Men who have sex with men (MSM) constitute a high-risk group for sexually transmitted infections (STIs), including HIV, in many parts of the world.1-4 Data on MSM are not routinely collected in India, however, where MSM is a taboo subject, prohibited by law, and often not recognized as a distinct social construct. Despite these restrictions, MSM are prevalent in India and a better understanding of this risk group is required.
HIV was first reported in India in 1986 among female commercial sex workers (CSWs). India is now estimated to have the highest number of HIV infections in the world, with more than 5.5 million persons currently living with HIV.5,6 Since 1987, when HIV sentinel surveillance and AIDS case identification in India began, the predominant mode of HIV transmission identified has been heterosexual sex. The prevalence of MSM and the role this group plays in the HIV epidemic are largely unknown, because surveillance for MSM has been limited.7-9 Although data are limited, studies suggest that MSM behaviors are occurring in complex and diverse ways beyond those who self-identify as homosexual.10-15
The objectives of our study were to identify the prevalence of MSM and to compare HIV and STI prevalence and risk behaviors of MSM with those of non-MSM attending STI public clinics in Pune, India over a 10-year period. STI clinics represent an important access point for persons who are infected with HIV or at high risk for HIV infection. Identifying the prevalence of MSM and risk behaviors unique to MSM is essential for the development of appropriately targeted prevention messages in controlling HIV and STIs among MSM in India.
Between May 1993 and August 2002, 13,249 persons (10,785 men) attending 1 of 3 referral STI clinics in Pune, Maharashtra (1 municipal STI clinic, 1 STI clinic in a state government teaching hospital, and 1 health care center in a red light district) were screened for HIV-1, HIV-2, and STI diagnosis and treatment. These clinics are affiliated with the National AIDS Research Institute (NARI) and are located in public sector hospitals that serve a largely urban indigent population with a low educational status. The project was approved by the Johns Hopkins University Joint Committee on Clinical Investigation and the Ethics Committee of the NARI. Both Institutional Review Boards have an assurance of compliance with the US Department of Health and Human Services Regulations for the Protection of Human Research Subjects with the Office for Protections from Research Risks (now Office for Human Research Protections [OHRP].
The methods for this study have been described in detail elsewhere.16 Eligibility criteria included age of 18 years or older, seeking care at the STI clinic, and willing to provide informed consent for study participation. After consent was obtained, a structured interviewer-administered questionnaire was used to obtain data on demographics, medical history, sexual and other risk behaviors, and knowledge of HIV. Participants underwent a physical examination and were tested for HIV and STIs as described elsewhere in this article. At the time of the interview and physical examination, HIV status was unknown. All participants received pre- and posttest counseling for HIV and STIs. We defined men reporting any sexual contact ever with another man as MSM. Hijras are persons identified as a third gender and are one of the following: castrated men, transvestites, transsexuals, or, rarely, hermaphrodites. The 84 hijras identified were excluded from our analysis of men, in keeping with current Indian cultural interpretation of hijras. The questionnaire was revised in 2000 to collect more detailed information on types of first and regular partners as well as on risk behaviors and condom use. For the 114 MSM seen between January 2000 and August 2002, we were able to define MSM further as men who ever reported having a male sexual partner but did not report having a regular male sexual partner as “MSM ever” and MSM who reported having a regular male sexual partner as “MSM regular.”
HIV and Sexually Transmitted Infection Diagnostic Criteria
Serum samples were screened with a commercially available enzyme-linked immunosorbent assay (ELISA) kit for identification of HIV-1 and HIV-2 antibodies (Recombigen HIV-1/HIV-2 Rapid Test; Cambridge Biotech, Galway, Ireland). Specimens testing positive by ELISA were confirmed using the Recombigen HIV-1/HIV-2 Rapid Test Device (Cambridge Biotech). Specimens that were discrepant by these assays were confirmed using a third ELISA or the HIV-1 or HIV-2 Western blot assay (Western blot assays were interpreted according to Centers of Disease Control and Prevention criteria17). Individuals with evidence of HIV-1 or HIV-2 infection were considered to be infected with HIV.
Other Sexually Transmitted Infection Diagnoses
Clinical diagnoses of genital discharge, warts, and ulcers were based on physical examination by a physician trained to make STI diagnoses as well as by the results of clinic-based laboratory tests (eg, Gram stain, dark field microscopy). A diagnosis of gonorrhea (GC) was based on Gram stain or cultures. Blood was screened for syphilis using a venereal disease research laboratory (VDRL) test and, in some cases, a rapid plasma reagent (RPR) test. Positive VDRL or RPR tests were confirmed by a Treponema pallidum hemaglutination assay (TPHA) using standard microbiologic methods.18 The diagnosis and treatment of STIs were according to the recommendations of the World Health Organization (WHO).19
We reviewed data collected from all male participants attending NARI clinics between 1993 and 2002 to determine HIV/STI prevalence and risk behaviors among MSM and compared these with those of non-MSM. Data included only the first participant visit to the STI clinic. We further compared changes in HIV, STIs, and risk behaviors over time among non-MSM and MSM using the Mantel-Haenszel test for trend.
Correlates of HIV infection were examined using χ2 tests for categoric variables, Mann-Whitney tests for continuous variables, and univariate logistic regression. Variables that were significantly associated at the P ≤ 0.1 level in the univariate analysis were included in the multivariate logistic regression analysis to identify factors independently associated with HIV infection. Covariates identified in published literature as important independent risk factors for HIV were also forced into our model regardless of statistical significance in the univariate analysis. Factors considered in the analysis included calendar time, demographics (age, marital status, education, and employment status), age at first sexual intercourse, lifetime number of partners, condom use, and past history and current diagnoses of STI. Condom use was not consistently asked about throughout the 10-year period, except among those who had a regular partner. Potential interactions between independent associations of HIV infection and calendar time were explored.
Finally, we compared the subset of MSM who reported having a regular male sexual partner (MSM regular) with those who ever reported having a male sexual partner (MSM ever) and with non-MSM. All statistical analyses were performed with the use of SAS software (release 8.02; SAS Institute, Cary, NC) and STATA (version 8; Stata Corporation, College Station, TX).
Characteristics of Men Who Have Sex With Men and Comparison With Men Who Do Not Have Sex With Men (1993-2002)
Of 10,785 men screened, 708 (6.6%) were ever MSM. The median age of MSM was 24 years (range: 18-60 years), 444 (62.7%) had less than a high school education, 253 (35.7%) were married, 189 (31.7%) had 10 or more lifetime partners, and 87 (13.3%) reported ever receiving money for sex (Table 1). A total of 643 (90.8%) MSM reported having multiple sexual partner types. One hundred sixty-three (23.1%) reported sex with hijras, 515 (72.7%) reported sex with female CSWs, and 403 (56.9%) reported sex with with female non-CSWs. Among 270 MSM for whom condom use data were available, 165 (61.1%) reported never using condoms.
We compared MSM with non-MSM to identify any significant differences in risk behavior. MSM were more likely to be younger (mean age of 26 years vs. 28 years; P < 0.0001), to have initiated sexual activity at <16 years of age (24.0% vs. 8.1%; P < 0.0001), to have had 10 or more lifetime partners (31.7% vs. 21.4%; P < 0.0001), to have had sex with a hijra ever (23.1% vs. 7.1%; P < 0.0001), and to have used condoms at least some of the time (36.9% vs. 25.8%; P < 0.0001) (Table 1).
HIV and Sexually Transmitted Infection Prevalence and Trends Among Men Who Have Sex With Men and Comparison With Men Who Do Not Have Sex With Men
MSM did not differ significantly in HIV prevalence (18.9% vs. 20%; P = 0.49) but were less likely to be diagnosed with syphilis (5.8% vs. 8.8%%; P = 0.009), GC (4.3% vs. 7.2%; P = 0.005) or genital ulcer disease (GUD; 21.5% vs. 32.6%; P < 0.0001) compared with non-MSM (Table 1).
When comparing MSM between 2 time periods, 1993 through 1997 and 1998 through 2002, HIV prevalence (18.5% vs. 19.5%; P = 0.74), GC (4.5% vs. 4.1%; P = 0.77), or genital warts (7.6% vs. 5.9%; P = 0.38) did not change significantly. GUD (7.3% vs. 32.4%; P < 0.0001) and syphilis (8.2% vs. 3.0%; P = 0.005) decreased significantly, however (Fig. 1A). Similar reductions in STI prevalence, except for HIV prevalence, were observed among non-MSM (Fig. 1B).
We also investigated whether sociodemographic characteristics or risk behaviors changed between these 2 time periods. MSM screened during 1998 through 2002 were less likely to have had at least a high school education (28.0% vs. 44.4%; P < 0.01), to have had fewer lifetime sexual partners (≥10 partners: 23.4% vs. 37.6%; P < 0.01), to have ever had sex with a hijra (27.1% vs. 20.3%; P = 0.04), to have used condoms at least some of the time (44.8% vs. 36.1%; P < 0.0001), and to have a greater awareness of AIDS (81.7% vs. 69.2%; P < 0.01). Like MSM, non-MSM screened during 1998 through 2002 were also less likely to have had at least a high school education (25.0% vs. 45.2%; P < 0.001), to have had fewer lifetime partners (≥10 partners: 15.7% vs. 23.8%; P < 0.01), to have used condoms at least some of the time (32.4% vs. 24.3%; P < 0.01), and to have a greater awareness of AIDS (78.9% vs. 61.4%; P < 0.01) than non-MSM screened during 1992 through 1997. Non-MSM, however, unlike MSM, were less likely to have had sex with a CSW in 1998 through 2002 (70.7% vs. 90.4%; P < 0.001) than in 1992 through 1997.
For MSM and non-MSM, the reason for the clinic visit in 1998 through 2002 was less likely for symptoms than in 1992 through 1997 (MSM: 63.8% vs. 77.4%; P < 0.01 vs. non-MSM: 67.2% vs. 83.4%; P < 0.01).
Correlates of HIV Infection Among Men Who Have Sex With Men
In multivariate analysis, factors independently associated with HIV infection among MSM were being employed (adjusted odds ratio (AOR) = 3.1, 95% confidence interval [CI]: 1.2 to 8.0), having a history of GUD (AOR = 1.9, 95% CI: 1.2 to 2.8), or being diagnosed with laboratory-confirmed syphilis (AOR = 2.1, 95% CI: 0.99 to 4.4) (Table 2). Similar associations were observed among non-MSM in univariate and multivariate analyses, with the exception that increased HIV prevalence was associated with never using condoms and number of lifetime partners. Education and marital status were not associated with HIV risk among non-MSM, however (data not shown).
Additional characteristics of Men Who Have Sex With Men (2000-2002)
Among the 114 MSM identified in 2000 through 2002, 38 (33.6%) were married to a woman, 48 (42.5%) reported that their first sexual partner was male, and 21 (18.6%) reported male lifetime sex partners exclusively. Regular alcohol use (daily or several times per week) was reported among 22 (19.5%) of MSM. Ninety-two (81.4%) reported ever having had insertive anal sex, and 39 (34.5%) reported a history of receptive anal sex. Twenty (26.6%) reported never using a condom.
A further comparison of MSM who reported ever having a male partner (MSM ever) in their lifetime versus MSM who reported having a regular male partner (MSM regular) revealed that MSM regular were more likely to never have been married, have their first sexual partner be a male, and have had receptive anal intercourse but were less likely to have had sex with a female CSW or a hijra (Table 3).
Our study represents data from more than 10,000 men and is the only surveillance study of men attending STI clinics that spans a decade of the HIV epidemic in India. Several important findings were identified. We found that high-risk sexual behaviors and ever having male-to-male sex were observed among men attending STI public clinics in Pune, India but that only a small proportion of these men reported having exclusively male sexual partners. MSM were more likely to have initiated sexual activity at age <6 years, to have had more lifetime partners, and to have had sex with a hijra. MSM were also more likely to be aware of HIV and were more likely to use condoms than non-MSM. Despite these differences, MSM and non-MSM had a prevalence of HIV that approached 20% and did not change significantly over the 10-year period. The number of partners and the prevalence of syphilis and GUD did significantly decrease for MSM and non-MSM over this time period, however. We also identified that many MSM and non-MSM have multiple sexual partner types, including hijras, female CSWs, and female non-CSWs. Finally, we identified at least 2 subsets of MSM: one that reported having multiple partner types, including women (82.5%), and the other reporting regular or exclusively male sexual partners (17.5%), suggesting that different outreach strategies are needed in India.
Relatively little has been written on the prevalence of MSM or associated behaviors in India. Prevalence estimates seem to vary widely, ranging from 1.5% to 37%.20,21 A study of 1573 male college students in Chennai, India found that 16% had experienced sexual activity with another man; however, almost half of such men had also had sexual activity with a woman.22 A North Indian study of 527 truck drivers found that 15% reported having had male-to-male sex.20 A survey of 2910 rural Indian men found that nearly 10% of single men and 3% of married men had had unprotected anal sex with a man in the past year.14 We found a prevalence of MSM ever of 6.6% among our group. This is similar to the 5.9% prevalence of MSM reported in a community-based study of urban slums in Chennai.12 Such numbers can be difficult to compare with one another, however, because different definitions of MSM have been used, including ever MSM, current MSM partner, or MSM behavior in the past 6 months.
We observed an HIV prevalence of 19%. Limited HIV surveillance among MSM from 2 other urban sites in India has found that in 2000 through 2003, 16% to 24% of MSM in Mumbai and 2% to 4% of MSM in Tamil Nadu were HIV-positive.7 When we compare our HIV prevalence rates with those obtained from India's national STI surveillance sites, such high HIV prevalence rates have been reported in only a few regions, including other cities in Maharashtra (ie, Mumbai [18%-30%]) and the states of Andhra Pradesh (19%-30%) and Tamil Nadu (9%-16%).7 Prevalence of HIV in the MSM and STI populations seems to be much higher than the 0.9% estimated HIV prevalence in the general population.
We confirmed the well-known association of STI and HIV risk,7,16,23-25 identifying an approximately 1.5- to 2-fold increase in HIV prevalence among MSM who had a prior or current STI, specifically GUD or syphilis. Interestingly, we found a 3-fold increased HIV prevalence among those MSM who were employed compared with those who were not. This could reflect an ability to spend more on commercial sex or to have greater mobility and opportunity to have partners away from home, but these associations were not identified in our analyses (data not shown). Finally, in terms of behavioral risks, we found that in our population of MSM, condom use increased over time, whereas the number of lifetime partners decreased, likely indicating a positive impact of prevention efforts in India. Nevertheless, condom use was still low among MSM, suggesting that continued emphasis on behavior modification is needed.
Approximately 73% of MSM reported sex with a female CSW, 57% reported having sex with a female non-CSW partner, 36% reported being married, and 23% reported sex with a hijra. Men having multiple partner types, including women, has been observed in other studies in India and elsewhere.1-3,13,15,21,26-29 In many of these studies, the epidemic spread to the heterosexual population has been thought to be partially attributable to the bridging effect of the MSM population, many of whom self-identified as heterosexual and had female partners. In our study, although we did not collect specific data on how people self-identified, we observed differences between MSM regular and MSM ever. MSM regular were more likely to have had receptive anal intercourse. There is also the suggestion that MSM regular had higher HIV and syphilis rates, similar to what has been described among MSM in the West, but our study sample of MSM regular may have been too small to observe a significant difference. In comparison, the MSM ever group was more likely to have multiple partner types, including female CSWs, female non-CSWs, and hijras. The MSM ever group, in particular, may represent a bridging group for the HIV epidemic in India. Therefore, at least 2 prevention efforts must occur: one that focuses on men whose regular partners are male and another that reaches out to men who have male and female partners and are unlikely to identify themselves as MSM. The MSM regular group is more likely to benefit from adapting prevention efforts that have been successful among MSM elsewhere.30 Some groups in India, such as the Humsafar Trust and Naz Foundation, are now engaged in reaching out to self-identified MSM and promoting a safer sex agenda in India. With regard to MSM ever, prevention and intervention efforts need to focus on educating individuals about the different HIV/STI risks associated with their sexual behaviors. In addition, it is similarly important to sensitize health care workers to ask individuals about sexual behaviors and not to assume that because they are married or have a female partner, they may not also be engaging in other risk behaviors that may place them and their partner(s) at risk for HIV and other STIs.
Our study has limitations that must be considered. Our data, although spanning a 10-year time period, were cross-sectional; therefore, temporality of exposure and outcomes cannot be ensured. Furthermore, we measured prevalence rather than incidence; thus, stable prevalence may represent increased survival, whereas overall new infections are decreased in the population. Because antiretroviral therapies were limited in India (<5%) during the study period, however, it is unlikely that there were substantial differences in mortality over time. Our male population did change somewhat over time in that they tended to be older in age, have fewer lifetime partners, and be more likely to use condoms at least some of the time. This may, in part, explain the reduction in STI prevalence noted during the 10-year period. Referral patterns and awareness of AIDS did change over time, and it is possible that men who already knew they were infected with HIV were coming to the clinic. Furthermore, given that homosexuality remains illegal and a tremendous societal stigma exists in disclosure of homosexual behaviors in India, it is likely that underreporting occurred because of lack of a provider's comfort in eliciting such behaviors and lack of a participant's willingness to disclose his sexual activities with other men (ie, participant's willingness to give socially desirable responses). Finally, our study's findings were based on a high-risk population of urban public STI clinic attendees from a single city in India, and therefore may not be representative of the trends occurring among non-STI populations or among STI populations seeking care in the private sector or in other geographic areas, particularly regions with low HIV prevalence. Persons seeking care in public STI clinics are typically indigent and have a low educational status; hence, risk estimates from such settings are likely overestimates of STIs, including HIV prevalence, in the general population. Despite these limitations, our data add important insights regarding MSM and support findings from other recent MSM studies in India.
India has among the highest number of men infected with HIV in the world. To date, India's surveillance, prevention, and intervention efforts have focused predominantly on the heterosexual transmission aspects of the HIV epidemic. Our data, however, highlight that a diverse range of MSM behaviors must be addressed. The next wave of surveillance, prevention, and intervention efforts in India must include a nonstigmatizing rigorous approach to understanding MSM; only then can HIV interventions be more wide reaching.
The authors thank Lidia Propper and Radhika Brahme for their contributions to this paper. They also thank the study staff of the NARI for their participation in data collection, clinical care, and laboratory network. Finally, many thanks to the study participants.
1. Beyrer C, Eiumtrakul S, Celentano DD, et al. Same-sex behavior, sexually transmitted diseases and HIV risks among young northern Thai men. AIDS
2. Caceres CF. HIV among gay and other men who have sex with men in Latin America and the Caribbean: a hidden epidemic? AIDS
. 2002;16 (Suppl 3):S23-S33.
3. Chan R, Kavi AR, Carl G, et al. HIV and men who have sex with men: perspectives from selected Asian countries. AIDS
. 1998;12(Suppl):S59-S65, S67-S68.
4. Zhang B, Liu D, Li X, et al. A survey of men who have sex with men: mainland China. Am J Public Health
5. National AIDS Control Organization. HIV/AIDS epidemiological surveillance and estimation report for the year 2005. Available at: http://www.nacoonline.org
. Accessed August 12, 2005.
8. Dandona L. Enhancing the evidence base for HIV/AIDS control in India. Natl Med J India
9. Mandavilli A. The coming epidemic. Nature
10. Nandi J, Kamat H, Bhavalkar V, et al. Detection of human immunodeficiency virus antibody among homosexual men from Bombay. Sex Transm Dis
11. National AIDS Control Organization. National baseline high risk and bridge population behavioural surveillance survey-2002 part-II (MSM and IDUs). Available at: http://www.nacoonline.org/publication.htm
. Accessed June 8, 2005.
12. Go VF, Srikrishnan AK, Sivaram S, et al. High HIV prevalence and risk behaviors in men who have sex with men in Chennai, India. J Acquir Immune Defic Syndr
13. Dandona L, Dandona R, Gutierrez JP, et al. Sex behaviour of men who have sex with men and risk of HIV in Andhra Pradesh, India. AIDS
14. Verma RK, Collumbien M. Homosexual activity among rural Indian men: implications for HIV interventions. AIDS
16. Mehendale SM, Rodrigues JJ, Brookmeyer RS, et al. Incidence and predictors of human immunodeficiency virus type 1 seroconversion in patients attending sexually transmitted disease clinics in India. J Infect Dis
17. Centers for Disease Control and Prevention. Interpretation and use of the Western blot assay for serodiagnosis of human immunodeficiency virus type 1 infections. MMWR Morb Mortal Wkly Rep
18. Larsen SA, Steiner BM, Rudolph AH. Laboratory diagnosis and interpretation of tests for syphilis. Clin Microbiol Rev
20. McKenna N. On the Margins: Men Who Have Sex With Men and HIV in the Developing World. Special Reports
. London: Panos Institute; 1996.
21. Khan S. MSM and HIV/AIDS in India. Available at: http://www.nfi.net/NFI
%20Publications/Essays/ 2004/MSM,%20HIV%20and%20India.pdf. Accessed June 3, 2005.
22. Hauser D. Sexual risk among male college students in Chennai, India: implications for HIV prevention strategies [abstract TuOrD437]. Presented at: XIII International AIDS Conference, 2000, Durban.
23. Stamm WE, Handsfield HH, Rompalo AM, et al. The association between genital ulcer disease and acquisition of HIV infection in homosexual men. JAMA
24. Plummer FA, Simonsen JN, Cameron DW, et al. Cofactors in male-female sexual transmission of human immunodeficiency virus type 1. J Infect Dis
25. Reynolds SJ, Risbud AR, Shepherd ME, et al. Recent herpes simplex virus type 2 infection and the risk of human immunodeficiency virus type 1 acquisition in India. J Infect Dis
26. Asthana S, Oostvogels R. The social construction of male ‘homosexuality’ in India: implications for HIV transmission and prevention. Soc Sci Med
27. Colby DJ. HIV knowledge and risk factors among men who have sex with men in Ho Chi Minh City, Vietnam. J Acquir Immune Defic Syndr
28. Khan S. Culture, sexualities, and identities: men who have sex with men in India. J Homosex
29. Tabet SR, de Moya EA, Holmes KK, et al. Sexual behaviors and risk factors for HIV infection among men who have sex with men in the Dominican Republic. AIDS
30. Johnson WD, Hedges LV, Diaz RM. Interventions to modify sexual risk behaviors for preventing HIV infection in men who have sex with men. Systematic Reviews, 2002, Issue 4. Art. No. CD001230.
Keywords:© 2006 Lippincott Williams & Wilkins, Inc.
India; men; MSM; homosexual; same-sex behaviors; male-to-male sex; HIV; sexually transmitted infections; syphilis; gonorrhea; genital ulcer disease; prevalence; risk factors; trends