Homosexual activity among rural Indian men: implications for HIV interventions

A cross-sectional survey of 2910 rural Indian men aged 18-40 years from five rural districts in five different states revealed that nearly 10% of single and 3% of married men had had unprotected anal sex with a man in the past year. Homosexually active men are not a separate sexual category, and report extensive mixing with female partners. They have more female partners than other men and they practised anal intercourse in 11% of their heterosexual contacts.

It is estimated that 4.58 million people in India were living with HIV in 2002 [¹]. A diffusion into the general population is now evident, with HIV prevalence among antenatal clinic attendees exceeding 1% in various south Indian states [¹]. In 2001, the prevalence among men who have sex with men (MSM) was 23.6% in Maharashtra and 2.4% in Tamil Nadu. The contribution of homosexual activity to the spread of HIV in India, however, is unknown, despite some evidence that sex between men, including anal intercourse, is not uncommon [²].

The HIV/AIDS prevention programme in India has created artificial dichotomies between heterosexual men and homosexual men (MSM) as an identity, and has an urban bias both of research and interventions. In India, where close to 76% of the population are rural, population estimates of homosexual activity and anal intercourse with men or women have so far not been available.

The data analysed in this paper were collected as part of a large rural sexual behaviour study in five states in India. This paper (i) reports on the prevalence of homosexual activity in a general rural population sample; (ii) describes sexual networking with female partners during the past year; and (iii) characterizes sexual practices with both male and female partners.

In 2001, we conducted a cross-sectional survey on the sexual behaviour of 2910 rural men aged 18-40 years in five districts in five states. The districts are: Mahendragarh (Haryana, northern region), Jalore (Rajasthan, western region), Jaunpur (Uttar Pradesh, central region), Puri (Orissa, eastern region), and Tumkur (Karnataka, southern region). In each of the five districts, 10 villages were randomly selected from those fulfilling three criteria: less than 5% tribal population, population size of more than 500, and at least 5 km away from the nearest town or any urban centre. A complete household listing was performed in each of the 50 villages. From the list of households with at least one man aged 18-40 years, 62 households were randomly selected. To avoid potential embarrassment and ensure confidentiality, only one man from each household was interviewed. Fieldworkers selected the youngest, second youngest etc. … eligible man in strict rotation in consecutive households.

Oral informed consent was obtained from the selected respondents, and the response rate was 94%. Men were interviewed by trained male fieldworkers using a structured and standardized questionnaire. In this study we defined men as homosexually active when they reported anal intercourse with a man in the past year.

The mean age of the men was 28.6 years (standard deviation 6.2), most were married (81%) and 95% were Hindu. Twenty-one per cent of men were illiterate, whereas 23% had completed secondary school. Most men (86%) reported sexual activity in the past year; 12.8% of married men reported an extramarital female partner, and 26.1% of single men had had sexual intercourse with women in the past year.

Homosexual activity was common: 9.5% of single and 3.1% of married men reported anal intercourse with a man in the past year. The prevalence was highest among the 20-24 year olds (6.3%), and declined to a level of 2.2% among men over 35 years. Homosexual activity did not vary by education or religion, but there were marked differences by district. In Puri (Orissa), 17.3% of single men and 9.4% of married men were homosexually active in the past year, compared with 6.0% and 1.7%, respectively, in the other districts combined.

Men with male partners also had more female partners. All homosexually active married men reported sex with their wife, and they were also more likely to report extramarital sex with women in the past year than other married men: 52.8 versus 11.6% (χ² = 105.8, P = 0.000) (Table 1). Similarly, 41.5% of single men with male partners reported female partners, compared with 24.5% of those not homosexually active (χ² = 7.2, P = 0.007). Commercial sex workers seemed to be the least important category of non-marital sex partners. However, men who reported commercial sex with a woman were also more likely to have a male partner (14 versus 4% among men with no commercial partners).

Table 1

Considerably greater mixing was reported by homosexually active men, with 56.9% of married men and 34.0% of single men having had three or more partners, compared with married men (5.0%, Fisher's exact test 138.4, P = 0.000) and single men without male partners (6.5%, Fisher's exact test 29.5, P = 0.000) (Table 1). Very few men (4%) had more than two male sex partners. Most of the male partners were currently single (91 and 78% reported by single and married men, respectively). Sex between men was mainly non-commercial, with only 18% of the most recent sex acts involving payment.

Sex with women was dominated by vaginal penetration, and was less varied within marriage than with other partners. Homosexually active men practised more anal sex with their wife (11.1%) than with other men (2.9%, Fisher's exact test 9.7, P = 0.002). The incidence of anal sex with other women was equally high (10.6%). The incidence of masturbation was much higher between men than with women. Oral sex with male partners was relatively uncommon, but anal intercourse was practised in 74% of last encounters. Seventy-two per cent reported insertive sex and as many as 29.6% reported receptive anal intercourse during their last sex act. None of these encounters with men were protected by condoms.

This first estimate on the scale of unprotected anal sex between men in rural India is worrying. Men who are homosexually active in rural India do not form a separate sexual network. Instead, they link to both circuits of high-risk activity and the general female population. The greater number of partners and the implied concurrency of partnerships (at least with spouse and regular partners) provide the conditions for an accelerated spread of HIV [³]. The frequent practice of anal sex (11%) in heterosexual contact adds to the vulnerability of partners. Heterosexual transmission of HIV in India may be less self-contained than commonly understood. The potential for sexual bridging among men in the general population who are homosexually active may be a 'diluted version' of the extensive sexual mixing with women reported among convenience samples of MSM in south Asia [^2-4], China [⁵] and Peru [⁶].

There seems to be little escape from challenging cultural norms and becoming explicit about risky sexual practices, and India's AIDS programme should start addressing the risk of unprotected anal intercourse with both men and women.

Acknowledgements

The authors would like to thank John Cleland and Carine Ronsmans for valuable comments on earlier drafts.

Sponsorship: The study was funded by UNFPA (grant no. IND/99/P61); the Ford Foundation (grant no. 950-1006) supported the consultants and resource persons.