Stigma and HIV risk among Metis in Nepal

Abstract

Similar to other parts of Asia, the HIV epidemic in Nepal is concentrated among a small number of groups, including transgender people, or Metis. This study was conducted to explore the social context of stigma among Metis in Nepal to better understand their risk for HIV. Fourteen in-depth interviews were conducted with Metis in Kathmandu, Nepal. We found that stigma from families leading to rural-urban migration exposed Metis to discrimination from law enforcement, employers and sexual partners, which influenced their risk for HIV. Specific HIV-related risks identified were rape by law enforcement officers, inconsistent condom use and high reported numbers of sexual partners. These data point to an immediate need to work with law enforcement to reduce violence targeting Metis. HIV prevention, housing and employment outreach to Metis in rural areas and those who migrate to urban areas is also needed. Finally, there is a need for more research to determine the prevalence of HIV among Metis, to explore risk within sexual networks and to better understand of the relationship between Metis and their families in order to develop future programmes and interventions.

Introduction

With the exception of several countries in sub-Saharan Africa, HIV prevalence is generally greater among men who have sex with men than among men who report sex with women exclusively (Laurence 2007). A recent systematic review of the literature in Asia found that men who have sex with men are 18 times more likely to be infected with HIV than other reproductive-age adults (Baral et al. 2009). Evidence of this trend exists in Nepal. Recent behavioural surveillance efforts conducted in 2009 found a 3.8% HIV prevalence among urban men who have sex with men (Family Health International 2007). Compared to the overall HIV prevalence in Nepal of .55%, men who have sex with men are disproportionately impacted and have been categorised as a group at high risk for HIV (Joint United National Programme on HIV/AIDS/World Health Organization 2006).

Transgender individuals are often included in the men who have sex with men category in HIV surveillance studies in Nepal and elsewhere, although this classification has been argued to be epidemiologically inappropriate due to sub-group variations in HIV risk by differences in identity, behaviour and sexual networks (Khan, Bondyopadhyay, and Mulji 2005). Meti is a Nepali word that can be loosely translated to mean transgender woman or a person born a man but who expresses a feminine gender identity. However, there is a range of identities among Metis that span a spectrum of masculinity and femininity, which may be represented in the variety of words used to refer to Meti individuals such as third gender, transvestite, hijra, TG, Meta and Pinky meta (Family Health International 2007). Compared to other men who have sex with men, transgender women have higher HIV infection rates in North America and Europe (Bockting et al. 2005). There is a reason to suspect that these patterns hold in Nepal but, given the lack of distinction between transgender women and men who have sex with men in local surveillance efforts, it is difficult to identify the extent of HIV infection within this sub-population. Nevertheless, a recent probability-based sample of men who have sex with men in four states of India, Nepal’s geographic and cultural neighbour, found that transgender individuals had the highest HIV prevalence of all individuals considered men who have sex with men (Brahmam et al. 2008).

We are aware of no published studies that have been conducted with Metis in Nepal to explore specific HIV-related risk behaviours (TreatAsia 2006). Yet researchers have come to realise the importance of the culture-specific social context of risk behaviours, especially as HIV increasingly affects ethnic and sexual minority groups who are stigmatised in society (Aggleton and Parker 2002; Rhodes et al. 2005).

Nepal is one of the poorest countries in the world, with 42% of its people living below the poverty line. The Maoist conflict that began in the mid-1990s and only recently ended contributed to Nepal’s impoverishment by displacing an estimated 50–70,000 people (Internal Displacement Monitoring Centre 2010). It is estimated that more than one third of the GDP is fuelled by remittances and development aid (Federal Research Division 2005; Seddon, Adhikari and Gurung 2002). Nepal is also a conservative patriarchal society, with religion, caste and marriage as central organising forces. The last 250 years of highly politically and socially conservative rule by the monarchy is the root of the contemporary form of conservatism in Nepal. Marriage has ancient roots in Hinduism, which is the predominant religion in Nepal. Marriage is considered universal, weddings are usually arranged and couples often live with the husband’s family (Caltabiano and Castiglioni 2008). Marriage likely serves many purposes: given the impoverishment of many Nepalese, it may be an important way to sustain the informal safety net through family ties as is done in other developing countries (Foster 2007). Marriage also upholds traditional values in Nepal that stress abstinence before marriage and sex only for procreation and within marriage (Chakrapani et al. 2007).

Men in Asia are seen as masculine through marriage and fatherhood, regardless of their sexuality (Asthana and Oostvogels 2001; Laurent 2005). Research in India and other parts of Asia has found that most societies are tolerant of male-to-male sex as long as it remains invisible (Asthana and Oostvogels 2001; Bardon et al. 2005). However, tolerance of Metis appears to be low. Although Nepal does not have any laws that criminalise homosexuality or transgenderism, there are numerous reports of law enforcement committing brutal violence against Metis (Human Rights Watch 2007). Metis are an easy target for law enforcement officers because they are visible in society by their feminine gender presentation.

The vulnerability of transgender women to HIV worldwide has been tied to the stigma and discrimination they face in society (Bockting, Robinson, and Rosser 1998; Sugano, Nemoto, and Operario 2006). Stigma can directly threaten the physical wellbeing of those who are stigmatised through acts of violence and through experiences of prejudice that create serious stressors to health and well-being (Link et al. 2001). In South Asia all men who defy traditional gender roles are targets of violence. Research with men who have sex with men in India has theorised that men with a feminine gender expression are targets of stigma and violence as a social punishment for defying gender roles (Chakrapani et al. 2007).

Stigma can also affect health indirectly. The disproportionate burden of HIV/AIDS among ethnic/racial and sexual minority groups in the USA has been tied to cultural barriers, power issues, socio-economic status, access to healthcare and a mistrust of the government, which are all linked to possessing a stigmatised status (Centers for Disease Control and Prevention 2007; Rhodes et al. 2005). HIV risk among transgender women may be similar to women’s risk for HIV infection, which is situated in a system of social inequalities based on gender differences (Amaro and Raj 2000). One study was identified in the literature that measures transphobia, described as discrimination towards the transgender community, which found a correlation between transphobia and HIV risk among transgender female youth (ages 15–25) (Sugano, Nemoto, and Operario 2006). Kothi-identified men who have sex with men in India who have a feminine gender presentation have been found to experience extreme violence and discrimination, which has also been tied to increased HIV risk (Chakrapani et al. 2007).

Stigma may be an important factor in the HIV risk environment of Metis in Nepal. Yet data are needed to understand the particular social context for Nepali Metis and how their stigmatised status may impact risk for HIV. Data from this study are presented by first exploring the social context of stigma through the words of Meti participants. Next, the direct and indirect impact of stigma on HIV risk-related behaviours for Metis are discussed, followed by ways in which these data may inform policy and a research agenda targeted to HIV prevention for Metis in Nepal.

Methods

Participants and procedures

Data for this analysis were drawn from a larger study to explore sexual and gender identity, sexual networks, risk perception and stigma among men who have sex with men in Nepal, which included 14 Metis. The larger study was an exploratory, formative study that used qualitative research methods to assess HIV risk among 20 men who have sex with men in Nepal. Data for this analysis were taken from interviews conducted with Metis who were either (1) service providers who served as key informants and research assistants for the larger study (n = 4) or (2) participants recruited from the community as part of the larger study (n = 10). Four of the five service providers interviewed for the larger study were young Metis (< 30 years old). Service providers participated in a semi-structured in-depth interview about HIV risk among men who have sex with men and Metis, in addition to their personal experiences as Metis. These data were used to identify sub-populations of men who have sex with men that would support a maximum variation sampling approach for the larger study, which it was decided would include men who identified as gay, men who have sex with men who did not identify as homosexual or gay but had sex with other men, Meti sex workers and Metis who were not sex workers. The large proportion of Metis was included based on the lack of data regarding Metis and the service provider observations that Metis were at extremely high risk for HIV. Data regarding service providers’ personal experiences as Metis were also used as texts for this analysis. All of the service providers had formal education in English allowing for interviews to be conducted in English by the US principal investigator (PI). The PI administered interviews face-to-face and donated US$50 to the organisation where the providers worked to compensate for their time and participation.

Based on the maximum variation sampling approach developed for the larger study, 10 Metis were recruited to participate in semi-structured, in-depth qualitative interviews. Half of the sample participants recruited were currently sex workers (n = 5) and the other half were not currently sex workers (though most had a history of sex work) (n = 5). In-depth interviews were semi-structured in order to maintain openness during data collection and were focused generally on gathering information that explored sexual and gender identity, risk perception and sexual networks in relation to HIV risk. All men who have sex with men participants were reimbursed US$10 for their time.

The PI conducted a structured training with the five service providers who participated as key informants to be trained as research assistants to recruit and conduct the in-depth interviews with participants. Their previous experience as participants and as service providers made these individuals ideal research interviewers due to the depth of understanding they had about the study goals prior to being trained. The training consisted of discussing the specific constructs of interest and the design of this study, determining a recruitment strategy based on the sub-populations identified in the provider data and visiting cruising sites to determine the feasibility of safely recruiting participants and conducting the study within the vicinity of these sites. The research assistants also participated in refining the interview guide using their insights as participants in the service provider interview that addressed similar constructs. Finally, the PI trained the research assistants on qualitative research methods that built upon their existing capacity as service providers. Research assistants were encouraged to incorporate their own probes throughout the interviews and were given leeway to modify the language used in the questions to be culturally relevant to participants. For example, when asking questions about sexual and gender identity, research interviewers often asked if participants felt as if they were more masculine or feminine, which was a culturally relevant way to discuss identity.

A purposeful recruitment strategy was used to gather information and perspectives on the constructs of interest from members of different men who have sex with men sub-populations. Sub-populations were identified in data collected from service providers and two of those sub-populations included Metis (i.e. Metis and Meti sex workers). The study used a convenience sampling approach. Participants were recruited by the interviewers at a weekly drop-in centre and STD clinic within the local collaborating nongovernmental organisation (Blue Diamond Society), local cruising sites in Kathmandu (e.g. a local park where many men who have sex with men meet informally after work hours) and during a community kick-off event for the study. Interviews were conducted at scheduled times in a discreet location (e.g. office space, room within a coffee shop). The interviewers approached individuals who either disclosed their Meti identity as clients of the local nongovernmental organisation, attended the community kick-off event that was targeted to Metis and men who have sex with men or presented in a feminine manner while at cruising areas. Since interviewers were also community members, they were easily able to identify potential participants. Potential participants were asked if they would like to be screened to determine eligibility for participation in the study at the time of recruitment. Inclusion criteria for the study were: (1) over age 18, (2) biologically male, (3) has had sex with men, (4) Nepali and (5) consents to participate. In order to provide anonymity and protect the identity of participants, verbal consent was given to participate in this study and no identifying information was collected. Interviews were translated and transcribed by a local consulting firm. Translation and transcription was done verbatim and local Nepali slang was left in the transcripts with explanations provided by the local transcriber.

Human subjects approval was obtained from the University of California, San Francisco and the Nepal Health Research Council. All interviews were conducted in the preferred language of the participants in a private office space and were digitally audio-taped.

Data analysis

The interviews ranged in length from 1.5 to 2.5 hours and all interviews were audio taped and transcribed. A phenomenological approach was used to explore HIV risk and stigma within conceptions of identity and the cultural experience specific to Nepal (Giorgi 1997). Using this approach, the essential themes from the raw narratives were taken and the participants’ experiences were condensed into narratives with public health implications. Data were coded using preliminary codes from the analysis of service provider data and then inductively through gradual analysis of the data and emerging themes (Wengraf 2001). The authors were able to bracket, or set aside, preconceived ideas in data analysis discussions by building on the diversity within the research team. Two authors had worked directly with transgender women in research or services settings, while the additional two authors had no prior research experience with the population from which data for this study were gathered. In analysis meetings those who were new to this population raised questions and requested explanations, a process that enabled the team to identify assumptions to be bracketed. Based on findings from the service provider data, particular attention was paid to identifying themes that illuminated HIV-risk perception, sexual networking and the influence of stigma on these phenomena.

An important component of the data analysis was memo-taking, which allowed for productive team discussions and an iterative process of analysis. Once all coding was complete and memos recorded, memos were used to guide discussions between the PI and other authors. The memos also allowed for the PI to reflect on the personal assumptions and subjective interpretations that may have influenced this analysis, which were then discussed with the other authors. Once separated, relationships between indexed ideas and themes regarding experiences related to HIV risk were mapped to help synthesise and interpret the data.

Findings

The social construction of stigma towards Metis and men who have sex with men: Hinduism and Marriage

Participants saw Nepali society as conservative and found that the Hindu religion was used to justify stigma towards Metis as unnatural. Yet multiple participants cited stories about revered Hindu gods who were transgender and the important role transgender individuals played in ancient Hindu societies, suggesting previous societal acceptance of people who are transgender. Participants thought that contemporary society re-interpreted ancient Hindu texts to fit the current social conservatism that instead saw Metis as deviants to be stigmatised:

Interviewer (I): Do you know about negative or positive thoughts toward homosexuals?

Respondent (R): I don’t know about negative thoughts. I know about the positive thoughts for gays and Metis in Hindu religion. I recently came to know the word “Kinner” which mean Transgender. We read in mythological books that Kinners are the prettiest girls and they are invited in functions. It is also mentioned that Mahabharata thought Arjun pretended to be transgender. Though it has been mentioned and accepted by the religion at past, people from today’s generation have been denying acceptance of third genders. They say that ancient third genders were gods, and we [modern Metis ] are human so we cannot be like gods. (Saloni, Community participant, 21 years old, non-sex worker)

Participants pointed to the importance of marriage, with roots in Hinduism, as fuelling stigmatisation towards Metis and others who did not conform to traditional gender roles (i.e. men who have sex with men). Hinduism stresses that an important role of marriage is to perpetuate one’s family through offspring (Laurent 2005). Meti identity was particularly challenging to families because parents were concerned about their ability to arrange a marriage for their son or force a son who was feminine to be a man. Most Metis in this study dressed as women at night (i.e. dressed in Saris or more modern dresses and skirts, wore makeup, styled their hair like women, had their nails done), particularly for social gatherings or to give performances, though many of these participants dressed as men during the day (i.e. jeans, t-shirts, sneakers). Two participants lived as women all the time. Most others had long hair and nails, which, despite their male apparel, made them visibly gender non-conforming because they expressed feminine characteristics. Metis who chose to live as women or present in a feminine manner went against traditional norms around appearance and identity. They also were seen as avoiding their social, religious and familial duty to carry on the family lineage:

According to our Hindu religion, there are roles of son and responsibilities of being a son including having children to ensure that the blood line is continued. We generally are going against these norms. (Monish, Community participant, non-sex worker, 30 years old)

In addition to marriage fulfilling a religious purpose, marriage is a way to mitigate economic stress in the family. A wife could contribute to the economic viability of the family through employment, childcare, chores and other family duties. This is particularly important in Nepal where most people live in poverty and the family structure is the only safety net upon which people can rely. A small minority of Metis fulfilled their marriage obligation and agreed to get married to women:

I: If you do not have feelings for females, on which grounds did you get married?

R: Marriage … it was more because of mother and father.

I: What did they do?

R: They said get married. They brought me a wife.

I: Was it love marriage or arranged marriage?

R: It was arranged.

I: Why didn’t you fall in love with a female?

R: I was small then, so I did not know. I was studying and I did not know then.

I: Did you marry with only two of you around or was it a social affair with family’s involvement?

R: If was a family affair. It happened like this. In the shop I had two sisters who married and went. After they went then mother and me, we were alone. My mother kept getting sick often. I use to go to the school so there weren’t anyone looking after our shop. Then mother started asking me to get married and bring my wife. (Aditya, Community participant, non-sex worker, 40 years old)

Despite the economic and social importance of family ties in Nepal, many Metis reported leaving the family. A number of participants reported abuse due to their gender identity or presentation:

I: In what way did you get mistreated by the family?

R: They ask me to behave as a guy and tell me not to behave as a girl. It’s natural for me to behave as a girl so I cannot change my behaviour, due to which they scold me and beat me. (Saloni, Community participant, non-sex worker, 21 years old)

Yet the primary driving force behind leaving the family was the fear of being forced to marry a woman. Most Metis reported trying to avoid being forced to marry by putting it off for as many years as possible, which created an intense pressure that eventually led to a decision to leaving the family and village for another life:

For example I am now 30 years, this is the time to get married, get settled and mould ourselves socially. They are putting pressure on me [to marry] but I am finding one excuse or another and am saying, ‘next year, next year, next year.’ This creates a lot of mental pressure and frustration. Due to this many do not want to go home and drift away from family. (Monish, Community Participant, non-sex worker, 30 years old)

The pressure by family to get married was not only driven by age, procreation and earning expectations, but also by the questioning of family members and villagers about the gender identity of participants. Participants expressed angst about the shame their family endured due to their gender expression and unmarried status and realised that much of the pressure to marry came from the pressure family members felt from villagers and extended relatives to prove their son was masculine:

And if they know or feel that their son is just like a lady, then the family forces marriage. Parents know about the sexuality, but they force them to marry a girl because they want some kids from their son. (Jasina, Service Provider, 24 years old)

Forms of HIV-risk related to the stigmatised status of Metis

Both married and unmarried Metis suffered consequences tied to gender-related migration that have important implications for HIV risk. Unmarried Metis left the family to avoid abuse and being forced to marry. Married Metis migrated to the city in order to find work and be able to support the family back in the village. Both married and unmarried Metis also sought a place where they could openly present as women while away from the village and family. Yet both groups experienced discrimination by employers due to their feminine gender presentation:

When we go for the job they don’t know that we are Métis they hire us, but once they know about us they fire us from the job. (Saloni, Community participant, non-sex worker, 21 years old)

As a result, many Metis became involved in sex work to survive or as a means to make money to support their families back home. In fact, all Metis interviewed for this study had engaged in sex work at some point in time.

Meti sex workers appeared to be at high risk for HIV due to the number of partners they had and their unique economic vulnerability within sex exchanges. Limited condom access was an important issue that reduced the likelihood of safe sex during sex exchanges. Metis reported having low access to condoms because they could not afford condoms and free condoms were not always accessible. Metis also reported engaging in unsafe sex despite the desire to use condoms because some partners would not pay for sex if condoms were required:

I: Do you know about condoms?

R: I know. I may not know many things, but I know that if I wear a condom, I can be safe from the diseases that I may get infected with.

I: Is it easy for you to get a condom?

R: If I go to buy, I may get it. But one thing is that I do not generally have money to buy. Secondly, [the Blue Diamond Society¹] provides condoms if we go on Saturdays.

I: Then why don’t you use it?

R: Even if I try using it, for me the most important thing is money. You know I don’t have a house. I need to pay house rent. I have to buy everything from salt … so … for this reason I do not use it.

I: You feel that you should use condoms or you feel that you should not use condoms?

R: I don’t know … if I wear … it should be required. … They say we get I-don’t-know-what kind of diseases, it is to be safe from these …

I: Don’t you tell your sex partners?

R: I ask them to wear it. But not all agree.

I: But it is also your responsibility to convince them, isn’t it?

R: I can make them convinced, but they say that if they are convinced, they will not give money. (Namita, Community participant, sex worker, 30 years old)

Metis were also at high risk for HIV due to having many sexual partners (ranging from 8–10 to 40–50 a month). Both sex workers and those who were not currently sex workers reported having many sexual partners. Non-sex workers reported difficulties finding a steady partner due to their stigmatised status as transgender women making partners unwilling to commit to a relationship. Metis reported that as a result of not having a committed partner they had many casual partners. As in the example below, the high number of partners many Metis reported may have increased their susceptibility to HIV simply through a high number of reported sexual contacts:

I: Do you have short-term or long-term affair with your sex partner?

R: I do meet the same guy and have sexual intercourse with him but I don’t get into long-term commitment.

I: Why do you have short-term affairs?

R: Most of the guys pretend as if they love us but after having sexual intercourse they don’t want to get in touch. So I don’t want to get involved with these guys …

I: To date, with how many guys did you have sexual intercourse?

R: I cannot tell the exact number. I had sexual intercourse with many guys.

I: You don’t have to be exact. You may tell us the approximate number.

R: I think it’s around 400–700 men with whom I must have had sexual intercourse …

I: Among the guys, with whom you had sexual intercourse, do they have sexual intercourse with only you or they have sexual intercourse with many partners?

R: I think they have sex with many partners. Some might only have sexual intercourse with me but most of them I think they have sexual intercourse with many guys. (Ambica, Community participant, non-sex worker, 25 years old)

Law enforcement in Nepal is made up of security personnel (i.e. army personnel), police and informal militants who are members of the Maoist party. Violence by law enforcement against Metis in Nepal is pervasive in media reports and data from participants in this study points to ways in which Metis are targeted due to their identity and perceived sexual availability:

Police officers are the ones who mistreat us the most. At Pashupati [Hindu temple] a police officer wanted to have sexual intercourse with me. When I refused to have sexual intercourse with him he took me the police station and accused me of having sexual intercourse in the jungle with guys. I told the senior officer that I was innocent and the officer is falsely accusing me. No one believed me and they locked me up in the police station. (Ambica, Community participant, non-sex worker, 25 years old)

Rape by law enforcement has direct implications for HIV risk among victims because they were not able to require condom use:

In the lockups police officers forcefully have sexual intercourse with us. During these sexual intercourses we request them to use condom but they disagree to use condom. Twice I was locked in police station, there 12 police officers beat me. They dragged me to the toilet and forcefully had sexual intercourse with me without using condom. When I requested them to use condom they threw [away] the condom that I had in my pocket. (Meena, Community participant, sex worker, no age reported)

Known rape and violence toward Metis also created fear of law enforcement, which had an indirect impact on HIV risk-related risk behaviours. One of the crimes Metis feared being falsely arrested for was condom carrying, which officers used as evidence of engagement in public sex and/or prostitution. When asked about condom use, Metis reported that there was a high risk associated with carrying condoms due to fear of being falsely arrested and then violated by law enforcement officers:

I: Isn’t that [carrying condoms] risky from police and other people?

R: Indeed yes. If they find us with condom, they think that we are having sex, then they take us in their custody, they beat us, or we have to have sex with them, give them money. It is risky. (Namrata, Service provider, no age reported)

Many Metis reported having sex in parks, ditches, public toilets and the jungle areas near temples, which had an important impact on condoms use. One reason given for engaging in public sex was that sex with other men was stigmatised and public places were often the only place where Metis could be with a male sex partner in secret. Most Metis either permanently or temporarily residing in Kathmandu did not have a residence or they lived in cramped, one-room apartments with other individuals where they did not have private space in which to have sex. Despite the risk of being arrested, precarious housing situations and a lack of resources to get a hotel room led many Metis to have sex in public. The decreased likelihood of having a condom available during public sex encounters along with the difficulty of implementing safe sex practises while in hurried sexual encounters often led to unsafe sex:

R: There is no fixed place. It’s sometimes in jungles, in parks. It’s a habit that has grown from my early age.

I: Is it possible to have safe sex during that time?

R: No. … Sometimes I use condoms, sometimes I do not use. How do you find condoms in the woods. (Nirmala, Community Participant, HIV positive, 30 years old)

Discussion

Data from this study describe the ways in which stigma by family, employers, sexual partners and law enforcement impacts HIV risk for Metis in Nepal. Metis in this study were stigmatised due to their gender presentation, feminine behaviour and unwillingness to fulfil traditional gender roles of a son in Nepali society. Participants described ways in which modern interpretations of Hinduism were the root of highly valued conservative notions around male gender roles that stressed the importance of fulfilling the obligations of being a son (i.e. to get married) or face rejection from family and society.

Most Metis in this study had migrated to the urban centre of Kathmandu where this study was conducted. Some Metis migrated to avoid being forced to marry, while others migrated to escape abuse from family and others for their feminine gender presentation. Metis who migrated to the urban area often ended up engaging in sex work, putting them at high risk for HIV (Brahmam et al. 2008; Herbst et al. 2008). Metis engaged in sex work because they faced gender discrimination that prevented them from securing employment. Khan et al. (2005) has linked economic vulnerability leading to sex work among gender-variant individuals in South Asia to pervasive genderphobia or ‘the fear of the gender variant and the those who are not acting out the normative masculinity’ (19). Due to their extreme vulnerability, Metis did not have the power to require safe sex with paying partners and were often forced to engage in unsafe sex despite the desire to use condoms.

Metis were also at risk for HIV due to having many sexual partners. Participants discussed the difficulties with being in a long-term relationship because their sexual partners were unwilling to commit to a relationship with a transgender woman. Participants attributed their sexual partners’ unwillingness to commit to the stigmatised status of Metis in Nepali society. Because Metis were unable to find committed partners, they had many sex partners to fulfil their sexual desire.

Overt stigma has been shown to affect the mental health of transgender women leading to increased HIV-risk behaviour (De Santis 2009). Research with transgender women in the US has found that transphobia causes some transgender women to look for gender affirmation in ways that increase their risk for HIV, notably through sex (Kammerer, Mason, and Connors 1999; Reback and Lombardi 1999). Research with hijras and other groups in India and Bangladesh that have been likened to Metis in Nepal has not identified this phenomenon (Brahmam et al. 2008; Chan and Khan 2007). However, hijras are a cultural group whose identities are more determined by the inability to have children than sexual behaviour, while Meti identity appears to be more tied to gender identity and sexual attraction to males. Though not reported by participants, stigmatised Metis in Nepal, especially recent migrants and those beginning the process of transitioning to another gender, may have been seeking affirmation of their female gender through sex, which in turn may have led to a high number of sexual partners.

Sexual risk for HIV may also have been high for Metis because they are part of a sexual network that is disproportionately impacted by HIV in Nepal (Wohlfeiler and Potterat 2005). HIV surveillance efforts have determined that men who have sex with men are one of four most at-risk populations for HIV in Nepal (New Era 2009). Metis, with many sexual partners who are counted in surveillance efforts as men who have sex with men, may be at increased risk for HIV due to the fact that their sexual network is known to have high rates of HIV.

The most direct form of stigma Metis experienced that had clear implications for HIV was harassment and violence by law enforcement. Metis in this sample reported personal experiences of being targets for false arrest, assault and rape by law enforcement officers because of their gender presentation. Chakrapani et al. (2007) theorised that police harassment towards Kothis in India is a form of punishment for transgressing traditional gender boundaries and an affirmation of the masculinity of the perpetrators. Nepal’s law enforcement officers, who at the time of the study were still employed by the conservative monarchy, may have also been acting to uphold the conservative societal values that stigmatise Metis because they transgress gender roles.

Violence towards Metis had direct and indirect impacts on Metis’ risk for HIV. Rape by law enforcement officers directly impacted HIV risk for Metis. Metis’ subsequent fear of law enforcement had an indirect effect on HIV risk. Participants, including sex workers, reported often having sex with their partners in public settings like local parks, ditches, public toilets or in the jungle where it was already difficult to implement condom use. And because Metis were fearful of being falsely arrested by law enforcement officers for carrying condoms, participants reported not carrying condoms at all, thus further reducing the chances that Metis had condoms on hand during risky sexual encounters.

A number of factors associated with this study prevent the generalisation of these findings to the Metis more generally in Nepal. These data were collected with a small convenience sample of Metis in Kathmandu who have distinct risks associated with their migration patterns to Kathmandu, including engagement in sex work. The sample also included service providers, who were articulate in their description of the vulnerabilities Metis faced, but may have biased the data because they also served as the research interviewers for the study. However, the benefit of a comprehensive understanding of the issues, study goals and the entrée research interviewers had as service providers to the community were key elements to the successful implementation of this study with a highly hidden and vulnerable population in Nepal. Another limitation was that most of these data were collected with primary oversight from afar as the PI was unable to be in the field throughout the duration of the study. The distance between the researcher and study site limited oversight and quality control of data collection; however, the co-author was in Nepal for the duration of the study and was able to provide oversight as a provider with experience conducting research in Nepal. Despite these limitations, these data do fill a gap in the literature about the plight of a highly vulnerable group of transgender women at risk for HIV in Kathmandu, Nepal that may serve to justify and inform future prevention efforts.

Implications

Data from this study highlight a number of ways in which stigma and discrimination results in HIV-related sexual risk behaviour and HIV risk among Metis in Nepal. There is an immediate need for the enforcement of the existing constitutional law to protect Metis from violence and discrimination in Nepal. The fact that those whose job it is to protect citizens are the primary perpetrators of violence towards Metis points to the need for a top-down approach by legislators and others in power that can force change in the culture of Nepal’s law enforcement agencies to prevent these acts of violence towards Metis. Strides in work with law enforcement may have an important impact on HIV risk among Metis by increasing condom carrying and use, especially among Meti sex workers. The cultural backdrop of stigma towards Metis within families is complex and prevention efforts that address families will need more data to inform creative strategies to tackle family dynamics either directly or indirectly. In the interim, outreach efforts in rural areas of Nepal are needed to reach those individuals with no outlet for discussing or expressing their true gender identity. Outreach efforts need to provide counselling, HIV-prevention information and support for those intending to migrate to Kathmandu or other urban areas of Nepal. The urban areas, in turn, need to have services to assist Metis in finding employment and housing to prevent engagement in sex work and precarious sexual encounters.

These data illuminate a number of ways in which Metis are at risk for HIV, yet accurate behavioural surveillance data are still needed to assess the HIV prevalence among this highly vulnerable population. An important consideration for surveillance efforts is the differentiation of Metis from men who have sex with men in order to determine risk for HIV within this particular risk population. Surveillance research may also consider studying the sexual networks of Metis in Nepal. These data speak to the fact that Metis are likely to have a high number of sexual partners. Data about the connections, if any, between these sexual partners along with prevalence data may enable researchers and interventionists to pinpoint Metis’ specific risks for HIV infection and target prevention strategies to intervene in specific sexual networks and work within partner dynamics (Wohlfeiler and Potterat 2005).