Willingness to Use HIV Self-Testing and Associated Factors Among Transgender Women in Malaysia

Abstract

Background: While global research on the implementation of HIV self-testing (HIVST) has increased exponentially, few studies have assessed information on HIVST uptake factors (i.e., willingness, preferences) in transgender women (TW) and none in the Malaysian context. We therefore sought to assess willingness to use HIVST among this understudied key population.

Methods: A cross-sectional survey was conducted among 361 HIV-negative Malaysian TW in 2017. Participants were recruited using flyers, announcements through social media, and direct referral from staff members of the community-based organizations serving the TW communities. Multivariable logistic regression was used to identify correlates of willingness to use HIVST.

Results: Nearly half of the participants (47.6%) reported that they would be willing to use HIVST. Multivariable analysis showed that willingness to use HIVST was positively associated with having experienced sexual assault in childhood (adjusted odds ratio [aOR]=3.239, p<0.001), having ever used mobile phone or apps to find sex work clients (aOR=1.677, p=0.040), and having engaged in condomless sex in the past 6 months (aOR=1.886, p=0.018). In contrast, living in Kuala Lumpur (aOR=0.559, p=0.032), having higher number of sex work clients per day (aOR=0.927, p=0.004), and current depressive symptoms (aOR=0.576, p=0.026) were negatively associated with willingness to use HIVST.

Conclusions: Findings in this study suggest that TW in our sample were moderately willing to use HIVST. Especially important here is the increased interest in HIVST among TW who are at higher risk for HIV infection. Overall, our findings underscore the need for additional research on how to most effectively implement HIVST for key populations, including TW, such that uptake and retention in regular HIV screening is sustained.

Introduction

Globally, transgender women (TW) are at significantly higher risk for HIV than people in the general adult population, with an estimated 19.1% (95% CI [confidence interval]: 17.4–20.7) of all TW currently living with HIV.¹ Similarly high rates of HIV in TW have been documented in the Asia-Pacific region.^2,3 In Malaysia, a Southeast Asian country of over 30 million people, an estimated 10.9% of all TW are HIV infected,⁴ over 28-fold higher than the 0.4% national HIV prevalence in the general adult population.⁵ Furthermore, recent data from a respondent-driven sampling study of TW sex workers in Greater Kuala Lumpur found an HIV prevalence of 12.4%.⁶ The high burden of HIV among TW reflects the complex and multilevel HIV vulnerabilities they face.

Worldwide, being TW is almost universally stigmatized, resulting in increased discrimination, gender-based violence and abuse, marginalization, and social exclusion that can limit TW's access to vital health services and placing them at greater risk for HIV.^7,8 Anti-transgender discrimination also results in TW having fewer economic opportunities, steering many into high-risk occupations such as sex work.⁹ Structural factors, including civil and religious laws that criminalize being transgender also elevates TW's HIV risk.¹⁰ In Malaysia, TW who are incarcerated in jails or prisons are often housed in male-only units or are placed in solitary confinement, leaving them vulnerable to physical and sexual abuse by other inmates and correctional staff.¹¹ Other structural barriers include laws that prohibit medical doctors from providing gender-affirming health services, such as hormone therapy (HT) and gender reassignment surgery, which has been illegal in Malaysia since 1982.¹¹ Anti-transgender stigma by medical doctors in Malaysia may result in a hostile environment for TW, resulting in reduced access and utilization of treatment and prevention services.¹²

HIV testing is the gateway to a range of HIV prevention and treatment options, such as pre-exposure prophylaxis (PrEP) for those testing negative and antiretroviral therapy (ART) for those who test positive. Moreover, HIV testing is the first critical step in the UNAIDS 90-90-90 strategy to eliminate new HIV infections. However, uptake of HIV testing services among key populations in Malaysia, including TW, remains low.^13–15 For example, a recent study in Kuala Lumpur, Malaysia's capital, found suboptimal rates of both lifetime (41.7%) and recent HIV testing (18.7%) among TW.¹³ Barriers to HIV testing services include multilevel factors such as stigma and discrimination, lack of anonymity, concerns about confidentiality, limited access to HIV care services, and mistrust of medical institutions.^16–18 TW face additional legal burdens, with frequent detention and prosecution under Malaysia's Islamic legal code for engaging in so-called “cross-dressing”—the wearing of stereotypically women's clothing in public, thus, resulting in a stigmatized population hidden from the public space.¹¹ Novel approaches to increase TW's linkage and retention to HIV testing are urgently needed.

One such strategy is HIV self-testing (HIVST), which offers an empowering and innovative way to improve HIV testing among marginalized and high-risk groups with limited access to traditional venue-based HIV care services.¹⁹ In 2016, the World Health Organization (WHO) recommended the scale-up of self-testing as an additional alternative to conventional HIV testing services.²⁰ With the characteristics of convenience, privacy, and confidentiality, HIVST may be particularly important in communities of TW who face high rates of stigma and discrimination. In addition to reaching individuals who have never tested for HIV or those who test infrequently, HIVST empowers users to test on their own, which normalizes screening and can facilitate partner testing.^19,21 These benefits of HIVST highlight the potential contribution of HIVST to close critical gaps in global HIV testing coverage and to help achieve the first of the UNAIDS 90-90-90 treatment goals—for 90% of all people with HIV to know their HIV status.²²

While global research on the implementation of HIVST has increased exponentially, few studies have explored self-testing in TW^23–26 and none in the Malaysian context. Currently, there is no licensed or approved HIVST kit available to consumers in Malaysia. Unregulated HIVST kits, often sold in unmarked packaging, can frequently be found for sale online. Although kits can sell for as little as 20 Malaysian Ringgit (∼$5 USD), there is no oversight in place to ensure that the kits are properly stored, packaged, and shipped—or even that the kits are testing for HIV antibodies. As the Malaysian Ministry of Health (MoH) has recently called for the development of national HIVST policy to improve uptake of HIV testing among key populations, information on HIVST uptake factors (i.e., willingness, preferences) will be crucial for the implementation of HIVST services for TW in the country. Therefore, we sought to directly assess willingness to use HIVST among this highly stigmatized and difficult-to-reach group.

Methods

Study design and participants

The data reported in this study are derived from a cross-sectional study of TW in Malaysia conducted in 2017. The original study was designed to explore acceptability of PrEP among TW, however, other health-related topics, including willingness to use HIVST, were also explored. Inclusion criteria were: (1) age of 18 years or older; (2) male sex assigned at birth and reporting gender identity as female or identified as a TW; (3) living in Malaysia; and (4) self-reported HIV-negative or HIV status unknown. Before conducting this survey, formative interviews were conducted with leaders from the Malaysian TW community to help identify regions around Malaysia with established communities of TW. Based on their feedback, we recruited TW in three out of nine states across peninsular Malaysia, including Selangor (inclusive of the Greater Kuala Lumpur region), Penang, and Seremban.

Study procedures

A total of 381 individuals were screened for study eligibility. After screening, 7 were identified as ineligible and 13 declined to participate, yielding a final analytic sample of 361. Participants were recruited using convenience sampling procedures, including posting of flyers in venues frequented by TW, announcements through social media, and by direct referral from staff members at community-based organizations serving the TW community. Participants were recruited and interviewed in three states of Malaysia, including Selangor (inclusive of the Greater Kuala Lumpur region), Penang, and Seremban. Screening was conducted in a private room by a trained research assistant who was also a member of the TW community. Individuals who met inclusion criteria and expressed an interest in participating were then guided through the informed consent procedures. After providing consent, participants completed a self-administered, anonymous survey using a tablet or laptop computer. The study survey was programmed into Qualtrics (Qualtrics, Provo, UT) online survey software and took ∼20 min to complete. All study materials were translated from English to Bahasa Malaya by trained translators. Back translation was conducted to ensure accuracy of all study materials. Participants were paid 20 Malaysian Ringgit (∼$5 USD) for their time. The study protocol was approved by the Institutional Review Boards of Yale University and the University of Malaya.

Measures

Primary outcome

The primary outcome variable, willingness to use HIVST, was measured using a single-item question, “Would you be willing to use HIV self-testing?” with a binary response option of “yes” or “no.” Before answering this question, all participants read a brief description of HIVST, which explained what HIVST is and how it differs from traditional venue-based HIV testing.

Independent variables

Independent variables from the survey included participant characteristics, sexual risk behaviors, drug and alcohol use, health measures, criminal justice involvement, and physical and sexual victimization items. Variables were selected based on a review of the relevant literature and knowledge of the Malaysian context.

Participant characteristics included age, ethnicity (Malay vs. other), religion (Muslim), relationship status (single), educational status (high school graduate), income, and living status (alone vs. with other people; living in Kuala Lumpur).

Sexual risk behaviors included engagement in sex work (“In the last 6 months, have you had sex with someone in exchange for money?”), age of first sex work (“How old were you when you first engaged in sex work?”), number of sex work clients, use of mobile phone or mobile apps to solicit sex work clients (“Have you ever used a mobile phone or mobile app to find clients for sex work?”), and engagement in condomless sex (CS) in the past 6 months (“In the past 6 months, have you had anal or vaginal sex without using a condom?”).

Drug and alcohol use

Drug and alcohol use measures included recent (past 30 days) use of amphetamine-type stimulants (ATS), defined as the use of any amphetamine, ecstasy/MDMA (3,4-methylenedioxymethamphetamine), or other psychostimulants, and recent use of alcohol (past 30 days). Prior injection of illicit drugs was also measured for the lifetime.

General and sexual health

General and sexual health measures were also included. Prior HIV testing was measured for lifetime and last 12 months. Prior sexually transmitted infection (STI) testing (i.e., syphilis, chlamydia, and/or gonorrhea) was measured for lifetime. Recent doctor visit was defined as having been examined by a medical doctor for any reason in the last 12 months. Current depression was screened using the 10-item Clinical Epidemiological Scale-Depression (CES-D).²⁷ Although the CES-D has not been validated for the Malaysian context, it is widely used in clinical settings for depression screening across Malaysia. Scores of ≥10 on the CES-D indicated moderate to severe depression symptoms (α=0.62). Recent use of HT was defined as any use of feminizing hormones in the last 90 days, including oral and injectable formulations.

Criminal justice involvement

Criminal justice involvement was measured using two variables, including previously in lock-up or jail and previously in prison (“Have you even been detained by the police, placed in lock-up, or imprisoned for any reason?”).

Violence and victimization

Experience of childhood physical and sexual violence and adulthood physical violence was measured using questions from the U.S. Centers for Disease Control and Prevention's Behavioral Risk Factor Surveillance System (BRFSS) questionnaire for violence and victimization.²⁸

Analyses

Analyses were performed using IBM SPSS 25.0. Estimates were evaluated for statistical significance based on 95% CIs and statistical significance of p<0.05. We computed descriptive statistics, including frequencies and percentages for categorical variables and means, standard deviations, and ranges for continuous variables. Bivariate logistic regression analyses were conducted between each of the independent variables and the primary outcome, willingness to use HIVST. Bivariate associations significant at p<0.10 were selected for inclusion in the multivariable logistic regression analyses. Childhood physical assault was excluded from the final model due to its high correlation with childhood sexual assault. The goodness of fit of the final model was assessed using the Hosmer–Lemeshow test.²⁹

Results

Participant characteristics

Table 1 shows participants' characteristics, stratified by willingness to use HIVST. Most participants were Malay (75.1%), high school graduates (69.3%), and had a monthly income of ≥MYR 1000 (86.1%). Regarding sexual risk behaviors, 269 (74.5%) participants had engaged in sex work in the last 6 months with an average of 3.5 sex work clients per day in the last month. Almost half (42.4%) of participants reported having used a mobile phone or an app to solicit clients for sex and one-third (31.0%) reported CS in the last 6 months.

Table 1.

Characteristics of Participants Stratified by Their Willingness to Use HIV Self-Testing (N=361)

Variables	Entire sample (N=361)		Willingness to use HIVST		OR (95% CI)	p
	n	%	No (n=189)	Yes (n=172)
Sociodemographic
Age (years)	35.3 (9.8)		36.1 (10.4)	34.5 (8.9)	0.984 (0.963–1.005)	0.134
Ethnic origin: Malay	271	75.1	142 (39.3)	129 (35.7)	0.993 (0.616–1.601)	0.977
Religion: Muslim	288	79.8	149 (41.3)	139 (38.5)	1.131 (0.675–1.894)	0.640
Relationship status: single	244	67.6	132 (36.6)	112 (31.0)	0.806 (0.518–1.253)	0.338
Completed high school	250	69.3	117 (32.4)	133 (36.8)	2.099 (1.322–3.331)	0.002
Income: ≥1000 MYR per month	311	86.1	162 (44.9)	149 (41.3)	1.080 (0.593–1.965)	0.802
Living alone	161	44.6	99 (27.4)	62 (17.2)	0.512 (0.336–0.782)	0.002
Living in Kuala Lumpur	245	67.9	139 (38.5)	106 (29.4)	0.578 (0.370–0.902)	0.016
Sexual risk behaviors
Engaged in sex work (last 6 months)	269	74.5	142 (39.3)	127 (35.2)	0.934 (0.582–1.500)	0.778
Age of first sex work	19.2 (5.6)		19.6 (6.0)	18.7 (5.2)	0.973 (0.932–1.016)	0.214
Number of sex work clients per day (last month)	3.5 (5.7)		4.2 (6.6)	2.7 (4.3)	0.949 (0.910–0.990)	0.016
Used mobile phone or mobile apps to find clients	153	42.4	67 (18.6)	86 (23.8)	1.821 (1.194–2.777)	0.005
CS (last 6 months)	112	31.0	42 (11.6)	70 (19.4)	2.402 (1.519–3.799)	<0.001
Drug and alcohol use
ATS use (last 30 days)	37	10.2	18 (5.0)	19 (5.3)	1.180 (0.597–2.330)	0.634
Alcohol use (last 30 days)	63	17.5	24 (6.6)	39 (10.8)	2.016 (1.155–3.520)	0.014
Drug injection behavior (lifetime)	10	2.8	4 (1.1)	6 (1.7)	1.672 (0.464–6.027)	0.432
General and sexual health
HIV tested (ever)	287	79.5	144 (39.9)	143 (39.6)	1.541 (0.915–2.594)	0.104
HIV tested (last 12 months)	221	61.2	114 (31.6)	107 (29.6)	1.083 (0.709–1.655)	0.713
Other STI tested (ever)	234	64.8	118 (32.7)	116 (32.1)	1.246 (0.808–1.924)	0.320
Seen by a doctor in last 12 months	307	85.0	162 (44.9)	145 (40.2)	0.895 (0.502–1.596)	0.707
Current depressive symptoms	202	56.0	116 (32.1)	86 (23.8)	0.629 (0.414–0.956)	0.030
Use of hormone (last 90 days)	200	55.4	94 (26.0)	106 (29.4)	1.623 (1.067–2.468)	0.024
Criminal justice history
Previously in lock-up/jail (lifetime)	132	36.6	63 (17.5)	69 (19.1)	1.340 (0.872–2.059)	0.182
Previously in prison (lifetime)	77	21.3	41 (11.4)	36 (10.0)	0.956 (0.577–1.583)	0.860
Violence and victimization
Childhood physical assault	135	37.4	51 (14.1)	84 (23.3)	2.583 (1.665–4.006)	<0.001
Childhood sexual assault	149	41.3	51 (14.1)	98 (27.1)	3.583 (2.306–5.568)	<0.001

Bold indicates statistical significance at p<0.05.

ATS, amphetamine-type stimulants; CI, confidence interval; CS, condomless sex; HIVST, HIV self-testing; OR, odds ratio; STI, sexually transmitted infection.

Almost four-fifth (79.5%) of TW reported having ever been tested for HIV, with 61.2% tested in the last 12 months. Furthermore, 64.8% reported prior STI testing and 56.0% met screening criteria for having moderate-to-severe depression. Regarding violence and victimization, over one-third reported having experienced physical assault (37.4%) and sexual assault (41.3%) in their childhood.

Willingness to use HIVST

Nearly half of participants (47.6%) reported that they would be willing to use HIVST. Among those willing to use HIVST, 85.5% preferred to receive the test results instantly, whereas 14.5% preferred to mail samples to a facility and receive results through mail. Table 1 shows the bivariate correlates of willingness to use HIVST. Completion of high school (odds ratio [OR]=2.099, p=0.002), living in Kuala Lumpur (OR=0.578, p=0.016), number of sex work clients per day in the last month (OR=0.949, p=0.016), use of mobile phone or mobile apps to find clients (OR=1.821, p=0.005), engagement in CS in the last 6 months (OR=2.402, p<0.001), current depressive symptoms (OR=0.629, p=0.030), and experience of childhood physical (OR=2.583, p<0.001) and sexual (OR=3.583, p<0.001) assault were significantly associated with willingness to use HIVST in bivariate analyses. Table 2 shows the independent correlates associated with willingness to use HIVST in the multivariable regression. Use of a mobile phone or apps to solicit sex work clients (adjusted OR [aOR]=1.677, p=0.040), having engaged in CS in the past 6 months (aOR=1.886, p=0.018), and having experienced sexual assault in childhood (aOR=3.239, p<0.001) were all associated with greater willingness to use HIVST, whereas living in Kuala Lumpur (aOR=0.559, p=0.032), having more sex work clients per day (aOR=0.927, p=0.004), and current depressive symptoms (aOR=0.576, p=0.026) were associated with lower willingness to use HIVST.

Table 2.

Multivariate Logistic Regression Analysis of Factors Associated with Willingness to Use HIV Self-Testing (N=361)

Variables	Willingness to use HIVST
	aOR	95% CI	p
Completed high school	1.589	0.942–2.680	0.082
Living alone	0.626	0.390–1.005	0.052
Living in Kuala Lumpur	0.559	0.329–0.951	0.032
Number of sex work clients per day (last month)	0.927	0.880–0.976	0.004
Used mobile phone or mobile apps to find clients	1.677	1.024–2.747	0.040
CS (last 6 months)	1.886	1.115–3.189	0.018
Alcohol use (last 30 days)	1.175	0.614–2.251	0.626
Current depressive symptoms	0.576	0.354–0.937	0.026
Use of hormone (last 90 days)	1.332	0.822–2.158	0.244
Childhood sexual assault	3.239	1.969–5.328	<0.001
Hosmer–Lemeshow test: chi-square=6.049; p=0.642

aOR, adjusted odds ratio.

Discussions

To our knowledge, this is the first study to assess willingness to use HIVST among TW in Malaysia, a key population with high vulnerability to HIV. Overall, our findings suggest that TW were moderately interested in HIVST, with 47.6% of our sample willing to use HIVST. Although this result may appear low relative to studies of HIVST interest in TW in the United States (68%),³⁰ Thailand (81%),³¹ Peru (95%),³² and Cambodia (100%),²⁵ our finding demonstrates that a solid base of TW would be willing to adopt HIVST as an alternative HIV testing modality. Given that participants in the current study were recruited from venues where HIV testing is routinely provided to TW (e.g., community-based organizations), it may be that these participants were already satisfied with their current venue-based HIV testing, as suggested by the higher rate of recent HIV testing in our sample. Nonetheless, a HIVST strategy that is patient centered and responsive to the needs of key populations is likely to be well received by patients in the Malaysian context. Ultimately, our results support that the need for Malaysia to rapidly develop national guidelines on HIVST to ensure key populations at high risk for HIV can access this alternative modality for testing, ultimately expanding and improving the uptake and frequency of HIV testing among groups with the greatest gaps in testing coverage.³³

Individuals' engagement in HIV risk behaviors has been associated with willingness to use HIVST among various risk groups.^34–36 Interestingly, we found that TW who engaged in a higher number of sex work clients per day were less willing to use HIVST. This result contrasts with prior research which found HIVST acceptability to be positively correlated with number of sexual partners.³⁷ Additional research has indicated that increasing a sense of agency or empowerment through HIVST may result in individuals changing the number of sexual partners.^38,39 This evidence does not, of course, explain why participants with more sexual partners are less willing to use HIVST, and thus warrants further research to disentangle this relationship.

Interestingly, TW who engaged in CS in the past 6 months reported greater willingness to use HIVST compared with TW who reported no recent CS. Those TW who are engaged in CS may be aware of their elevated HIV risk and, in response, are seeking alternative modes for HIV testing, such as HIVST. They may also view HIVST as a technology that could aid them in making informed decisions, given that knowing one's HIV status is imperative to HIV prevention. Prior research with men who have sex with men has shown that they use HIVST to screen sexual partners before engaging in CS,^36,40,41 which is also known as “enhanced serosorting.” It is, however, important for testers to be aware that most HIV rapid test kits, particularly those which employ antibody-only assays, have limited ability to detect acute HIV infection (infections occurring in the last 21 days).⁴² Should someone be in the acute stage of HIV infection, use of HIVST for “enhanced serosorting” may result in a false negative result, which could put their sexual partners at risk of infection. In the future, HIVST Kits may integrate assays that can detect the p24 antigen, which would improve early detection of HIV in the acute infection stage. Nonetheless, use of HIVST for serosorting activity should be addressed during pre- and posttest counseling materials to ensure testers are aware of the limitations of HIVST Kit assays.⁴³

Findings further revealed that mobile phone or apps are popular means for meeting sexual partners among Malaysian TW (42.4%). This finding is consistent with previous studies with other risk groups, which investigated the role of geosocial networking (GSN) apps for meeting potential sex partners.^44–49 More importantly, TW who reported using their mobile phone or apps to solicit sex work clients had greater willingness to use HIVST. It is possible that individuals who solicit clients online are making a rational judgment about their own risk levels and are willing to use HIVST. This may indicate not only a concern about risk of HIV infection but also a self-management response to their HIV risk behaviors,⁵⁰ allowing them to conveniently learn their HIV status and link to appropriate care early in the process. Furthermore, the high use of mobile GSN applications to find sex clients in this population suggests that GSN apps may be an important platform for delivering education about HIVST services, including information about where and how to obtain HIVST Kits.

We found that willingness to use HIVST was negatively associated with TW participants reporting living in Kuala Lumpur, and those exhibiting depressive symptoms, which was highly prevalent (56.0%) in this risk group. Prior research has suggested that one of the reasons for nonutilization of HIVST is the psychological distress that might arise from receiving a positive HIVST result without immediate personal or professional support present. It is further documented that the decision to seek HIV testing is influenced by numerous psychological factors (e.g., insight into one's susceptibility to HIV, perceived capacity to access and effectively utilize resources, and coping strategies).⁵¹ Individuals with moderate-to-severe depression and a lack of social support may choose not to use HIVST due to triggers of fear from positive test, which may not be adequately addressed before or after self-testing. Providing individuals with access to resources for counseling and support services pre- and post-HIVST, for example through 24-h hotlines, contact information for local mental health providers or access to virtual counseling services, could mitigate some of the potential psychological distress associated with HIVST within this population. Alternatively, HIVST programs for key populations could be designed to include online counseling and support through video chat programs, such as WhatsApp or Zoom, which may alleviate some of the psychological fears testers may have. In fact, one study of TW in Thailand found that HIVST with online support was the most preferred form of HIV screening for this population.⁵²

Furthermore, having experienced sexual assault in childhood was associated with greater willingness to use HIVST in our sample. Many studies have established a strong association between experiences of childhood sexual violence and later engagement in sexual risk-taking behaviors, such as having multiple sexual partners, inconsistent condom use, and participation in transactional sex.^53–57 It is possible that TW who experienced such trauma during childhood have heightened awareness of their vulnerability to HIV.⁵⁸ These TW may be more willing to use HIVST as it afford them the ability to test discreetly and conveniently and get linked to appropriate care (e.g., PrEP if negative, ART if positive). Additional research is needed to further elucidate this relationship to characterize and guide appropriate intervention development tailored specifically to this at-risk group.

The results of this study must be considered in the context of certain limitations. First, participants were recruited at venues where HIV testing services were being provided (community-based organizations), which may explain the higher rate of recent HIV testing in our sample. Furthermore, participants were recruited using direct referrals from staff members of the local community-based organizations, which may have biased our sample toward TW who are already connected to clinical resources and services. TW in regions without such community-based organizations may have less access to outreach and support and, in turn, may be more willing to adopt HIVST.⁵⁹ Furthermore, our sample comprised only TW and did not evaluate the acceptability HIVST among transgender men, gender nonbinary, or other minority gender populations. Second, participants were not asked about their prior knowledge of HIVST, previous use of HIVST, or various attributes related to HIVST (e.g., preferred method of distribution, associated cost, type of kit, and pre- and posttest counseling and linkage to prevention/treatment services). The findings reported in this study are therefore constrained by the lack of contextual information related to HIVST and its potential influence on actual uptake and prompt for future research. Third, all data were collected through self-report, introducing the potential for social desirability bias (i.e., underreporting of sexual and drug use risk behaviors) and recall bias. Fourth, the use of a cross-sectional study design limited our ability to make causal inferences. Nonetheless, our findings represent important data about TW's self-perceived willingness to use HIVST in Malaysia—a topic and population warranting further study given the significantly lower rate of linkage to HIV prevention and treatment services in the Malaysian context. The findings in this study contribute to a growing evidence base that suggests HIVST has the potential to greatly enhance uptake and retention to HIV testing in TW while also providing novel posttest pathways for linking individuals to prevention and treatment services.

Since WHO's release of the guidelines on HIVST in 2016, 77 countries have adopted formal policies on self-testing, while many others are currently developing them. Unfortunately, Malaysia has yet to establish guidelines for the use of HIVST, which also means there is no regulated and approved HIVST available to consumers in-country. As a discreet and convenient approach, HIVST might be most useful in reaching TW who are reluctant or unable to access traditional venue-based HIV testing services because of concerns about privacy, stigma, and discrimination, and, in some instances, criminalization. The findings from this study could be relevant as the market for HIV self-tests takes shape in the country, with the MoH in the process of developing necessary strategic plans, regulations, implementation guidance, and standard operating procedures needed for scale-up. Important considerations must be given while selecting service delivery models and support tools. For example, it is important to empower and effectively engage stakeholders and key population members in developing and adapting an HIVST service delivery model. The remote services that will complement the convenience and privacy of HIVST, such as new digital, social media, video or messaging platforms, toll-free hotlines, can be key for the scaling up and implementation of HIVST. Furthermore, it is important to raise community awareness about the benefits of HIV testing—including self-testing—and subsequent linkage to prevention or treatment services.

Conclusions

As the HIV epidemic continues to disproportionately affect TW, HIVST represents a promising strategy that could help decentralize testing, safeguard privacy, and make HIV service delivery systems more responsive to the needs of key populations, including TW. This study provided the first assessment of willingness to use HIVST among TW in Malaysia. Overall, the findings suggest that TW in our sample were moderately willing to use HIVST. Especially important here is the increased interest among those who are at increased risk for HIV infection. These results underscore the need to examine how best to implement safe and effective HIVST service delivery models to ensure utility of HIVST as an additional alternative to increase HIV testing rates among this marginalized and hard-to-reach group. As the MoH plans to scale-up HIVST among stigmatized and hard-to-reach groups, further research needs to be done to identify approaches to deliver HIVST in a safe and effective model to meet the diverse needs and preferences of those wishing to use this technology. Additional studies to determine optimal subsidy levels for self-tests as well as support options to link to appropriate services after HIVST is also needed.

Abbreviations Used

aOR

adjusted odds ratio

ART

antiretroviral therapy

ATS

amphetamine-type stimulants

BRFSS

Behavioral Risk Factor Surveillance System

CES-D

Clinical Epidemiological Scale-Depression

CI

confidence interval

CS

condomless sex

GSN

geosocial networking

HIVST

HIV self-testing

HT

hormone therapy

PrEP

pre-exposure prophylaxis

STI

sexually transmitted infection

TW

transgender women

WHO

World Health Organization

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This research was supported by a grant from the Yale Bates Summer Fellowship, the Yale Global Health Seed Funding Award, Yale Global Health Studies Scholarship, and a grant from the National Institute on Drug Abuse (K01 DA038529 for J.A.W. and R01 DA041271 for F.L.A.).

Cite this article as: Shrestha R, Galka JM, Azwa I, Lim SH, Guadamuz TE, Altice FL, Wickersham JA (2020) Willingness to use HIV self-testing and associated factors among transgender women in Malaysia, Transgender Health 5:3, 182–190, DOI: 10.1089/trgh.2019.0085.