HIV Testing Preferences Among Trans Men and Trans Women: A Retrospective Study of Client Records from 2017 to 2019 in a Community-Based Transgender Health Center in Metro Manila, Philippines

Background: Transgender individuals are considered as high-risk for contracting HIV infection. Integrating HIV testing and counseling (HTC) services into current transgender health programs is necessary to increase its uptake. Our study aimed to describe the characteristics of trans men (TM) and trans women (TW) who accessed HTC services in a community-based transgender health center in Metro Manila, Philippines, and to examine the relationship between gender identity and their HIV testing preferences. Methods: We conducted a retrospective study of TM and TW seeking care from 2017 to 2019. Medical records of clients were reviewed to ascertain their age, gender identity, year and frequency of clinic visits, lifestyle factors, and HIV testing preferences. The effect of gender identity on HIV testing preferences was estimated using a generalized linear model with Poisson distribution, log link function, and a robust variance, adjusted for confounding variables. Results: Five hundred twenty-ve clients were included in the study, of which about four out of ve clients declined the HTC services being offered. In addition, the prevalence of non-uptake of HTC services was 48% higher [adjusted Prevalence Ratio (aPR): 1.48; 95% Condence Interval (CI): 1.31–1.67] among TM compared to TW. Clients who initially consulted in 2017 had a 25% higher prevalence of refusal for HTC services (aPR: 1.25; 95% CI: 1.08–1.43) than clients who initially consulted in 2019. Approximately 4% and 11% of the TM and TW, respectively, who accessed the HTC services were reactive and linked to antiretroviral therapy treatment. Conclusion: HTC service uptake of TM and TW is low. HIV program implementers should strategize solutions to reach this vulnerable population for increased and better HTC service uptake and linkage to care.

and poor mental and sexual health outcomes [7]. This observation parallels the lack of direct research evidence, existing levels of stigma and discrimination, and narrowed options for healthcare such as gender a rmation services, preventive health screenings, and mental health interventions [8].
Transgender people are reported to have signi cantly inferior lifetime rates of HIV testing relative to cisgender gay and bisexual men. Conversely, HIV testing rates are likely to be lower among transgender adolescents [9]. It is also reported that sexually active transgender youth had STI prevalence between 1.40-2.80% [10]. Increased levels of discrimination such as denial of medical services and harassment in healthcare settings [11,12] as well as expected discrimination have been associated with postponement or delay of medical services among the transgender population [13,14].
To address the HIV/AIDS burden, local HIV prevention combination programs that are trans-inclusive are increasing [15]. Strategies include HIV self-testing, Pre-Exposure Prophylaxis (PrEP) and Post Exposure Prophylaxis (PEP), condoms and lube, and other biopsychosocial methods. Engagement in these programs and services will help mitigate the prevalence of HIV, suicide, and violence across the transgender community [16]. However, the Philippines, where resources are in scarcity, regrettably, struggles to address the unique set of healthcare needs the transgender community requires. Moreover, the current healthcare system in the country does not necessarily function effectively for the transgender population, speci cally for HIV testing and counseling (HTC) services. As a result, inclusive surveillance and data collection methods across the national transgender communities remain a challenge. Integrating the transgender population into the current Philippines HIV/AIDS surveillance system may modify this current state. Hence, it is essential to establish evidence to support the health outcomes of Filipino transgender people that will help inform program development and interventions explicitly targeted for this key population. Our study aimed to describe the characteristics of trans men (TM) and trans women (TW) who accessed the community-based transgender health center's HTC services in Metro Manila, Philippines. Moreover, we examined the relationship between gender identity and HIV testing preferences of the transgender population.

Study Setting
Victoria by Love Yourself Inc. (VLY), the Philippines' rst community-based transgender health center, was established in 2016. This initiative came about in response to the needs of the transgender community particularly, on access to comprehensive and quality transgender healthcare services. It is a one-stop shop that provides holistic care that integrates transgender health and sexual health.
The VLY services include free HTC services, HIV treatment care and support, Sexually Transmitted Infection (STI) consultation and treatment, PrEP, and PEP. They also offer gender-a rming services, such as gender transitioning counseling, pre-gender a rming surgery assessment and consultation, hormone administration, medically supervised gender-a rming hormone treatment, and even a support group for transgender people.

Study Design and Population
A retrospective study of TM and TW clients seeking care at VLY Community Center from March 2017 to December 2019 was conducted. Using routinely collected clinic data, we determined their issues relating to their sexual health, particularly their HIV testing preferences (consented or accepted HIV testing; refused or declined HIV testing). All client records of TM and TW who accessed the services of VLY were screened and included in the study using the following criteria: (1) 18-60 years old and (2) those who identify as transgender, and (3) not those who identify as otherwise including but not limited to cisgender, questioning, or genderqueer.

Data Collection
A review of medical charts was done to ascertain information from the study participants, including their age, gender identity, initial year and frequency of clinic visits, smoking and drinking statuses, use of recreational drugs, and uptake of offered HTC services. Moreover, data extraction of medical records was carried out following a developed case report form. Encoders were trained and ensured to have su cient expertise, particularly in handling medical records. To identify inaccuracies and discrepancies during the encoding, a small subsample of at least 10% of the total records was reassessed to validate the data encoded into the developed database.

Statistical Analysis
Descriptive statistics for the clients' demographic pro le, gender identity, and the HIV testing preferences outcome were calculated. Associations between every ascertained covariate and the HIV testing preference of the client (consented or accepted HIV testing; refused or declined HIV testing) were estimated using bivariable generalized linear models (GLMs) with a Poisson distribution, log link function, and a robust variance; a suitable method for cross-sectional data with a common outcome [17][18][19].
In addition, a multivariable GLM with also a Poisson distribution, log link function, and a robust variance was performed to estimate the adjusted effect of gender identity on the clients' HIV testing preference [17][18][19]. In the multivariable GLM, we controlled for age (15-24 years old; 25-34 years old; 35 years old and above), gender identity (TM; TW), frequency of clinic visit (1 visit; 2 to 3 visits; 4 visits and above), drinking status (never drinker; ever drinker), recreational drug use (never user; ever user), smoking status (never smoker; ever smoker) and year of initial consult (2017; 2018; 2019). The clients' characteristics included in the model were chosen a priori as potentially important predictors of HIV testing preference [20,21]. Moreover, the GLM was tted to account for the heterogeneity in the clients' preference on HIV testing.
For clients who availed of the HTC services of VLY, descriptive statistics were also calculated and strati ed by their HIV test results (reactive vs. non-reactive) to summarize the client's study characteristics. Furthermore, the treatment status of reactive clients was tabulated and strati ed by gender identity.
Crude (cPR) and adjusted (aPR) prevalence ratio with a 95% con dence interval (95% CI) were used to report the effect size estimates for the effect of gender identity, year of initial consult, and other confounding factors on HIV testing preference. STATA 17 software (www.stata.com/stata17/) was used to carry out all statistical analyses.

Ethical Approval and Consent to Participate
Following the national guidelines, the study's research protocol received ethical approval from the University of the Philippines Manila Research Ethics Board (UPMREB) (CODE: 2021-105-01). The data gathered and client information were kept con dential and private following the Philippine Data Privacy Act of 2012. Written informed consent form, from the participants, was not required in our study. The need for informed consent was waived by the UPMREB. Table 1 shows the descriptive statistics of the characteristics of the study population. A total of 525 TW and TM were included in the study. The clients have a mean age (± SD) of 25.8 ± 5.8 years old. Most of them belonged to the 15-24 years old age bracket (46.7%) and 25-34 years old age bracket (46.1%). Approximately 13 out of 20 of the clients were identi ed as TM, while the rest were TW. The year 2019, as the initial consult recorded the highest number of clients (56%), while only 10% were recorded during the rst year (2017) of the VLY (for details, see Table 1). Regarding the uptake of HTC services, approximately eight out of 10 clients refused or declined HIV testing. Conversely, among the 93 patients who consented or accepted HIV testing, 27 of them were TM (29%).   Clients who consented or accepted HTC services from the VLY were further described in Table 3 and strati ed according to their HIV test results. Approximately nine out of 10 clients were non-reactive, and most of them were TW (69%). In terms of year of consult, most of the reactive patients were observed in 2019, and the same year also recorded the most number of clients who availed the HTC services (57%). In addition, most clients who tested reactive were on treatment (88%), and one TW client was lost to follow up. Out of the eight reactive clients, three were already virally suppressed, and the other four were on ART (See Table 4 for details).

Discussion
Several factors may increase the risk of transgender populations for HIV infection. Trans women were identi ed as having more signi cant risks for acquiring HIV infection than other transgender populations. They were also least likely to receive any HIV treatments or interventions and other preventative services [2,[22][23][24]. The transgender community is also known to experience an increased risk for sexual behaviors, family rejection, stigma, discrimination, and safety concerns [25][26][27]. In addition, numerous individual, social, and interpersonal factors provide an interplay in terms of the experiences the transgender community experiences [28,29].
Report on the education and training for health professionals in the Philippines provided information on the adequacy of the current health curricula in terms of the HIV response [30]. Moreover, the Integrated HIV Behavioral and Serologic Surveillance embedded in the Health Sector Plan for HIV and STI 2015 to surveillance, suggested actions to increase HIV and HIV-related services both for the transgender (TG) and men-having-sex-with-men (MSM) populations [31,32]. However, the guidelines for the increase in uptake for HTC services among the TW and TM should be further strengthened because of the existing barriers to testing. [33] Our study aimed to identify the factors that enable the TG populations to refuse or decline HIV testing services. Through medical records review, our study showed that most TM did not consent or accept HIV testing services from the VLY, and they are more likely to refuse HIV testing services compared to TG.
Our results conformed with the prevalence report of the Center for Disease Control (CDC) recommended guidelines for HIV and Sexually Transmitted Infection (STI), wherein suboptimal trends on HTC services were observed among TG [34]. This nding is congruent with the results that TM did not know their HIV status [35]. However, one study on TG youth showed that TW was signi cantly less likely to get tested for HIV compared with TM [36]. Contrary to this nding, a recent publication on an extensive survey from the United States consisting of 26,927 TG respondents in 2015 revealed that TW had signi cantly higher odds of reporting their HIV status than TM [37], which was also seen in our study. In addition, the most common reason for never testing for HIV among TM was a low-risk perception of their sexual activities.
Low-risk perception as a signi cant barrier to HIV testing was also seen in previous studies [38 -41], not only among TG populations. Others reasons for TM or TW not getting their HIV testing also included fear of HIV-related stigma and discrimination [42,43], insu cient knowledge on HIV/AIDS or poor health literacy [44,45], and limited availability due to lack of time [46]. Further investigation on why TM and TW in VLY not knowing their HIV status because of refusal should be conducted to engage more TM and TW clients in HTC services. Moreover, increasing willingness for HIV self-testing among TM or TW to ensure one's safety and con dentiality is an approach that can also be explored [47,48].
The third-year since the launch of VLY in 2017 recorded the highest number of TM and TW clients consenting to HIV testing. Furthermore, our results showed that these clients in 2019 were more likely to get themselves tested for HIV compared with clients during the formative years of VLY. As an exclusive health center for the transgender community under the supervision of LoveYourself Inc., Victoria was initially established to provide HTC [49]. Over the years, through community consultations, partnerships with LGBTQIA + organizations, and transgender health capacity building of the community center, VLY o cially rolled out their gender-a rming services, the rst in the Philippines. The one-stop-shop model of integrating sexual health services and transgender health could translate to the gradual increase of HIV testing uptake among VLY clients. This strategy further establishes that gender-a rming care services can be an entry point in accessing HIV services.
Similarly, research has suggested that a gender-a rmative integrated care framework complemented by peer navigation effectively addresses the HIV burden experienced by the transgender population [50]. Our results also showed that the later years of VLY operations had established trust and con dence with its TM and TW clients to encourage them to avail themselves of the HTC services. Building trust and rapport between physicians, HTC service providers, and clients are crucial in all central HIV testing practices [51].
The integral approach in establishing trust is, to begin with, simple steps, to take part in clients with small successes, and to show dedication and commitment through continuous communication [52]. VLY used this strategy to build trust and rapport with TW and TM clients, which was also seen in previous studies conducted by other HTC providers [53,54]. The establishment of VLY as a community-based transgender health center for TM and TW provides an essential avenue for these populations to avail HTC services in con dence and without the stigma.
HIV testing lacking good motivational counseling and linkage to care may not be effective [55]. Our study provided information on reactive TG clients in which almost all of them were linked to care, particularly in

Conclusion
The role of early HTC services in the reduction of increasing HIV cases is an essential approach in the HIV care spectrum, especially to vulnerable populations such as the TG community. In our study, HTC service uptake of TM and TW is low. Our study, which demonstrated the refusal rate of HIV testing among transgender populations, particularly among TM, presented an opportunity for the HIV program implementers in the Philippines to reach this group to provide the HTC services they need.   Figure 1