Background
Transgender people are disproportionately affected by HIV. The term “transgender” refers to a diverse range of people whose gender identities differ from the sex assigned to them at birth [
1,
2]. A recent review by Baral et al. shows that transgender women have a pooled HIV prevalence of 19.1% [
3], which is a 49-fold increased odds of HIV infection compared with non-transgender adults of reproductive age [
2].
An increasing amount of epidemiological research has highlighted factors that increase HIV vulnerability among transgender people. Gender-based sexual violence that transgender people experience due to their gender identity [
4] increases risk of HIV infection [
5]. Stigma prevents access to HIV services and care [
6], and is often associated with inconsistent condom use in anal intercourse and sex work among transgender women [
7,
8]. Coinfection with perianal and other sexually transmitted infections (STI) could also facilitate acquisition and transmission of HIV [
2,
9]. In addition, recent studies suggest ways in which gender affirming hormone use could impact HIV vulnerability [
2]. Furthermore, sex work is often associated with higher prevalence of HIV among transgender people [
10].
Whereas evidence showing the high vulnerability to HIV of this population is increasing, transgender people are generally underserved in national HIV responses [
11]. For instance, the majority of countries do not include transgender in the national HIV plans and strategies [
3]. This is often confounded by lack of adequate data related to transgender people in many countries [
2]. Population-based representative studies and routine surveillance among transgender people are generally limited globally [
12]. In addition, there is a paucity of data to guide clinical care of transgender people on antiretroviral and gender-affirming hormone therapies [
13]. Furthermore, transgender people are often subsumed under men who have sex with men (MSM) in many countries [
11,
14], which makes it difficult to provide tailored responses to their behaviors and vulnerabilities distinct from those of MSM [
7,
11]. As a result, HIV programs are not effective for this population [
7].
Research and dissemination of findings related to transgender people are required to inform policy and programming for these particularly vulnerable people. In this article, we present descriptive findings from a large-scale national integrated biological and behavioral survey among transgender women in Cambodia.
Methods
Study design, settings, and participants
The cross-sectional study was conducted between December 2015 and February 2016 in 13 major sites (1 capital city and 12 provinces) of Cambodia. Respondent Driven Sampling (RDS) method was used to recruit participants with support from community-based implementing partners of KHANA, the largest national NGO providing community-based HIV prevention, care, and support services in Cambodia [
15]. Among the 13 study sites, participants were recruited in 20 specific locations (six locations in Phnom Penh and 14 locations in the provinces). The number of the selected locations was determined based on the proportion of the required sample size in each study site, and the estimated population of transgender women in each site. Eligible participants were individuals who: (1) were biologically male at birth and self-identified as a woman, (2) were aged 18 years or over, (3) reported having sex with at least one man in the past 12 months, (5) could speak Khmer, and (6) were able and willing to provide written consent to participate in the study.
Four initial seeds were identified at each site by outreach workers from KHANA’s implementing partners based on age (two seeds aged 18 to 24 and the other two seeds aged 25 or older). These seeds had to meet the eligibility criteria for participation and have an established and large social network comprising about 10 or more other transgender women in their given location. Eligibility to participate as a seed was determined by the leader of data collection team using a paper-based eligibility form. Each seed was given three coupons and asked to refer three additional transgender women. US$2 was given to each seed for a successful referral. Each seed was expected to extend to 3–6 “recruitment waves” in each site. If the initial seeds did not recruit participants or if enrollment was halted because all recruitment chains had “dried up” (i.e. stopped recruiting), additional seeds were selected based on the above criteria. In total, 80 seeds were selected by the implementing partners’ outreach workers initially.
Data collection and training
Data were collected by two teams with eight personnel each that included one field supervisor, five interviewers, one lab technician, and one counselor from the Provincial AIDS and STI Program. The field supervisor conducted eligibility screening of the participants. Each consenting participant was assigned a unique personal identification number, which was used to link all data collected from each participant. The unique personal identification number was not linked with any personal data to protect confidentiality. The counsellor then explained objectives of the study in details, including the process of HIV testing, potential risks and benefits of participation, and obtained written informed consent. This was followed by HIV counselling conducted by the counsellor, and HIV testing conducted by the lab technician. Finger-prick-based HIV testing was conducted using Determine™ test, according to the national protocol [
16]. HIV testing and counseling were performed in accordance with all applicable national guidelines.
After the HIV test was conducted, an interviewer administered the questionnaire in a private room using an Android tablet. At the end of the interview, the counsellor provided the HIV test result and post-test counselling. Participants with a reactive result were referred to the nearest HIV clinic for a confirmatory test, and were invited to come back to the implementing partner site for ongoing support and follow-up services. The field supervisor then provided a coupon incentive to the participants. Each participant received US$4 in cash to compensate for their time, and a package of three condoms. This amount of incentive was determined reasonable and equivalent to a typical meal for one person and the cost of transportation.
Prior to data collection, all interviewers and field supervisors were trained for three days on data collection methods and tool pretesting and reflection, to safeguard consistency, quality, and validity of the data. The training included interview techniques, confidentiality, and privacy, and provided opportunities for the study team the opportunity to rehearse questionnaire administration and other study procedures. During data collection, review sessions with interviewers were conducted regularly to review progress and communicate any problems or issues which required solving.
Variables and measures
The questionnaire was developed using standardized and validated tools adapted from previous literature, the most recent Cambodia Demographic and Health Survey, and other studies among HIV key populations in Cambodia. It was initially developed in English and then translated into Khmer, the national language of Cambodia. Another translator then back-translated it into English to ensure that the “content and spirit” of every original item was maintained. Clear instructions and explanations were included to avoid any confusion during the interviews.
To design the study and develop the tools, consultative meetings were held with representatives of transgender women, community people, non-governmental organizations (NGOs), donor agencies, government officers, as well as researchers and practitioners working on HIV and AIDS in Cambodia. Prior to data collection, the questionnaire was pretested to ensure that the wording and contents of the questionnaire were culturally suitable, acceptable, and clearly understood by the study participants before it was finalized. In the pilot study, we conducted face-to-face interviews with 20 transgender women recruited from Phnom Penh to assess the contents, format, length, language, and appropriateness of the questionnaire. Necessary modifications were made based upon feedbacks from the pilot study and from the consultative meetings. The final version of the questionnaire was used for the main data collection.
The questionnaire collected the following information: (1) Socio-demographic characteristics; (2) transgender identity and related experiences; (3) sexual behaviors and condom use in different relationships; (4) accessibility to condoms and lubricants; (5) HIV/STI screening and care seeking behaviors; (6) substance use (alcohol, illicit drugs); and (7) experiences of stigma and discrimination in communities and health facilities.
Data management and analyses
Data from the questionnaires collected via Android tablets were transferred onto a secure server after completion of each day of data collection. After synchronization, the data were cleared from the tablet. At the conclusion of the two-month data-collection, the data were downloaded from the secure server, tabulated, and transcribed into a Microsoft Excel format. Data were then imported into STATA (Version 12.0) for analyses. Prior to the analyses, data were weighted to account for the effect of RDS method. Descriptive analyses were conducted to calculate the counts and proportion (%) for categorical variables and means with standard deviations (SD) for continuous variables. Where applicable, Chi-square test or Student’s t-test was used to compare between groups.
Ethical considerations
Participants were required to provide written consent after being informed in details about the study objectives, risk, and benefits. Participants were informed that they could withdraw from the study at any time. Interviews and HIV testing and counseling were conducted at private locations within drop-in centers, private houses, or offices of KHANA’s implementing partners, and confidentiality was ensured by assigning a personal identity number (PIN) to each participant and removing all personal identifiers. The study protocol was approved by the National Ethics Committee for Health Research (NECHR) of the Ministry of Health, Cambodia (No. 420 NECHR) and FHI 360’s Protection of Human Subjects Committee (PHSC No. 713897).
Discussion
There is growing interest in understanding the epidemiology of HIV among transgender people. This study set out to explore demographic, behavioral and health-related characteristics of transgender women in Cambodia. Results showed that the prevalence of HIV was 5.9% among the sample. In addition, the results provide useful data on high-risk sexual behaviors, STI symptoms, substance use, HIV testing, discrimination experiences, and willingness to use PrEP among this population.
The HIV prevalence found in our study is slightly higher than the 4.2% found by a recent study by Weissman et al. [
17], but both are significantly higher than an earlier study conducted in 2010 which found a prevalence of 2.6% among transgender women [
18]. However, direct comparison of findings from this study with those in the previous ones is difficult because of the differences in characteristics of the samples and measures. Our study was conducted in 13 sites (one city and 12 provinces), while the study by Weissman et al. was conducted in six sites [
17] and the Bros Khmer study in seven sites [
18]. The timeframes used to measure key variables, such as HIV testing and condom use, were also not consistent. Studies from Cambodia have consistently reported higher levels of HIV prevalence among transgender people compared to MSM [
17,
18], underscoring the importance of separating transgender people from MSM so as to have an accurate picture of HIV epidemic among this group.
Since 2013, transgender women have been recognized as a separate population from MSM within the Standard Operating Procedures of Boosted Continuum of Prevention, Care and Treatment [
19]. However, even within transgender people, differences were noted in HIV prevalence based on residence. The prevalence was significantly higher among transgender women in urban communities compared to those in rural areas, which suggests the need to better focus on cities and urban areas. These findings also buttress calls for geographic localization of HIV prevention efforts [
10], given that HIV rates were much higher in Banteay Meanchey, Siem Reap, Phnom Penh, and Battambang provinces compared to other provinces.
An important implication from the study was related to combination prevention. The majority of the participants were sexually active, with 97.9% reporting having intercourse with men, mostly anal sex. However, a significant number of these did not use condoms consistently. Given the high prevalence of HIV in this group, these findings suggest that expansion of HIV prevention interventions for this group is warranted. This is particularly relevant given that two thirds of the participants had not heard of PrEP, but 88.9% were willing to use it. Combined with the high proportion (94.9%) of HIV positive participants who were on ART, these findings suggest that provision of PrEP could reduce incidence of HIV among this population. The high prevalence of sex work in exchange of money among this sample provides another reason to consider PrEP for transgender sex workers. Other studies have highlighted the heightened vulnerability and the needs for better tailored interventions for transgender sex workers [
20,
21].
The low rate of HIV testing observed among transgender women in this study is of concern and suggests that significant effort is required to increase HIV testing uptake. In particular, HIV education and benefit of HIV testing should be widely promoted among this population. The lack of HIV awareness could influence HIV sexual behaviors, including the use of condoms, or commercial sex [
22]. In addition, HIV testing is the first step towards early access to treatment, which also has high preventive benefits. Results from this study indicate that participants were willing to consider HIV self-testing, which could provide additional flexibility in access to testing. HIV self-testing has potential to reach individuals who are least likely to seek testing in a public clinic, hospital, voluntary confidential counseling and testing center (VCCT), or with a community-based non-governmental organization. Results from other settings indicate that self-testing may be acceptable to transgender people [
23] and should be made available while considering issues related to confirmation and linkage to care, cost [
24], quality assurance, as well as support needs of different potential users of self-tests [
25].
Our study indicates that 6.5% of the participants reported having sex during or after using illicit drugs, so called “Chem sex”, which may increase sexual risk taking [
26,
27]. Furthermore, 45% of the participants had ever used gender affirming hormones, which included through injections in 18% of the cases. Our results also indicate that the injecting practices were often unsafe (by self-injection or non-skilled providers) and included sharing of needles. Therefore, HIV prevention efforts should explore provision of PrEP and education regarding the use of drugs during sex, while considering how safe needle exchange interventions can be integrated for those who inject drugs or hormones.
Our study shows the importance of community-based outreach services in increasing access to HIV prevention services. Peers and outreach workers were the sources of condoms for 70% of the participants. In addition, findings related to STI suggested that the majority of the participants that experienced STI symptoms shunned public facilities as a source of treatment, with the majority seeking health care from NGO or private facilities and pharmacies. Furthermore, the majority of participants accessed HIV testing from community-based NGOs. These preferences may be related to the demand creation, peer-support building or less stigmatizing nature of community-based organizations and services in the study context. Evidence from other contexts indicate that when services, often community-based, are non-stigmatizing, gender-affirming, trans-specific and leveraged on local peers and resources, transgender people access them [
28‐
30].
Conversely, these findings highlight a need to better understand how the capacity of public health facilities to provide friendly services to transgender people can be strengthened. In 2014, the World Health Organization released a global guidance on HIV programming for key populations, including transgender people [
31]. More recently, implementation tools developed to assist the provision of comprehensive community-led responses for transgender people have been launched [
32]. In Cambodia, these guidelines and implementation tools need to be adapted and implemented.
Stigma and discrimination are powerful drivers of inequitable access to HIV and health services among transgender [
6,
33] and other [
34] populations. Yet HIV cannot be eliminated unless transgender persons can have universal access to HIV and health services. Our study shows that a significant number of transgender women experienced discrimination based on their transgender identity, including problems getting services from an HIV prevention program (8.5%) and other health and medical services (9.1%). In a similar Cambodian study, 54.8% of participants reported having experienced discrimination related to transgender identity in their lifetime [
17]. Consequently, significant effort will be required to sensitize, and train health care providers so that they provide appropriate care with non-judgmental attitudes to transgender women. General practitioners, nurses and non-clinical health professionals working in public hospitals and HIV clinics need to be equipped with appropriate clinical skills including clerking, counselling, and examination skills so as to provide appropriate care to transgender women. This is particularly important because 39.2% of the participants reported experiencing sexual abuse or assault, which can increase HIV vulnerability and requires access to health and medical services.
More broadly, the intersectionality of multiple deleterious determinants of health, including drug use or abuse, violence or stigmatization, poor mental health, internalized transphobia, and economic hardships may potentiate HIV vulnerability [
35], reduce uptake of services [
34], and contribute to poor HIV treatment outcomes [
2] among different populations, including transgender people. Hence, multi-sectoral interventions to mitigate these complex interacting syndemic factors in an integrated fashion are required. Importantly, this approach should include sectors beyond physical health, such as mental health, legal, policing, and social protection sectors, to mitigate arbitrary arrests [
6,
36], lack of employment [
37], criminalization, victimization [
38], violence [
5,
38], and other structural drivers of vulnerability to HIV among this population.
Findings from this study should be cautiously interpreted in light of the design, recruitment, methods of data collection, and sample profile. As this was a cross-sectional study, it captures a snapshot view of the population studied, and may not document changes in measured variables over time. Our study sample had relatively educated participants, which may not be representative of all transgender women in the country, and were initially recruited from those who had some contact with outreach workers. Future studies should be more cautious about selecting seeds that are more representative of sub-populations of transgender women by considering their socio-demographic characteristics such as education levels, occupations, incomes, etc. Sexual behaviors and other sensitive information were collected from the participants through a self-reported questionnaire, which is often at high risk of bias [
39]. Despite these limitations, this paper provides important information that can be built upon to inform programming and policy.
Acknowledgements
This study was financially supported by the United States President's Emergency Plan for AIDS Relief (PEPFAR)/the United States Agency for International Development (USAID) through HIV/AIDS Flagship Project. We thank all data collection team members, KHANA’s implementing partners, community support volunteers, outreach workers, and all participants for their excellent contribution to this study. Disclaimer: Content of this paper is the responsibility of the authors and does not reflect the view of USAID or our respective institutions.