risk of transmission to others within the first 3 months of diagnosis, as was the situation for almost half of our sample. Others have found significant reductions in transmission behavior among MSM after being diagnosed with HIV,23 and our findings show that such changes can happen rapidly. Whether or not and for how long such changes may persist, however, remain unanswered.

Also encouraging is that more MSM reported sex with a seroconcordant partner at follow-up than at baseline (87% vs. 63%) and more reported UAI with seroconcordant than with serodiscordant or serostatus unknown partners (87%, 60%, and 31%, respectively). Although superinfection remains a concern for individuals during acute and early HIV infection,24 a modification in behavior may result in reduced HIV transmission to uninfected individuals. Interventions that promote and provide opportunities for HIV-positive individuals to choose new partners who are also positive should be further investigated.

This study suggests that there is potential for significant interruption of HIV transmission during early infection if individuals are diagnosed early and willing to modify their behavior after diagnosis; however, the sample may not be representative of newly infected North American MSM. Most participants are well educated, report white ethnicity, and must initiate contact with our study through self-referral from advertisements (posted in venues catering to gay identified men) or clinician referral. This suggests that these men may be more concerned with their health and the health of others than MSM who are recently infected with HIV but not included in this study. Another issue is that partner serostatus is reported by the study participant and may not be accurate, because individuals tend to be able to report more accurately on activities done together (eg, UAI)25,26 than on their partner's disease status27 or what sexual behavior their partner practices with others.28,29 Finally, we recognize that the reduction in the practice of risk among MSM in this cohort may be attributable to how they feel emotionally while adjusting to a diagnosis of HIV rather than to their response to counseling. Yet, even if the reduction in the numbers of partners was a reflection of their emotional status, it is encouraging that among those who continued to have a lot of sex, harm reduction behaviors such as serostatus mixing and condom use during anal intercourse were reported, thereby reducing their potential for transmitting.

To have a significant impact on HIV incidence among MSM in the United States, the behavior changes demonstrated in this study and previous studies23 need to be more pronounced. Moreover, identification of highly infectious individuals with early HIV infection must be enhanced to intervene early. The findings of this study of men recently infected with HIV illustrate that interventions designed specifically for MSM in main partnerships are needed to help them negotiate a balance between protecting their partner (ie, preventing infection of an HIV-negative partner or superinfecting an HIV-positive partner) and achieving the intimacy perceived to be acquired from unprotected intercourse. As access to technologies that allow for detection of early HIV infection increases, new prevention strategies for individuals diagnosed with recent HIV infection that support them in achieving rapid behavior change within existing and new partnerships are urgently needed.


1. Vittinghoff E, Douglas J, Judson F, et al. Per-contact risk of human immunodeficiency virus transmission between male sexual partners. Am J Epidemiol 1999;150:306-311.
2. Quinn TC, Wawer MJ, Sewankambo N, et al. Viral load and heterosexual transmission of human immunodeficiency virus type 1. Rakai Project Study Group. N Engl J Med 2000;342:921-929.
3. Chen SY, Gibson S, Weide D, et al. Unprotected anal intercourse between potentially HIV-serodiscordant men who have sex with men, San Francisco. J Acquir Immune Defic Syndr 2003;33:166-170.
4. Rietmeijer CA. Resurgence of risk behaviors among men who have sex with men: the case for HAART realism. Sex Transm Dis 2005;32:176-177.
5. van der Snoek EM, de Wit JB, Mulder PG, et al. Incidence of sexually transmitted diseases and HIV infection related to perceived HIV/AIDS threat since highly active antiretroviral therapy availability in men who have sex with men. Sex Transm Dis 2005;32:170-175.
6. McGowan JP, Shah SS, Ganea CE, et al. Risk behavior for transmission of human immunodeficiency virus (HIV) among HIV-seropositive individuals in an urban setting. Clin Infect Dis 2004;38:122-127.
7. Erbelding EJ, Stanton D, Quinn TC, et al. Behavioral and biologic evidence of persistent high-risk behavior in a HIV primary care population. AIDS 2000;14:297-301.
8. Do AN, Hanson DL, Dworkin MS, et al. Risk factors for and trends in gonorrhea incidence among persons infected with HIV in the United States. AIDS 2001;15:1149-1155.
9. Centers for Disease Control and Prevention. Increases in unsafe sex and rectal gonorrhea among men who have sex with men-San Francisco, California, 1994-1997. MMWR Morb Mortal Wkly Rep 1999;48:45-48.
10. Crepaz N, Hart TA, Marks G. Highly active antiretroviral therapy and sexual risk behavior: a meta-analytic review. JAMA 2004;292:224-236.
11. Glass TR, Young J, Vernazza PL, et al. Is unsafe sexual behaviour increasing among HIV-infected individuals? AIDS 2004;18:1707-1714.
12. Gorbach PM, Galea JT, Amani B, et al. Don't ask, don't tell: patterns of HIV disclosure among HIV positive men who have sex with men with recent STI practising high risk behaviour in Los Angeles and Seattle. Sex Transm Infect 2004;80:512-517.
13. Semple SJ, Patterson TL. HIV-positive gay and bisexual men: predictors of unsafe sex. AIDS Care 2003;15:3-15.
14. Hays RB, Kegeles SM, Coates TJ. Unprotected sex and HIV risk taking among young gay men within boyfriend relationships. AIDS Educ Prev 1997;9:314-329.
15. Gross M, Buchbinder SP, Celum C, et al. Rectal microbicides for U.S. gay men. Are clinical trials needed? Are they feasible? HIVNET Vaccine Preparedness Study Protocol Team. Sex Transm Dis 1998;25:296-302.
16. Golden MR, Brewer DD, Kurth A, et al. Importance of sex partner HIV status in HIV risk assessment among men who have sex with men. J Acquir Immune Defic Syndr 2004;36:734-742.
17. Van de Ven P, Kippax S, Crawford J, et al. In a minority of gay men, sexual risk practice indicates strategic positioning for perceived risk reduction rather than unbridled sex. AIDS Care 2002;14:471-480.
18. 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.
19. Wawer MJ, Gray RH, Sewankambo NK, et al. Rates of HIV-1 transmission per coital act, by stage of HIV-1 infection, in Rakai, Uganda. J Infect Dis 2005;191:1403-1409.
20. Galea JT, Gorbach PM, Whittington WL, et al. An ethnographic study of sexual practices and partnering characteristics of MSM with gonorrhea in Seattle, WA. Presented at: International Society of Sexually Transmitted Disease Research; 1999; Denver.
21. Gorbach PM, Galea J, Amani B, et al. Don't ask, don't tell: patterns of HIV disclosure among HIV positive men who have sex with men (MSM) with recent STI practicing high risk behavior in Los Angeles and Seattle. Sex Transm Infect 2004;80:512-517.
22. Prochaska JO, DiClemente CC, Norcross JC. In search of how people change-applications to addictive behaviors. Am Psychol 1992;47:1102-1114.
23. Colfax GN, Buchbinder SP, Cornelisse PG, et al. Sexual risk behaviors and implications for secondary HIV transmission during and after HIV seroconversion. AIDS 2002;16:1529-1535.
24. Smith DM, Wong JK, Hightower GK, et al. Incidence of HIV superinfection following primary infection. JAMA 2004;292:1177-1178.
25. 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.
26. Seage GR III, Mayer KH, Horsburgh CR Jr, et al. Corroboration of sexual histories among male homosexual couples. Am J Epidemiol 1992;135:79-84.
27. Niccolai LM, Farley TA, Ayoub MA. HIV-infected persons' knowledge of their sexual partners' HIV status. AIDS Educ Prev 2002;14:183-189.
28. Stoner BP, Whittington WL, Aral SO, et al. Avoiding risky sex partners: perception of partners' risks vs partners' self reported risks. Sex Transm Infect 2003;79:197-201.
29. Drumright LN, Gorbach PM, Holmes KK. Do people really know their sex partners? Concurrency, knowledge of partner behavior, and sexually transmitted infections within partnerships. Sex Transm Dis 2004;31:437-442.

behavior change; acute HIV infection; men who have sex with men; HIV; sexual partners

© 2006 Lippincott Williams & Wilkins, Inc.