Introduction

Labor migration is a global human phenomenon fueled by a wide array of factors such as globalization, demographic shifts, conflicts, income inequalities and climate change. The current statistics indicate that there are approximately 232 million international migrants [1, 2], and that more than 90% of international migration is related to seeking employment [3]. Thailand has achieved tremendous progress in its economic development. In less than a generation, the country has shifted away from a low-income status to become an upper-income country [4]. As such, Thailand has become a magnet country, attracting an increasing number of migrants from its less-developed neighboring countries, namely Myanmar, Cambodia, and Laos. There are approximately 3.6 million migrant workers (MWs), from Myanmar, Cambodia, and Laos, in Thailand, including both documented and undocumented migrants, with or without work permit [5]. Thailand, being a member of the AEC (ASEAN Economic Community), an organization that endeavors to promote economic integration in the region, including free movement of labor [6], is expected to experience a continued increase in the number of its migrant workers, with downstream impact on its public health system [7].

Thailand has a success story in reducing rates of HIV infection among the Thai population, which is overwhelmingly cited as an example of an effective HIV prevention program [8–10]. However, the country still holds one of the highest rates of adult HIV prevalence in Asia [11], and HIV infection has remained fairly high among high-risk individuals, including intravenous drug users (IDUs), men who have sex with men (MSM), and commercial sex workers [12]. In Thailand, MWs are recognized as a priority group for HIV prevention, treatment and care [12]. The vulnerability of MWs to HIV infection is widely established. Several studies in Asia have documented a number of risk factors among MWs including but not limited to low levels of AIDS knowledge, low perceived risk for HIV/STIs (sexually transmitted infection), high levels of risky sexual behaviors, and limited access to health care [13–20]. Similarly, many studies in Thailand showed that MWs engaged in behavioral patterns that increased the risk of HIV infection. Low rates of condom use [21–23] and high frequency of visits to sex workers [21, 22], compounded with poor access to health care services [24, 25], were commonly observed among migrant workers in Thailand.

Despite the documented profile of risky sexual behaviors among MWs in Thailand, there are still extensive knowledge gaps regarding HIV testing in this population. HIV testing is the gateway to effective HIV prevention and treatment. Many studies have shown that awareness of HIV status is associated with substantial reduction of risky sexual behaviors among HIV-infected individuals, and persons with HIV-negative status are less likely to engage in risky sexual behaviors after they become aware of their HIV status [26]. HIV testing fosters timely linkage to antiretroviral therapy (ART), which in turn is associated with positive clinical outcomes, and reduced risk of HIV transmission at both individual [27, 28] and population levels [29]. However, the proportion of individuals who know their HIV status remains significantly low. At the global level, only 50% of the population were aware of their HIV status as of the end of 2012 [30]. In Thailand, studies on HIV testing in Thailand, as a reflection of the concentrated nature of the epidemic in the country [12, 31], have mostly focused on populations most at risk (youth and adults) including MSM, female and male sex workers, and IDUs, with HIV testing rates ranging from 23–76% [32–36]. Little is known about HIV testing behavior among MWs in Thailand, including factors that facilitate and restrain the uptake of HIV testing services in this population. Such information is crucial for guiding interventions to increase HIV testing among MWs in Thailand, and ultimately, to prevent HIV infection and promote timely linkage to HIV care and treatment.

The current study uses data from the baseline survey of the Prevention of HIV/AIDS among MWs in Thailand (PHAMIT-2) project to document the prevalence of, and factors associated with, HIV testing among MWs from Myanmar residing in Thailand. PHAMIT-2 is a GFATM (Global Fund to Fight AIDS, Tuberculosis, and Malaria)-funded project designed to reduce new HIV infections among MWs mainly from Myanmar, Cambodia, and Laos [37]. The current analysis is restricted to MWs from Myanmar, which is the largest group of MWs in Thailand.

Methods

Results

Participant characteristics

Table 1 displays socio-demographic and behavioral characteristics of participants, and Fig 1 outlines the flow of the analysis. In total, 2,169 MWs from Myanmar completed the survey (the refusal rate was estimated to be less than 1%), of which 1,049 participants consisted of those who knew a place to test for HIV and 1,120 of those who did not or were not certain where to test for HIV. Of the participants who knew an HIV testing place, most had a primary or lower education level (73.8%), and over half lived with a partner (57.0%), and had resided in Thailand for less than 5 years (54.5%). The majority of participants had a work permit (85.3%) and a legal ID (92.9%), worked in factories (57.4%), lived outside of their working sites (62.1%), and were sexually active (74.4%). The median age in this group was 27 years [Interquartile range (IQR): 23–33], and there was a relatively higher proportion of males than females (58.8% vs 41.2%).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Flow of the data analysis.

https://doi.org/10.1371/journal.pone.0154669.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Socio-demographic and behavioral characteristics of participants filtered by the item “do you know any places proving counseling and testing for HIV.”

https://doi.org/10.1371/journal.pone.0154669.t001

Of the 1,120 participants who did not know or were uncertain about a place to test for HIV, 825 (73.7%) had a primary education level or lower. The median age was 25 years (IQR: 20–32). Most participants were either single (50.2%) or lived with a partner (43.2%). The majority of participants in this group had resided in Thailand for less than 5 years (69.5%), had a work permit (76.3%), a legal ID (85.4%), worked in factories (60.8%), and had sexual experience (53.6%). Among the sexually active participants in this group, 537 (90.3%) had a spouse or regular partner, of whom the majority (86.2%) never used condoms with their spouse or regular partners. Compared to participants who knew a place for HIV testing, those who did not know (or were uncertain) appeared, to some extent, to have a lower risk of HIV/STI infection. The percentage of condom use in the past 12 months for those who last had sexual intercourse with a regular partner was higher (85.2% vs 60.3%) than that of participants who last had sexual intercourse with a non-regular partner. (1.7% vs 4.1%) Nevertheless, condom use with a regular partner (9.9%) and consistent condom use with a non-regular partner (20.0%) of this group of participants were fairly lower than those of participants who knew an HIV testing place (18.4% and 34.7%, respectively).

Bivariate associations between independent variables and past HIV testing experience

The analysis to identify correlates of “ever been tested for HIV” included a total of 1,034 participants who knew a place to test for HIV. Fifty-five participants (5.3%) reported having been tested for HIV. MWs with a secondary or higher education level [odds ratio (OR), 1.90; CI, 1.06–3.42; P = 0.028)], those who had resided in Thailand for five years or more (OR, 2.80; CI, 1.56–5.04; P = 0.000), lived outside of the work site (OR, 2.04; CI, 1.08–3.85; P = 0.025), and who were able to speak Thai (OR, 2.92; CI, 1.41–6.03; P = 0.003), were more likely to have been tested for HIV. Participants who were single were less likely to ever have been tested for HIV (OR, 0.45; CI, 0.23–0.873; P = 0.016) compared to those who lived with partner. MWs who worked in the fishery sector were more than two times likely to report ever having been tested for HIV compared to those who worked in factories (OR, 2.51; CI, 1.35–4.67; P = 0.003). (Table 2)

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Bivariate associations socio-demographic and psychosocial variables with ever have been tested for HIV.

https://doi.org/10.1371/journal.pone.0154669.t002

Table 3 displays reasons for testing for HIV by gender. Most MWs (75.9%) cited pregnancy as the reason for testing, followed by curiosity (13.8%), and as part of the job application process (10.3%). Regarding male migrant participants, curiosity (46.4%), blood donation (25.0%), and job application process (21.4%) were the most frequently cited reasons for testing.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Reasons for testing for HIV.

https://doi.org/10.1371/journal.pone.0154669.t003

Multivariate analysis

In the multivariable analysis (Model 3), having a secondary or higher education level (AOR, 2.58; CI, 1.36–4.90; P = 0.004), having resided in Thailand for at least five years (AOR, 2.45; CI, 1.25–4.78; P = 0.009), knowing someone who died of AIDS (AOR, 1.81; CI, 1.00–3.27; P = 0.048), and working in the fishery sector (AOR, 2.51; CI, 1.28–4.92, P = 0.007) remained significant. Being female (AOR, 1.96; CI, 1.05–3.66; P = 0.033) and not having a work permit (AOR, 3.71; CI, 1.36–10.13; P = 0.010) were associated with increased odds of ever having been tested for HIV. Additionally, there was a trend for participants who resided outside the work site to more likely have been tested for HIV compared to those who lived in the work site (AOR, 1.82; CI, 0.93–3.59; P = 0.079). (Table 4)

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Multivariate analysis of factors associated with ever have been tested for HIV.

https://doi.org/10.1371/journal.pone.0154669.t004

Discussion and Implications

There is a remarkable dearth of studies addressing HIV testing among cross-border MWs in Thailand, and in Southeast Asia at large. The current study leveraged data from the PHAMIT-2 Project to document the prevalence of, and factors associated with, HIV testing among MWs from Myanmar.

In this study, the odds of ever having been tested for HIV was higher for participants who had at least a secondary education level. Our finding corroborates results from a previous study in Kenya in which individuals with secondary and higher education attainment were more likely to have been tested for HIV than those with no or lower education levels [40]. There is a general trend for individuals with higher education levels to use more health care services [41–43]. In other settings, higher education levels were associated with higher knowledge on HIV/AIDS prevention, transmission and HIV testing [23, 44], and the latter is documented as a predictor for HIV testing [45, 46]. However, in our study, there was no association between knowledge on HIV/AIDS and HIV testing. Nonetheless, interventions that aim to improve HIV/AIDS-related knowledge and awareness on the benefits of HIV testing are an important component of HIV prevention, and should be delivered in a way that is accessible to those with little or no formal education.

We also found that females were almost two times more likely to have been tested for HIV compared with males. This is in line with findings from other studies showing gender differentials in HIV testing behavior, and fits well with the common observation that females have more contact with health care systems, particularly because they attend routine antenatal care as part of the sexual and reproductive health services [47–49]. Additional evidence in support of gender differentials in HIV testing in our study includes the fact that female MWs overwhelmingly cited pregnancy as the reason for testing, contrasting with male MWs who mostly tested for HIV out of curiosity, for blood donation, or as a part of the work application process. The uptake of HIV services by men has proven to be a challenge in many settings, and there exists evidence showing that migrant men are less exposed to HIV testing and less willing to be tested [50]. Our study suggests the need to consider gender differences in the design of intervention programs to improve the uptake of HIV testing services among MWs. Interventions such as those that strengthen antenatal care services as an entry point for HIV testing can potentially benefit both females and males if they are delivered in a framework that targets couples for testing. Interventions should also target female and male MWs outside the context of antenatal care and/or sexual reproductive health services. For example, mobile HIV testing outlets organized in the vicinity of, or at worksites could be a promising strategy to boost HIV testing in male MWs.

In our study, MWs without a work permit had higher odds of ever having been tested for HIV compared to those who possessed a work permit. It is not clear why this was the case; however, considering the characteristics of participants presented in Table 1, 265 of 419 participants (63.2%) without a work permit reported that they did not know any place providing counseling and testing for HIV. This proportion was fairly higher than that of participants who possessed a work permit, 855 of 1,750 participants (48.8%), implying that “knowing a place for HIV testing” was still an important barrier of accessing HIV testing for MWs without a work permit than those with a work permit. Thus, we might have selected in the group of participants without a work permit, those who were more likely to have been tested for HIV because our analysis was restricted to the group of participants who knew a place for HIV testing. There is however still need for future research to explain the observed difference in HIV testing between MWs with and without a work permit in Thailand. In contrast, MWs with legal ID were more likely to have been tested for HIV; however, the association was not statistically significant. In other settings, legal factors such as the lack of entitlement to health service for undocumented factors, or the fear that disclosure about HIV may adversely affect legal status were documented as important factors that increased the vulnerability to HIV infection among the population [51].

Knowing someone who is infected with, or who died of HIV/AIDS may influence individuals’ perceived risk of HIV risk, which in turn, depending on studies, was shown have been associated with increased or decreased odds of testing for HIV [52–56]. These variables have been meaningfully used as an indicator to help contextualize individual HIV risk perception in many settings [57–59]. In our study, participants who knew someone who died of AIDS had increased odds of ever having been tested for HIV. A previous study showed that, across many African countries, participants who reported that they did not know someone who has AIDS or died of AIDS were much less likely to report ever having been tested and to receive their HIV test results [60]. However, HIV risk perception was not associated with ever having been tested for HIV in our study. This is an indication that factors other than risk perception may mediate the association of knowing someone who died of AIDS with ever having been tested for HIV. Moreover, there is a possibility of residual confounding due to insufficient sensitivity of our measure of HIV risk perception. Furthermore, although there is a well-established relationship between self-perception of HIV risk and uptake of safer behaviors [60], this association suffers from a lot of inconsistencies [61–63]. Nevertheless, interventions designed to increase risk perception, more specifically HIV risk personalization, should be one of the prongs of strategies to promote HIV testing among MWs.

In this study, working in the fishery sector (fisheries/deep water seafarer) was independently associated with increased odds of ever having been tested for HIV. Migrant fishermen in particular have been identified as a group at high risk of HIV infection among the MWs in Thailand. For example, seafarers are more likely to visit sex workers than other migrant groups and to report lower use of condoms particularly when acquainting with non-regular partners [21, 64]. In a qualitative study, migrant seafarers reported that many of them went for testing because they felt susceptible to HIV; however, many others were reluctant to testing due to fear of a positive result or other inconveniences such as time and money [22]. The same study indicated that seafarers were aware of their risk of HIV infection especially from sex workers, and most had observed their peers become sick and die from AIDS. This could to some extent explain our finding that knowing someone who died of AIDS was associated with increased odds of ever having been tested for HIV.

Although not statistically significant, there was a trend for MWs residing outside of the working site to more likely report ever having been tested for HIV compared to those who lived in the work site. This finding is of particular relevance as it emphases the need for interventions to reach out to MWs who work and live at their workplace, and as a consequence, have limited exposure to health programs that are routinely provided to the community. As outlined earlier, provision of mobile HIV testing and counseling to MWs at their workplace can possibly increase the proportion of HIV testing in this population. Although not related to MWs, results from a community-based intervention strongly demonstrated the efficacy of mobile voluntary HIV counseling and testing (MVCT) provided free of charge and associated with community mobilization to increase the HIV testing rate in Northern Thailand rural communities [65]. In the contexts of the PHAMIT-2 project, with collaborations between NGOs, the local Public Health Office and the hospital located in the area, MVCT also operated with a focus on migrants living in the community and in the workplace. The latter group was, however, more difficult to access for the project, in most cases due to non-cooperation or disapproval from the employer.

Previous research has linked language barriers to limited access to health services, including uptake of HIV-testing service [49]. Poor or limited ability of the language in the host country poses a serious challenge for MWs to accessing health care services. In our study, however, Thai speaking ability was not associated with HIV testing in the multivariate models. It is important to note that although the majority of participants who were tested for HIV in the last 12 months, and received pre-and post-test counseling, reported that counseling was offered in the Burmese language, and that most understood the content of both pre-and post-counseling sessions, a substantial proportion of MWs did not receive pre-and post-test counseling. This suggests that language barrier or unavailability of the counseling in the MWs’ native language is still a possible reason to explore further, and that provision of migrant-friendly services, such as those delivered in the migrant’s native language, remains important (See S1 Table). In addition to providing language-sensitive HIV testing services, the quality of HIV testing services will greatly increase by ensuring confidentiality of the test results. In our study, the majority of test results were not provided in strict confidentiality (S1 Table).

Overall, we found that a low proportion of MWs from Myanmar had previously tested for HIV. Other studies in Asia have reported a low proportion of HIV testing among the migrant population. For example, in China, only one out of 100 male migrants had previously been tested [13, 66]. In our study, the actual proportion of those who had ever been tested could even be lower than that reported under the provision that those who did not know any place providing HIV test are considered as never having been tested for HIV. The low proportion of HIV testing in MWs from Myanmar in this study is a strong call for strategies to strengthen HIV testing and counseling services for the MWs in Thailand in general. Interventions should be tailored to the specific needs of MWs, and should be grounded in outreach and community mobilization that integrate social, cultural, and structural contextual factors that shape the health-seeking behavior of MWs in Thailand. In the PHAMIT-2 Project, one of the key goals is to increase the HIV testing rate of MWs in Thailand through a range of multi-level and multi-stakeholder activities. For instance, the PHAMIT-2 Project intends to increase the uptake of VCT services through field staff and peer outreach who refer MWs to VCT services at drop-in centers or community hospitals, strengthened by mobile VCT services operated by hospitals or health center staff. These activities are synergistically linked with other core activities of the Project, which indirectly have an impact on HIV testing behavior [37].

There are a number of limitations worth mentioning. Firstly, our study is cross-sectional by nature; thus, causation cannot be inferred. Secondly, our outcome measure "ever been tested for HIV", while being useful indicator to assess the overall reach of HIV testing services, does not capture recent trends in HIV test-seeking behavior as would a time-bound indicator like "tested in the last 12 months". Thirdly, participants were not randomly selected and are, therefore, unlikely to be representative of all MWs from Myanmar although those included may have been more likely to have been tested given the manner of their selection. Fourthly, because of the small proportion of participants who reported ever having been tested for HIV in our sample, we may not have sufficient statistical power to assess the relationship between the predictors and our outcome of interest. Moreover, the survey questionnaire was designed in such a way that participants who were asked if they had ever been tested for HIV were filtered from those who knew a place to test for HIV. Consequently, the analysis to identify correlates of "ever been tested for HIV" was conducted in only around half of the participants that were recruited. We recognize that this could be a source of significant bias; however, the degree to which our results may have been affected could not be determined. When assuming that participants who did not know any place providing HIV testing had never been tested for HIV, the proportion of those who reported "ever been tested for HIV" was substantially reduced (2.5%), and we could not perform the multivariate analysis. Fifthly, Of the 2,169 Myanmar respondents, 42.6% were Burmese, 22.9% were Dawei or Tavoya, 21.6% were Mon, 9.4% were Karen, 1.8% were Pa-oh, 1.3% were Ya-kai, and 0.3% were Shan. The fact the interviews were conducted in Burmese, at the exception of the Mon who were interviewed in Mon language, could be a possible source of selection bias given that those with no or limited ability to speak Burmese might have been excluded from the study. Additionally, the variable “ethnicity”, although epidemiologically important for our study, was not included in our analysis given that some ethnic groups who had no or too few outcomes, responded yes on the HIV testing item. There is a possibility that social desirability and recall bias might have affected our results given that the collected data were self-reported. Lastly, issues such as “health insurance status”, “discrimination”, “stigma”, and “security”, which are important for the uptake of HIV testing, were not addressed in the survey; therefore, could not be controlled in our models. Further research is needed to complement our findings by addressing the above limitations.

Conclusion

The main finding from our study is that a significantly low proportion of Thailand’s MW population from Myanmar has ever been tested for HIV. This sounds the alarm for the need to rapidly develop strategies to improve the up-take of HIV testing services in this population, and thus secure the benefits associated with awareness of one's HIV status. We have identified a number of individual-level factors (such as gender, education, and lack of knowledge about HIV testing sites) and structural barriers (limited availability of HIV test counseling in migrants’ native language) that can inform future interventions to improve HIV testing among MWs in Thailand. Such interventions should include culturally-sensitive HIV/AIDS and health education that encourages safe sexual practices, facilitates the flow of information about places for HIV testing sites, and ensures availability of language assistance and confidentiality of HIV testing.

Supporting Information

Acknowledgments

The authors would like to thank the Mahidol Migration Center (MMC), and the research team at the Institute for Population and Social Research (IPSR), Mahidol University for the kind permission to use the dataset from the PHAMIT-2 Baseline Survey 2010. Appreciation is also extended to the Raks Thai Foundation, the supporter of the survey. We thank the JSPS Program for Advancing Strategic International Networks to Accelerate the Circulation of Talented Researchers, "The Japan-ASEAN Collaborative Research Program on Innovative Humanosphere in Southeast Asia: In Search of Wisdom toward Compatibility Growth and Community in the World." for supporting the collaboration between Kyoto University and the IPSR, Mahidol University. Finally, the authors would like to thank Larissa Popp (Senior Program Officer, FHI 360) for helping edit the current paper