Preexposure Prophylaxis in Latin America and the Caribbean—The Barbados Experience

Abstract

This study explored oral pre-exposure prophylaxis (PrEP) awareness, and sexual behavior of gay, bisexual, and other men who have sex with men (gbMSM) in Barbados. Factors associated with PrEP use were examined using Pearson chi-square and logistic regression. Of the 171 male participants who completed the survey, 22% were under 25 years old, 70% identified as gay/homosexual, and 59% were single. PrEP awareness was 77%, and of these 29% were current PrEP users. Users were more likely to have stable living arrangements (p < .05) and to identify as gay/homosexual (p=.03). Uptake by current PrEP users was attributed to non-judgmental and confidential PrEP services. Barriers identified by former PrEP users included cost and risks of STIs. Since the PrEP program is free, as is general health care for STIs, this suggests that there may be a knowledge gap. Further studies are needed to explore perceived barriers to the program.

Overall, there has been a general decrease in the incidence of HIV in Latin America and the Caribbean, but that decrease has not been stable and may not be large enough to change the epidemic in the region.¹ In 2008, 20,000 new HIV cases were recorded in the Caribbean (WHO, 2009),² and new cases decreased to 12,000 in 2012.³ In 2018, 16,000 new HIV cases were reported in the Caribbean, and in 2019, 13,000 new HIV cases were reported.^4,5 Gay, bisexual, and other men who have sex with men (gbMSM) continue to be heavily affected by the HIV epidemic even though the prevalence of HIV infections is decreasing in that group.⁶ Globally in 2019, 23% of new HIV infections were among gbMSM; the risk of acquiring HIV was 26 times higher for gbMSM compared with the rest of the adult male population.⁷ In the Caribbean in 2017, gbMSM accounted for 23% of the new HIV infections, and in 2019 they accounted for 26% of new infections.^6,7,8 In 2016 in Barbados, it was estimated that 11.8% of gbMSM were HIV-positive, compared with 1.6% of the general population aged 15–49 years.⁹ While there is a paucity of information on transgender women in the Caribbean, it was reported that they are also strongly affected by the HIV epidemic. In 2019, transgender women accounted for 5% of all new HIV infections in the Caribbean and in some countries such as Jamaica, the prevalence of HIV infections among this population was estimated to be greater than 50%.⁸

Pre-exposure prophylaxis (PrEP) through the oral use of a once-daily tablet containing tenofovir-emtricitabine (called PrEP onwards) for HIV prevention was recommended by the World Health Organization for MSM in 2014,¹⁰ although WHO had earlier recommended its use in demonstration projects for MSM and other at-risk groups^,11 Tenofovir-emtricitabine (PrEP), used by participants in our study, can be 99% effective in reducing HIV transmission when taken correctly.¹² Since 2015 WHO has recommended oral PrEP for people at substantial risk of HIV, and its use has become widespread among at-risk groups in many parts of the world.¹³ However, PrEP use has been limited in Latin America and the Caribbean, being offered as a public health intervention by six countries by the end of 2019.¹⁴ At the time of the study, PrEP was only formally available in the English-speaking Caribbean countries of Barbados and The Bahamas, being introduced in Barbados in 2018 and The Bahamas in 2017.⁹ According to PrePwatch, Guyana and Jamaica have been added to the list of Englishspeaking Caribbean countries offering PrEP.¹⁵

The Barbados PrEP program is coordinated by the Ministry of Health and Wellness. The program mainly targets gbMSM who are 18 years old or older, who may benefit from PrEP (according to the HIV Incidence Risk Index for MSM (HIRI-MSM) scale) and who meet a standard clinical profile. Additionally, they must express a willingness to use PrEP as prescribed, including periodic testing for HIV and STIs.⁹ At the time of this study, PrEP was not available in private physician's offices but at two health clinics/sites in Barbados.

This study explores the awareness of PrEP, behaviour towards PrEP, and factors associated with PrEP uptake among gbMSM and transgender women in Barbados.

Methods

Study design and recruitment

This was an online cross-sectional survey conducted among gbMSM residing in Barbados. In collaboration with an LGBTQ non-governmental organization (NGO) in Barbados, participants were recruited through email blasts sent by the organization to members. Participants could click on a hyperlinked URL, which would take them to the survey. The first email went out on May 29th, and reminder emails were sent out on June 19th and July 1st, 2019. Participants were encouraged to forward the survey link to others in their social network. Eligible participants had to be 18 years or older and living, attending school, or working in Barbados since these were required to be eligible for the PrEP program. While the study was originally designed to recruit gbMSM, we later opened it up to transgender women and female sex workers. No compensation was offered to take part in the study.

Instrument

The online questionnaire elicited sociodemographic information such as age, gender, education, income, housing arrangements, and religiosity. We assessed housing stability by determining whether the participants lived with close family members or a spouse/partner or whether they lived with a friend. Anecdotally, living with a friend is often considered a transient living arrangement in the region. For relationship status, we did not ask participants to specify the gender identity of their married partner. It should be noted that the only union socially or legally accepted in Barbados is that between a man and a woman. Awareness of PrEP, PrEP access, and reasons for initiating or stopping PrEP were also asked about. Questions on sexuality and sexual behaviours including numbers of sexual partners and condom use were also included. Sexual health including HIV status and sexually transmitted infection (STI) history, as well as drug and alcohol use were included. Questions used to determine facilitators and barriers to PrEP included multiple response questions such as, What where your reasons for discontinuing PrEP? (options included potential side effects of the drugs, cost, etc.), Where would you prefer to obtain PrEP?, What is most important to you if you were to/(or when you) obtain PrEP (options include such items as confidentiality of the service, non-judgmental service, ease of access, cost). While access to the program is without cost to nationals, there may be nominal costs involved for follow-up blood chemistry for non-nationals. Our questionnaire was adapted from an instrument created for a PrEP study conducted in Toronto.¹⁶

The HIRI-MSM, an HIV risk assessment tool, was included to determine PrEP eligibility.¹⁷ The scale is a Likert-like scale of seven questions assessing sexual risks in the past six months. A total score of 0 to 47 can be attained. Those scoring 10 or more are recommended to initiate PrEP.¹⁶ There are no data on the psychometrics of the scale for the Caribbean, however HIRI-MSM scores are used by the Barbados Ministry of Health and Wellness to determine PrEP eligibility.

Ethics

The study received ethical approval from the University of the West Indies–Cave Hill (IRB No. 190305-A). Participants were asked not to provide any information that could identify them such as their names, addresses, or workplaces. Participants were told that they could skip questions or sections that they did not feel comfortable completing.

Statistical analysis

For this exploratory study, analysis included univariate (frequencies, percentages, means, and standard deviations), chi-squared tests to describe the association between PrEP awareness, PrEP use, and other variables. Data for the HIRI-MSM scale were dichotomized to reflect the cut-off value of 10. We also used a binary logistic regression model to test the association between PrEP uptake and other variables using Stata software (version 15, StataCorp, College Station, TX, USA) for cross-sectional studies.

Results

Sociodemographic characteristics

As the survey was mostly distributed by our NGO partner, and partially through snowball sampling, we are unable to calculate the response rates. The survey was completed by 188 participants of whom 158 identified as males, 15 as females, 12 as transgender women and three as transgender men. As our main outcome variable was HIV risk based on the HIRI-MSM, and the HIRIMSM was designed for use with MSM, we limited our sample to MSM and transgender women (n=171). Our final sample included 11 participants (6.4%) who identified as transgender women. Due to the small number of transgender women in our sample, in order to protect anonymity we were unable to analyse the results for this group. Sociodemographic information of the sample is presented in Table 1. Most participants (86.5%) were younger than 40 years old and described themselves as gay/homosexual (70.2%) (note, while the word homosexual is stigmatizing, some participants chose to identify themselves using that term). Almost half of the participants had completed a bachelor's degree or higher (48.5%) although less than half earned a monthly salary of $2,500 BDS or more (Average salary in Barbados at the time of the study was $2,500 BDS/month).¹⁸ More than half of the participants worked full-time (56.7%). Thirty-one percent were married or in a partnership.

Sexual behaviour and drug use among MSM for use with the HIRI-MSM

Of the gbMSM participants who reported any sexual behaviour in the last six months (n=151), the majority (n=122; 80.8%) reported having anal sex with one to five male partners. (Table 2). Of those who responded to the question on condomless receptive anal sex in the last six months, 81 (56.6%) reported doing so between one and three times. Twenty-five (17.5%) reported having anal sex with an HIV-positive male partner and of these men, 13 (52.0%) had condomless sex with an HIV-positive partner at least once in the last six months. In the last six months, 10% had used methamphetamine and 8.6% had used poppers (amyl nitrate). As determined by the HIRI-MSM cut-off score for PrEP eligibility, the majority of the sample (145/151, 96.0%) had a score of 10 or more.

PrEP awareness, procurement, and use

Figure 1 outlines the extent of PrEP awareness and use. More than three in four people were aware of PrEP. Of those who were aware of PrEP (n=68), 39.8 % (95% CI 35.6–51.7) were PrEP-experienced. Most of the PrEP-experienced men (50/68, 73.5%) (95% CI 59.6–80.9) were current PrEP users.

Table 3 explores PrEP use among current (n=48) and former users (n=20). More than two thirds (77.6%) of the current users, used PrEP for three to six months compared with 44.4% for former PrEP users. No current PrEP user had been on therapy for over six months but 16% of former PrEP users were so experienced. Among the current users, 89.8% used PrEP daily while 55.6% of the former users had used PrEP daily. Among the current users, 87.8% obtained PrEP from the Barbados PrEP program whereas 55.6% of former users had obtained PrEP from that program. Six former PrEP users had obtained PrEP from friends and three had obtained it from the U.S. or other private programs. By age group, those older than 40 years were more likely to have used PrEP (48.0%). Slightly more than half the sample who were aware of PrEP but were not PrEP-experienced (n=64), were interested in using PrEP (n=35, 54.7%) (Data not shown in the table).

Facilitators and barriers to PrEP use

We asked several questions about barriers and facilitators to obtaining PrEP among current and former users as well as those who had never used PrEP but were interested in its use (data not shown). Among current PrEP users, facilitators included confidentiality (38.0%), ease of access (30.0%) and a non-judgemental service (26.0%). Similarly, for those interested in using PrEP confidentiality (27.9%), ease of access (22.1%) and non-judgemental service (19.1%) were important.

Table 1.

SOCIODEMOGRAPHIC CHARACTERISTICS OF STUDY PARTICIPANTS (N=171)

Table 2.

SEXUAL BEHAVIOUR AND DRUG USE AMONG PARTICIPANTS^a

Barriers to PrEP among current users included stigma attached to PrEP use (47.1%), potential side-effects (42.6%), cost (36.8%), perceived attitude of PrEP program staff (22.1%), and increased risk of other STIs (19.1%). For former PrEP users, barriers included cost (45.0%), perceived risk of contracting another STI (30.0%) and side effects (25.0%). Among those interested in PrEP, more than half (69.2%) reported that they were unaware of the local free PrEP program. Concerning location for receiving PrEP services, current users preferred service from a non-governmental organization

Figure 1.

PrEP awareness and use among participants in the entire sample.

Table 3.

PREP PROCUREMENT AND USE AMONG CURRENT VERSUS FORMER PREP USERS

(NGO) (59.0%) while those interested in PrEP use would prefer to obtain PrEP from a general practitioner's office (64.7%).

Current PrEP users were asked about their reasons for using PrEP through a multiple response/open-ended question. The most frequent reasons given for PrEP use were HIV prevention, being recommended by friends, the desire to have many sex partners, and because of their sexual practice (data not shown).

Table 4 shows the unadjusted (bivariate) associations between current and former

Table 4.

BIVARIATE LOGISTIC REGRESSIONS FOR SOCIODEMOGRAPHIC ASSOCIATIONS TO ANY PREP USE

PrEP use and sociobehavioural variables. Stable living arrangements (living with family)—Odds Ratio OR: 3.28, 95%CI: 1.22, 8.85; or partner—OR: 3.30 95% CI: 1.12, 9.70, and identifying as gay / homosexual—OR: 10.15, 95% CI: 1.27, 81.07; were associated with PrEP use, (p-value <.05).

Discussion

This exploratory study focused on PrEP awareness and PrEP use including facilitators and barriers to PrEP use among a sample of gbMSM in Barbados. To maintain anonymity of the data, and out of respect to the lives of the transgender women, their information was subsumed under gbMSM. The study used one main source for recruiting participants, which increased the possibility that the sample could be representative of MSM and transgender women who frequent the NGO. However, it is not possible to determine whether these individuals are representative of all gbMSM and transgender women in Barbados. At the time of the study, there were approximately 134 PrEP users (126 cisgender men, 6 transgender women and 2 cisgender women) of the country's PrEP program.⁹ Respondents in the study were highly educated but had lower incomes than the country's average.¹⁸

Participants in the study exhibited a high level of awareness of PrEP despite the recent introduction of the program. Barbados MSM's level of awareness was on par with, or higher than other studies that observed PrEP awareness in recently introduced PrEP programs.^19,20,21 For example, in a systematic review of global PrEP data, the proportion of MSM aware of PrEP was 50% (95% CI: 44.8–55.2).¹⁹ In a U.S. study, only 39% of Black MSM and transgender women were aware of PrEP two years after it was introduced,²¹ although PrEP awareness continues to rise in this and other MSM populations in the U.S.²² The high level of PrEP awareness among participants in our study may be due to the small geographic and dense population advantage of the country. Further, as the participants were mostly drawn from a group that facilitates both social and health care for the LGBTQI population, it is more likely that this group would be knowledgeable about PrEP. At the time of the study, the PrEP program was not publicly advertised.

We used the HIRI-MSM to indicate eligibility for PrEP. As assessed by the HIRIMSM, 96% of our participants scored in the PrEP-indicated category but less than a third were on PrEP. The reason for this high PrEP-indicated category may be a function of our recruitment strategy as the majority of the participants were recruited through an LGBTQ NGO that caters to men who are young and may be more comfortable in exploring their sexuality. Young age is a known risk for high number of sexual partners and condomless sex.²³ Another reason is the fact that the HIRI-MSM is not sensitive to the type of relationship and does not account for the respondent's HIV partner's viral load (example, a monogamous couple with an HIV positive partner with an undetectable viral load would score high on the scale). There is ample evidence showing that there is very little risk of HIV transmission when the viral load is undetectable.^24,25 Further, from an anti-discrimination perspective, categorizing all condomless sex as high risk may be counterproductive. Participants who have condomless sex with an HIV-positive partner are judged to be in the PrEP-indicated category according to the HIRI-MSM scale. However, if these individuals are virologically supressed then transmission would be negligible and they would be low-risk.^24,25 The HIRI-MSM scale only uses information on sex with men and excludes sex with women which also has the potential for HIV transmission (in addition to other STIs), among bisexual men. It is known that levels of virus can be different in semen vs. in vaginal fluids (bisexual/transgender men) and the blood.²⁶ Further, individuals may need probing to determine whether they have sex with non-main partners, as transmission among men in serodiscordant relationships may be from non-main partners.^25,27 Some men may also not be forthright about being the receptive partner in anal sex, as in some communities being the receptive partner bears higher stigma.²⁸ The HIRI-MSM has also been shown to have lower sensitivity for some groups such as Black MSM since it does not take into account sexual networks (sex with older sexual partners) and the local HIV epidemiology—which may be important for incident HIV among young Black MSM.²⁹ Therefore, it is up to the medical provider to take an unbiased history from the patient before deciding on the course of action for PrEP.

Users of PrEP believed that using PrEP would reduce their risk of HIV infection. This is an important observation as it validates a PrEP program. While we did not test the importance of PrEP use in improving the psychological well-being of individuals, other studies have noted this important trend.³⁰ Our results are similar to other research noting that people who use PrEP are aware of associated sexual risks and take preventative measures to reduce the probability of HIV transmission.^31,32 Other facilitators to PrEP uptake within our study included the recommendation of PrEP from friends and others. This is also an important communication piece because the program was not publicly advertised. This was also similar to studies that revealed that PrEP facilitators included influence from peers / friends.^33,34,35

Willingness to use PrEP for the uninitiated was high and is comparable to the results of a meta-analysis of willingness to use PrEP in global gbMSM populations where it was measured at 58.6% (95% CI: 54.8–62.4).¹⁹ In another meta-analysis of data from low and middle-income countries, six studies measured willingness to use PrEP at less than 50%, nine studies recorded levels between 50%–70%, and five studies recorded levels of over 80%.²⁰ Based on our discussion with local PrEP program implementers, the program was kept small to ensure that demand did not exceed capacity. Yet there appeared a greater need than was met by the program. Importantly, studies have shown that willingness to use PrEP may not predict PrEP uptake and adherence due to other factors such as clinic operation time, and other structural barriers to services, which were not investigated in our study.^36,37

For people who were not PrEP-experienced, the main reasons for not using PrEP was being unaware of the free PrEP program. As already discussed, the program was deliberately contained by not publically advertising so that capacity did not exceed demand. Stigma may also have affected the use of PrEP as seen in other studies.^36–39 While the early days of the PrEP program were marred by the concept of "Truvada whores" (i.e., use of PrEP for the main purpose of enhancing sexual pleasure), stigma around PrEP use is still an important barrier.^37,39 Taking PrEP can be seen as confirmation of being gay which is a problem for men who do not want to be labelled as gay.⁴⁰ Even among those who identify as gay, issues around morality, shame, and guilt of sexual pleasures may prevent them from using PrEP.^41,42 Therefore, it is important that PrEP providers de-emphasise morality and/or discuss sex in a sex positive manner when addressing PrEP services.

The main reasons for discontinuing PrEP among former PrEP users, were related to cost of PrEP and associated side effects. Our study did not explore the meaning of "cost" to participants. Notably, while PrEP dispensation in Barbados is free, there may be nominal costs related to follow-up medical visits for non-nationals, which are required in order to be in the program. PrEP is distributed centrally, and transportation could have been a problem in accessing the service although this was not explored in the study. At the time of the study, PrEP was dispensed nationally at two locations, one of which was not along the main bus route. Further, while bus fare was only $1.50 USD per route, some participants may need to change several buses to reach the clinic locations which would result in additional time expenditure. Our results also showed that some former PrEP users procured PrEP from friends and other networks more frequently than current users. The reasons for doing so may be multifactorial, but these were not explored in the study.

Participants of all groups of PrEP users indicated that confidentiality of the PrEP service and being provided with nonjudgemental service were important to PrEP uptake. Participants' ratings of the PrEP service were equivocal. Whether or not participants experienced barriers, it is important that their concerns are valued. Therefore, providers of PrEP service should be cognizant to cues of possible barriers to PrEP uptake.

As already discussed, PrEP was only available at two clinic sites in Barbados and not available at public or private clinics and other health care providers' offices. While access to primary care providers has been shown to be important in PrEP access and uptake,^43,44 this important information could not be explored in the study. It should be noted that Barbados has a universal health care system for citizens, encompassing one tertiary level hospital and eight polyclinics, some of which provide secondary levels of care. There are also private clinics, hospitals and medical laboratories. Drugs in country can only be procured by one government agency which procures pharmaceuticals and medical supplies through an open tender annually, procuring drugs from local manufacturers and, from sources in the U.S., Canada, South America, and Europe.⁴⁵

As a study using convenience sampling based on a narrow sampling frame there are inherent biases that limit the results of the study. Further, we could not reliably establish a sampling frame as our main recruitment organisation email list was not up-to-date which limited our ability to determine full membership. As in all internet-based self-reported studies, participation was limited to those with access to a computer, laptop or a smartphone. Data rates are expensive in Barbados, which would have been a barrier to completing the study by phone. Additionally, the study was long to complete on a smartphone. While we were able to provide printed copies of the questionnaire to participants upon request, only four participants chose that route. These limitations are well described in other internet-based studies.⁴⁶ To reach a broader group of gbMSM who may not be connected to the NGO used as the primary source for recruitment, we used a snowball technique. However, our data were not collected in such a way that we could determine referral patterns.

The original study was not designed for transgender women and therefore lacked a nuanced approach to determine this population's specific PrEP needs. A study specifically designed for this population is needed. While the results of the study may not reflect the full spectrum of the populations studied, we believe that the results provide a basis from which other Latin American and Caribbean countries can bolster their PrEP services or institute PrEP programs.