|Year : 2020 | Volume
| Issue : 5 | Page : 53-60
HIV/AIDS-Related risk behaviors, HIV prevalence, and determinants for HIV prevalence among hijra/transgender people in India: Findings from the 2014–2015 integrated biological and behavioural surveillance
Shobini Rajan1, Pradeep Kumar2, Bhavna Sangal3, Arvind Kumar4, Shreena Ramanathan5, Savina Ammassari6
1 Assistant Director General, Strategic Information Management Division, National AIDS Control Organization, Ministry of Health and Family Welfare, Government of India, New Delhi, India
2 Consultant, Strategic Information Management Division, National AIDS Control Organization, Ministry of Health and Family Welfare, Government of India, New Delhi, India
3 Former Technical Officer, Strategic Information Management Division, National AIDS Control Organization, Ministry of Health and Family Welfare, Government of India, New Delhi, India
4 Associate Consultant, Strategic Information Management Division, National AIDS Control Organization, Ministry of Health and Family Welfare, Government of India, New Delhi, India
5 Former Consultant; Strategic Information Division, The Joint United Nations Programme on HIV/AIDS (UNAIDS), New Delhi, India
6 Former Senior Advisor, Strategic Information Division, The Joint United Nations Programme on HIV/AIDS (UNAIDS), New Delhi, India
|Date of Submission||19-Nov-2019|
|Date of Decision||20-Feb-2020|
|Date of Acceptance||27-Feb-2020|
|Date of Web Publication||14-Apr-2020|
Dr. Pradeep Kumar
National AIDS Control Organization, Ministry of Health and Family Welfare, Government of India, 36 Janpath Road, New Delhi - 110 001
Source of Support: None, Conflict of Interest: None
| Abstract|| |
Background: Hijra or transgender (H/TG) people are significantly affected by HIV in India. HIV prevalence among H/TG is the second highest after people who inject drugs. Effective interventions require understanding about various risk behaviors and associated factors for high prevalence. Objectives: This study analyzes the known risk behaviors and vulnerabilities of HIV-positive and HIV-negative H/TG people to identify the determinants of HIV seropositivity in this high-risk group. Methods: Using secondary data from India's 2014 to 2015 Integrated Biological and Behavioural Surveillance survey, this analysis was conducted among 3325 H/TG people across seven states. Probability-based sampling methods were used to recruit H/TG people. Informed consent was obtained for the collection of behavioral information and blood samples for HIV testing. Multivariable binary logistic regression analysis was undertaken to identify the determinants of HIV seropositivity. Results: HIV prevalence for this group of respondents was 9.5%. Multivariable analysis of survey data revealed higher odds of HIV infection if H/TG had regular male partners (adjusted odds ratio [AOR]: 1.81, confidence interval [CI]: 1.07–3.06), were living in the states of Maharashtra (AOR: 6.08, CI: 3.02–12.22) and Odisha (AOR: 2.91, CI: 1.05–8.06), and were members of self-help groups (AOR: 2.08, CI: 1.04–4.14). None of the demographic or behavioral correlates of risk were found to be associated with HIV infection. Conclusion: The findings suggest that community and structural factors, which are inadequately covered in surveys such as IBBS, play a more important role than individual behavioral factors.
Keywords: Determinants, Hijras/transgender people, HIV, sexual behavior, surveillance
|How to cite this article:|
Rajan S, Kumar P, Sangal B, Kumar A, Ramanathan S, Ammassari S. HIV/AIDS-Related risk behaviors, HIV prevalence, and determinants for HIV prevalence among hijra/transgender people in India: Findings from the 2014–2015 integrated biological and behavioural surveillance. Indian J Public Health 2020;64, Suppl S1:53-60
|How to cite this URL:|
Rajan S, Kumar P, Sangal B, Kumar A, Ramanathan S, Ammassari S. HIV/AIDS-Related risk behaviors, HIV prevalence, and determinants for HIV prevalence among hijra/transgender people in India: Findings from the 2014–2015 integrated biological and behavioural surveillance. Indian J Public Health [serial online] 2020 [cited 2020 May 29];64, Suppl S1:53-60. Available from: http://www.ijph.in/text.asp?2020/64/5/53/282416
| Introduction|| |
Transgender (TG) people bear a disproportionately higher risk for HIV and sexually transmitted infections (STIs)., The term “TG” refers to individuals whose gender identity or expression differs from the sex assigned at birth. TG people, who were born male, but self-identify as female, are at much higher risk for HIV infection as shown by evidence collected around the world.,, Through a systematic review of data from 15 countries, Baral et al. estimated HIV prevalence in TG at 19.1% with a 49-fold increase in odds of infection in this group compared to the general adult population. Research suggests that HIV infections among TG are driven by multiple factors. At the individual level, sex work, unprotected anal sex, multiple sex partners, coinfections with perianal STIs, limited HIV knowledge, gender-affirming hormone use, and interplay of psychosocial health problems, such as depression and substance use/abuse, play an important role in increasing HIV vulnerability and risk., At the community level, social determinants such as stigma, discrimination, violence, and victimization due to TG identity, together with high levels of HIV within the group, are major factors of HIV risk.,,
In India, HIV prevalence in TG or Hijras (H), as they are commonly called in the national context in addition to many other names, is also much higher (3.14%) than in the general population (0.26%)., For a long time, Hijra or transgender (H/TG) people have been the second most affected population group in the country.,, HIV Sentinel Surveillance (HSS 2017) confirmed that H/TG people have a higher HIV prevalence (3.14%) than men who have sex with men (MSM) (2.69%) and female sex workers (1.56%). Only people who inject drugs have a higher HIV prevalence rate (6.26%) than H/TG people. In 2012–2013, a mapping study in 17 states estimated around 62,137 H/TG people, with the majority (61%) living in Andhra Pradesh, Maharashtra, Odisha, Uttar Pradesh, and West Bengal.
H/TG people in India continue to be organized into Gharanas which are systems of social organization like a fraternity or clan., Their traditional occupation is “badhai” (i.e., blessing of newborn babies and newly married couples), singing, and dancing. Nowadays, however, due to lack of education, job opportunities, and economic constraints, H/TG people may resort to sex work or begging for a livelihood., Engagement in these stigmatized occupations contribute to further marginalization and discrimination of H/TG.,,
India's National AIDS Control Programme (NACP) has recognized H/TG as a high-risk group. Originally, H/TG people were seen as a subpopulation of men who have sex with men (MSM), but then it became apparent that H/TG people are at higher risk for HIV. In NACP-IV (2012–2017), prevention interventions specifically tailored for H/TG people were already introduced. As some of the specifically targeted interventions started only around 5 years ago, and HIV prevalence in H/TG people remains higher than in other key populations, a better understanding is needed of HIV risk and vulnerability in this group.,, This study presents the results of an analysis of known risk and vulnerability factors in HIV-positive and HIV-negative H/TG persons. The determinants of HIV infection are also examined to identify individual predictors of risk in H/TG using data from national biobehavioral surveillance.
| Materials and Methods|| |
Study design, setting, and data source
Data used in this study were drawn from India's 2014–2015 Integrated Biological and Behavioural Surveillance (IBBS) survey conducted among H/TG people across 14 domains in 11 states. For this analysis, data from only seven states were used, including Andhra Pradesh and Telangana, Karnataka, Maharashtra, Delhi, Odisha, Tamil Nadu, and West Bengal. These states were selected due to an H/TG prevalence above the average of 3.14% (HSS 2017). The present analysis using the secondary data of IBBS 2014–2015 was undertaken during January–March 2019.
Study subjects, sampling, tools/techniques
By definition, the survey covered H/TG “aged 15 years or more, whose self-identity does not conform unambiguously to conventional notions of male or female gender roles, but combines or moves between them.” H/TG people were recruited into the survey across eleven domains, each comprising a single district or a conglomerate of neighboring districts with similar sociocultural characteristics. In each domain, 400 H/TG people were to be recruited using a conventional cluster sampling method for fixed sites (i.e., homes) and time-location cluster sampling approach to recruit mobile H/TG. Where the number of H/TG people was deemed too limited to easily reach the desired sample size, a “take all” approach was used to recruit respondents. Altogether, 3325 H/TG people from seven high-prevalence states are included in the current analysis.
Behavioral information was collected through a questionnaire using computer-assisted personal interview tools. Blood specimens for HIV testing were collected through a finger prick test using dried blood spot filter paper cards. All tools used in the survey were standardized and translated into different vernacular languages. Field teams who implemented the survey were rigorously trained with the help of technical guidelines and manuals to ensure consistency in data collection across multiple sites. Details of the survey methodology are available elsewhere.
Ethical approval was obtained from the National AIDS Control Organisation (NACO) Ethics Committee vide letter number T-11020/20/2008-NACO (R&D) dated September 26, 2013. Informed consent was obtained from all participants following administration of the participation information sheet. For respondents aged 15–17 years, assent was obtained from local guardian. Confidentiality of respondents was given utmost priority; data specimens in IBBS had unique respondent numbers and cannot be linked to any respondent. Referral to the nearest HIV counseling and testing center for free HIV counseling and testing was offered to each participant.
The primary outcome measure for the current analysis was HIV status (dichotomized as HIV positive and HIV negative), determined based on laboratory testing using two-test protocols. Specimen testing was done using enzyme-linked immunosorbent assay kits, and samples reactive in both tests were labeled HIV-positive.
The independent measures considered in this analysis to assess the HIV risk and vulnerability among HIV-positive and HIV-negative H/TG people and to identify the determinants of HIV infection include individual profile characteristics, alcohol and drug use, sexual risk behaviors, stigma/discrimination, experience of violence, HIV/AIDS knowledge, program exposure, and community collectivization.
Profile characteristics included age (grouped into 15–18, 20–24, 25–34, and 35 + years); education (no schooling, secondary, and high school and above); marital status (never married, currently married, and widowed/separated/divorced); self-identification (Akwa or operated, Nirvan or nonoperated, and others); occupation (sex work, traditional [such as dancing, singing, begging, and “badhai”] and others); and state of residence.
Variables for sexual behavior included age at first sexual encounter with a male (≤14, 15–17, 18–24, and 25+ years); forced sex at first sexual intercourse (yes/no); partner types (regular [lover/boyfriend or live-in partners], paying [respondent receives cash or kind from a partner for selling sex], paid [respondent pays cash or kind to a partner when buying sex], and casual [those other male partners besides the regular partner who does not pay the respondent for sex]); consistent condom use (defined as use of condom at each sex act, every time, in the past 12 months) with each of these different partners (yes/no); had sex with male/Hijra partner during travel outside their current place of residence (yes/no); and having reported to have had an STI (yes/no).
Other key independent variables included substance use (alcohol and drugs [yes/no]); experience of physical and sexual violence (yes/no); stigma and discrimination (yes/no); self-risk perception (high, moderate, low, and no risk); HIV/AIDS knowledge and program exposure including HIV testing and collection of test results (yes/no); comprehensive knowledge of HIV (dichotomized [yes/no] and defined as (i) knowing two major ways of preventing the sexual transmission of HIV [using condoms and limiting sex to one faithful, uninfected partner], (ii) rejecting two most common local misconceptions about HIV transmission, and (iii) being aware that a healthy-looking person can be infected with HIV); awareness about medication for prevention of mother-to-child transmission of HIV (yes/no); exposure to any prevention interventions (yes/no); and comprehensive exposure to interventions (dichotomized and includes those who received all of the four services, i.e., information on STI/HIV/AIDS from peer educators or outreach workers, condoms from peer educators or outreach workers, checkup and counseling for STIs, and referral to other services [i.e., STI clinics, HIV testing, and detox centers]).
Factors on community collectivization included membership in a self-help group, membership in any MSM/TG collective (yes/no), perception of H/TG on participation of other H/TG for problem resolution (all H/TG will work together, most, some or none will work together and don't know), and negotiated with/stood up against: (i) police (yes/no), (ii) goons/local leaders (yes/no), (iii) fellow H/TG or MSM (yes/no) to support/help other H/TG members in the community.
Data gathered from the seven selected states were combined, and national weights were calculated and used for the whole data analysis. Multivariable binary logistic regression analysis was undertaken using SVY commands in STATA 13 (Stata Statistical Software: Release 13. College Station, TX: StataCorp LP) to identify factors independently predictive of HIV infection. Adjusted odds ratios (AORs) and 95% confidence intervals (95% CIs) were generated. Factors significantly associated with HIV infection (P < 0.05) in the bivariate analysis were included in the final model along with known confounders, including age, education, occupation, marital status, self-identity, and state of residence irrespective of their significance levels. Due to the skewed distribution, except for the state variable, all others were dichotomized for the multivariable analysis. Missing cases, no response options, or do not know responses were removed.
| Results|| |
HIV prevalence in the samples (3325) drawn from seven states was 9.5% (CI: 7.8–11.5). The prevalence was highest in Maharashtra (17.3%, CI: 13.4–22.1), followed by Odisha (9.1%, CI: 5.9–13.9), Tamil Nadu (8.1%, CI: 2.2–25.5), Karnataka (6.1%, CI: 4–9.0), NCT of Delhi (5.3%, CI: 3.2–8.6), Andhra Pradesh (5.2%, CI: 3.4–7.8), and West Bengal (4.7%, CI: 2.8–7.9). Comparison of HIV status at the bivariate level revealed that a higher proportion of HIV-positive H/TG people were 35 years or older (25.9% vs. 15.0%, P = 0.010) and engaged in sex work (46.4% vs. 43.6%, P = 0.016) and traditional occupations (19.6% vs. 11.8%, P = 0.016) than HIV-negative H/TG people who were younger or mainly involved in other occupations. Similarly, a higher share of HIV-positive than HIV-negative H/TG people reported having a regular partner (59.8% vs. 48.8%, P = 0.022) [Table 1]. This was also the case of H/TG people who were aware of ART (76.2% vs. 67.2%, P = 0.025), had a comprehensive knowledge of HIV (63.7% vs. 54.8%, P = 0.041), and were aware of medications for the prevention of mother-to-child transmission of HIV (73.6% vs. 63.9%, P = 0.043). Further, a higher proportion of HIV-positive than HIV-negative H/TG people were members of self-help groups (49.1% vs. 32.8%, P = 0.001); believed that if there was a problem in the H/TG community, all will work toward its resolution (67.6% vs. 52.9%, P = 0.003); and in the past 12 months negotiated with/stood up against the police (51.4% vs. 36.3%, P = 0.002), goons/local leaders (36.7% vs. 28.3%, P = 0.049), and fellow MSM or H/TG (71.0% vs. 61.9%, P = 0.025) to help/support other H/TG community members [Table 2].
|Table 1: Distribution of profile characteristics and sexual behavior among Hijra/transgender persons according to HIV status, Integrated Biological and Behavioural Surveillance, 2014-2015*|
Click here to view
|Table 2: Distribution of contextual risk and vulnerability factors among Hijra/transgender persons according to HIV status, Integrated Biological and Behavioural Surveillance, 2014-2015*|
Click here to view
Multivariable analysis found that higher odds of HIV positivity if H/TG people had a regular male partner (AOR: 1.81, CI: 1.07–3.06), were living in the state of Maharashtra (AOR: 6.08, CI: 3.02–12.22) or Odisha (AOR: 2.91, CI: 1.05–8.06), and were members of self-help groups (AOR: 2.08, CI: 1.04–4.14). In contrast, odds of positivity were lower if H/TG people had travelled outside their current place of residence and had sex with male/Hijra partner at the place they visited last (AOR: 0.59, CI: 0.35–0.99) [Table 3].
|Table 3: Association of HIV seropositivity with profile characteristics and other contextual risk and vulnerability factors among Hijra/transgender persons, Integrated Biological and Behavioural Surveillance, 2014-2015*|
Click here to view
However, demographic or behavioral correlates of risk, such as education, marital status, condom use, alcohol/drug use, stigma and discrimination, violence, self-perceptions of HIV risk and HIV testing, and collection of results, were not found to be significantly associated with HIV infection. Neither was an association found with variables related to prevention program exposure including having received one or more HIV/AIDS services or a comprehensive package of services.
| Discussion|| |
HIV prevalence in H/TG people at 9.5% in seven high prevalence states confirms that this group continues to be highly affected by the epidemic in India. Earlier surveys recorded even higher HIV prevalence levels among H/TG people across India ranging from 9.8% to 45.2%. Maharashtra was already known to have very high HIV prevalence in this group, but higher odds of infection in H/TG people in Odisha compared to those in other states are unexpected.,, This finding calls for a stronger targeting of prevention interventions, especially at these locations where the epidemic is concentrated in H/TG people to avert new infections and provide adequate care and treatment.
Having a regular partner and membership in a self-help group were the only factors resulting in a higher likelihood of HIV infection. An ethnographic research conducted by Khan et al. in Bangladesh helps explain why H/TG people who have a regular partner may be more likely to get infected with HIV. For H/TG people who face a lot of stigma and discrimination in their community and society more at large due to their gender variant identity, having a regular male partner is of great importance as this is a source of moral and emotional support. The desire to have a sustained relationship with a man, similar to the one he would have with a woman, may lead a H/TG person to make concessions and neglect consistent condom use to please and maintain a regular partner., The practice by which H/TG people tend to engage in multiple concurrent partnerships, because if one partner leaves, he/she will have another partner whom he/she can rely upon, has also been documented. These aspects should be taken into consideration in the design of prevention intervention for H/TG people and their partners, both regular and casual.
In this study, H/TG people who reported being a member of a self-help group were twice as likely to be HIV-infected than nonmembers. This result could be due to sampling or because H/TG people who are at higher risk may be more likely to take part in self-help groups which could be the result of efforts made through targeted prevention programs. This finding needs to be further investigated including thorough analysis from programs and qualitative data to better understand realities on the ground. Self-help groups and community collectives have been created by the national program as a platform for key populations to discuss their needs and become involved in the development, implementation, monitoring, and evaluation of programs., That HIV-positive H/TG may be overrepresented in these groups and may have greater access to services and information than HIV-negative H/TG, would not be surprising. This could explain why the former were found to be better aware of HIV and of availability of AIDS treatment as shown by this analysis.
Besides membership in a self-help group, measures relating to problem resolution in the community and negotiation (i.e., with police and others who posed a threat to the community) were found to be associated with HIV infection in the bivariate analysis. This association, however, was not found to be significant in the multivariate analysis. Regardless, these factors, operating at the community level, seem to be important and should be examined more in depth in the future. Research should be conducted to better understand the role of community-level factors in influencing risk-taking tendencies among H/TG people (e.g., peer norms on safe sex, condom use, HIV testing, and STI treatment). Information on these aspects was not collected in the IBBS and therefore their impact on HIV infection unknown. Future bio-behavioral surveys should take community-level factors into consideration to enhance the understanding of HIV vulnerability and risk among H/TG.
Against expectations, like in earlier studies from southern India,, this multivariable analysis shows a lack of association of HIV infection with engagement in sex work, consistent condom use, HIV testing, substance use/abuse, stigma and discrimination, and violence. However, key factors of HIV risk and vulnerability should not be underestimated. The analysis found that nearly half of the samples from the seven selected states reported sex work as their main occupation. Around one in three of HIV-positive H/TG people indicated using condoms consistently with paying/paid partners. Keeping consistent condom at a high level is a major priority, especially among H/TG people and people who inject drugs who are at highest risk for HIV. Similarly, there is a need to scale-up initiatives aimed at reducing their vulnerability due to alcohol use, as IBBS results show that such consumption is high before or during sex act. Sexual and other violence, stigma and discrimination have to continue to be addressed, although not found to be directly associated with HIV infection in the logistic regression analysis. Such measures are in any case important to reduce H/TG people's vulnerability and ensure their adequate access to vital services including HIV testing and care and treatment.
This study has some limitations which are mainly related to the design and implementation of the 2014–2015 IBBS. The selection of domains in the IBBS was not done randomly; hence, the sample may be somewhat biased as is suggested by the analysis of results. Four hundred H/TG people were to be recruited in each domain, but in some, this sample size could not be achieved. This has impacted statistical significance of results, as smaller sample sizes are sufficient for behavioral estimates, but do not provide robust enough estimates of HIV prevalence. As the IBBS was conducted only in select geographical areas, its results cannot so easily be generalized to the whole country. Further, analysis of behaviors is largely based on self-reported data, which may suffer from social desirability bias. Long recall periods used in some questions may also have affected the quality of responses. Investigators put in place measures to address these biases from the outset of the survey, including through rigorous training of field enumerators and other measures aimed at ensuring best possible survey results.
| Conclusion|| |
The findings of this study have significant implications for the H/TG interventions in the next phase of India's NACP. The prevalence of HIV in this group remains the second highest nationally and is alarmingly high in some states such as Maharashtra. Thus, H/TG need to continue to be a major target of prevention efforts in India. Given the complex interaction of multilevel risks in and across these groups, and absence of an association between known proximate determinants and HIV infection, additional factors predictive of HIV status at community and structural level will need to be studied to inform prevention policies and programs.
The authors wish to thank the researchers/staff of the national and regional institutes (AIIMS, New Delhi; ICMR-NIMS, New Delhi; ICMR-NARI, Pune; ICMR-NIE, Chennai; ICMR-NICED, Kolkata; PGIMER, Chandigarh; and RIMS, Imphal) and State AIDS Control Societies for their relentless support during the different phases of the IBBS. Finally, we are grateful to the respondents who took their time and participated in the survey.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
World Health Organization. Policy Brief: Transgender People and HIV. Geneva, Switzerland: World Health Organization; 2015. p. 1-28.
Poteat T, Scheim A, Xavier J, Reisner S, Baral S. Global epidemiology of HIV infection and related syndemics affecting transgender people. J Acquir Immune Defic Syndr 2016;72 Suppl 3:S210-9.
Baral SD, Poteat T, Strömdahl S, Wirtz AL, Guadamuz TE, Beyrer C. Worldwide burden of HIV in transgender women: A systematic review and meta-analysis. Lancet Infect Dis 2013;13:214-22.
Mayer KH, Grinsztejn B, El-Sadr WM. Transgender people and HIV prevention: What we know and what we need to know, a call to action. J Acquir Immune Defic Syndr 2016;72 Suppl 3:S207-9.
Poteat T, Wirtz AL, Radix A, Borquez A, Silva-Santisteban A, Deutsch MB, et al
. HIV risk and preventive interventions in transgender women sex workers. Lancet 2015;385:274-86.
Chakrapani V. HIV and STI Prevalence, Vulnerability and Sexual Risk among Hijras and Other Male-to-Female Transgender People in India: A Research Synthesis and Meta-Analysis. New Delhi: UNDP India; 2014.
Chakrapani V, Newman PA, Shunmugam M, Logie CH, Samuel M. Syndemics of depression, alcohol use, and victimisation, and their association with HIV-related sexual risk among men who have sex with men and transgender women in India. Glob Public Health 2017;12:250-65.
Ganju D, Saggurti N. Stigma, violence and HIV vulnerability among transgender persons in sex work in Maharashtra, India. Cult Health Sex 2017;19:903-17.
National AIDS Control Organization. HIV Sentinel Surveillance 2014-15: A Technical Brief. National AIDS Control Organization, Ministry of Health & Family Welfare, Government of India; 2015.
National AIDS Control Organization. HIV Sentinel Surveillance 2016-17: Technical Brief. Delhi: National AIDS Control Organization, Ministry of Health and Family Welfare; 2017.
National AIDS Control Organization. HIV Sentinel Surveillance 2010-11: A Technical Brief. National AIDS Control Organization; 2011.
National AIDS Control Organization. National Integrated Biological and Behavioural Surveillance (IBBS) 2014-15: High Risk Groups National AIDS Control Organization, Ministry of Health & Family Welfare, Government of India; 2015.
National AIDS Control Organization. National Integrated Biological and Behavioural Surveillance (IBBS), Hijra/Transgender people, India 2014-15. New Delhi: National AIDS Control Organization, Ministry of Health & Family Welfare; 2016.
Thilakavathi Subramanian VC, Selvaraj V, Noronha E, Narang A, Mehendale S. Mapping and size estimation of Hijras and other trans-women in 17 states of India:First level findings. Int J Health Sci Res 2015;5:1-10.
National AIDS Control Organization. Operational Guidelines for Implementing Targeted Interventions among Hijras and Transgender People in India. National AIDS Control Organization; 2015.
United Nations Development Programme. Hijras/Transgender Women in India: HIV, Human Rights and Social Exclusion. United Nations Development Programme; 2010.
Hijras KG. The unique transgender culture of India. Int J Cult Ment Health 2012;5:121-6.
Khan SI, Hussain MI, Parveen S, Bhuiyan MI, Gourab G, Sarker GF, et al
. Living on the extreme margin: Social exclusion of the transgender population (Hijra) in Bangladesh. J Health Popul Nutr 2009;27:441-51.
Operario D, Nemoto T, Iwamoto M, Moore T. Unprotected sexual behavior and HIV risk in the context of primary partnerships for transgender women. AIDS Behav 2011;15:674-82.
Shaikh S, Mburu G, Arumugam V, Mattipalli N, Aher A, Mehta S, et al
. Empowering communities and strengthening systems to improve transgender health: Outcomes from the Pehchan programme in India. J Int AIDS Soc 2016;19:20809.
Shaw SY, Lorway R, Bhattacharjee P, Reza-Paul S, du Plessis E, McKinnon L, et al
. Descriptive epidemiology of factors associated with HIV infections among men and transgender women who have sex with men in South India. LGBT Health 2016;3:292-9.
Saravanamurthy PS, Rajendran P, Priya MM, Ashok G, Sai RS, Julia AH, et al
. A cross-sectional study of sexual practices, sexually transmitted infections and human immunodeficiency virus among male-to-female transgender people. Am Med J 2010;1:113-9.
Chakrapani V, Newman PA, Shunmugam M, Dubrow R. Barriers to free antiretroviral treatment access among kothi-identified men who have sex with men and aravanis (transgender women) in Chennai, India. AIDS Care 2011;23:1687-94.
[Table 1], [Table 2], [Table 3]