International Journal of STD & AIDS >
Volume 19, Number 12 >
Pp. 833-837
Original research articles |
* Shoklo Malaria Research Unit (SMRU), PO Box 46, Mae Sot, Tak 63110;
Mahidol-Oxford Tropical Medical Research Unit (MORU), Faculty of Tropical Medicine, Mahidol University, Bangkok 10400, Thailand;
Centre for Vaccinology and Tropical Medicine, Churchill Hospital, Oxford OX3 7LJ, UK
Correspondence to: Dr Rose McGready Email: rose{at}shoklo-unit.com
 | Summary |
|---|
 Top Summary INTRODUCTIONMETHODSSTUDY DESIGNRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Key Words: HIV • migrants • pregnancy • refugees • syphilis
| INTRODUCTION |
|---|
|
TopSummaryINTRODUCTIONMETHODSSTUDY DESIGNRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Burma has one of the largest and most long-standing problem of displaced people in Asia.14 In Tak province (Figure 1), Thailand, there are an estimated 150,000 migrant workers from Burma and 120,000 refugees living in camps along the Thai-Burmese border. Non-Governmental Organizations (NGOs) provide health care to refugees within the camps. Migrant workers access a variety of health facilities including Thai private and public health systems, the Mae Tao clinic and traditional healers. The purpose of this study was to determine the seroprevalence of HIV and syphilis, and assess the knowledge, attitudes, beliefs and behaviours surrounding HIV and STIs, to direct further programme control measures in refugee and migrant women on the Thai-Burmese border.
|
| METHODS |
|---|
|
TopSummaryINTRODUCTIONMETHODSSTUDY DESIGNRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Refugees from Burma began arriving in Thailand in 1984 and currently live in 10 major camps situated on the Thai-Burma border. Migrant workers in Tak province either hold official work permits (approximately 30%) or are considered illegal. This study focuses on pregnant women attending antenatal clinics (ANCs) in Mae La refugee camp or migrant clinic sites of Mawker Tai (MKT), Wang Pa (WPA) and Walley (WAL) (Figure 1). Mae La is the largest camp for refugees with a population of approximately 50,000 at the time of the survey in 2005. Karen is the predominant ethnic group. The Shoklo Malaria Research Unit (SMRU) has conducted weekly ANCs since 1986 and in migrant villages since 1998. Condom promotion has been an integral component of the ANC programme since 1996.
| STUDY DESIGN |
|---|
|
TopSummaryINTRODUCTIONMETHODSSTUDY DESIGNRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
In 1997, a cross-sectional survey of all pregnant women in Mae La refugee camp using non-specific seroreactivity against Treponema pallidum infection was performed using Venereal Disease Research Laboratory (VDRL) test and confirmed by a T. pallidum haemagglutination assay (TPHA). All pregnant women dually reactive to VDRL and TPHA were considered as having active syphilis. Consent was verbal and every woman had the right to refuse. All women with a positive result, and their partners, were appropriately counselled and treated.
Unlinked and anonymous HIV testing was conducted on frozen sera from a sample of Karen women who previously attended ANCs in Mae La camp. HIV seroprevalence was determined by testing previously frozen serum samples for HIV antibodies using enzyme-linked immunosorbent assay.
In 2005, a repeat cross-sectional survey of all pregnant women in Mae La refugee camp, using the same method as in 1997 was done. Anonymous HIV seroprevalence data were provided by the PMTCT (Prevention of Mother to Child Transmission) programme, which commenced in Mae La refugee camp in 2002. All HIV-positive antenatal women had access to antiretroviral treatment including triple therapy, or AZT plus nevirapine, depending on the CD4 count. At the time of this survey, the system was ‘opt-in’ with uptake rates in the region of 77%. Migrant women were not offered HIV testing through SMRU's ANCs due to a lack of funding.
Qualitative research interviews
In 2005, interviews were conducted with the two most senior local, trained counsellors heading the PMTCT programme in Mae La camp. Discussions focused on factors that might contribute to a low HIV/STI prevalence and behaviours that may lead to increased HIV transmission in the camp. The counsellors' responses were based on information gathered from discussions with thousands of pregnant refugee women attending the PMTCT clinics. The PMTCT programme was conducted as part of the antenatal care programme and all newly registered women attended group counselling, followed by one-on-one counselling before being tested. Questions from counsellors and from pregnant women were an integral part of these sessions. The counsellors were involved with all Mae La's HIV-positive pregnant women.
In 2005, focus group discussions were held with 17 groups of pregnant women attending ANCs (95 women), and three groups of HIV-positive women attending the monthly social support gathering for refugees with AIDS (14 women), in Mae La camp. Each focus group consisted of four to six refugee women and discussions were conducted in either Karen or Burmese language. Groups were separated by ethnicity (Karen, n = 12; Muslim, n = 4 and Burmese, n = 4 group) and parity as well as HIV status.
All discussions used a semi-structured interview guide that included six open-ended questions regarding HIV/AIDS and STI knowledge, attitudes, beliefs, behaviours and contraceptive use. All discussions lasted between 30–45 minutes, were taped and held in a private room. Discussion notes were taken actively by the facilitator (LK or MM) and the observer (KP). All notes taken in Karen or Burmese were translated to English following the discussions and verified from the recordings.
Quantitative knowledge, attitude, belief and behaviour survey
Following focus group discussions, women were encouraged to participate in a private room in a one-on-one survey consisting of 17 close-ended questions which are detailed in the Results. SMRU first conducted HIV/STI knowledge surveys in 1996 in the camps and these formed the basis of the survey.
Continuous normally distributed data were described by the mean (standard deviation) and non-normally distributed data by the median (min–max). Percentages were given for categorical data. Categorical data were compared using the chi-squared test or by Fisher's exact test, as appropriate. Odds ratio (OR) was calculated as a measure of risk of HIV by univariate indices. Student's t-test was used to analyse means, the Mann-Whitney test was used to analyse medians.
To assess independent predictors of HIV, a multivariate logistic regression model was fitted using the variables that were significantly associated with HIV from the univariate analysis. A two-sided P value of less than 0.5 was deemed to be statistically significant. Data were analysed using EpiInfo version 6, SPSS version 14 for Windows (SPSS Inc.) and STATA version 10.0 (StataCorp., College Station, TX, USA).
| RESULTS |
|---|
|
TopSummaryINTRODUCTIONMETHODSSTUDY DESIGNRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
In November 1997, a total of 404 pregnant Karen refugees were screened for syphilis in Mae La refugee camp. None of these women, (0 of 404) (95% CI 0–0.9) were VDRL-positive. In December 2005, 0.4% (three of 741) (95% CI 0.1–1.2) of pregnant women in Mae La were VDRL- and TPHA-positive for syphilis while the corresponding number in migrants was 0 (0 of 234) (95% CI 0–1.6) (Figure 1). One case of syphilis was HIV-negative and the other two were HIV-positive. Reported syphilis seroprevalence in published studies among pregnant women in Thailand were similar (0.2%).12 Data since 1998 from Burma suggest much higher rates of 9.1% (Figure 1).15 Mae Tao Clinic reported a rise in syphilis rates in pregnant women from 1.2% in 1999 to 2.5% in 200316 (Figure 1).
In 1997, 500 anonymous samples from Mae La refugee camp pregnant women tested for HIV revealed one seropositive sample 0.2% (95% CI 0–1.1). In December 2005, the seroprevalence of HIV was 0.4% (two of 500) (95% CI 0.1–1.4). Tak Province (2005) reported a rate of 1.3%.17 A rise in HIV was reported by Mae Tao clinic among women attending ANC from 0.8% to 2.2% from 1999 to 2005.18 Seroprevalence in ANCs in Burma was reported as 2.0% in 200419 (Figure 1).
Qualitative research interview
PMTCT counsellors
The two counsellors interviewed had a collective experience of 10 years and had counselled and screened a total of 7792 pregnant women from May 2002 to December 2005. They suggested a lower rate of HIV/STI in Mae La refugee camp than surrounding areas resulted from religious faith, isolation, conservative attitudes of the camp governing body, heavy punishment for intravenous (i.v.) drug use and a culture of obedience to elders. They reported sexual taboos: sexual intercourse was accepted only within the bounds of marriage (monogamy) and reprimands were enforced, e.g. forced marriage in the case of premarital sexual relations or refusing schooling if individuals were found to be sexually active or pregnant.
The counsellors identified three risk factors they thought could increase HIV transmission in the future. These include: leaving the camp to look for paid employment rather than serve camp authorities; an increase in sexual relations between women from the camp and non-camp residents; Mae La being a larger, less remote and less isolated camp than previous refugee camps. This has contributed to decreased control by camp authorities leading to access to DVDs and youth magazines with broader media content. These are sources of preoccupation of the young.
The responses from the 20 focus group discussions were obtained from 95 HIV-negative pregnant women and 14 HIV-positive women. There were obvious differences in demography between HIV-infected and uninfected women (Table 1). Women with HIV were older, had a higher risk of having had more than one husband (OR 1.4 [95% CI 1.1–1.8]) and of being multigravid (OR 1.1 [95% CI 1.1–1.3]). Their husbands also had a greater number of marriage partners and their level of education tended to be lower (non-significant). HIV was present in women of all ethnic groups.
|
Most primigravid groups, 75% (six of eight) gave no response when asked what STIs were, but 40% (eight of 20) of groups reported they were diseases that are sexually transmitted to women from husbands who have visited sex workers. False beliefs were expressed by 25% (five of 20) of groups. They stated that STIs were transmitted through sweat, sitting in the same place as someone who had discharge and sharing cigarettes. When discussing how pregnant women in Mae La camp contracted syphilis, most (55% [11/20]) had no idea and, 70% (14/20) had no theories as to why the prevalence of syphilis was so low.
Standard interview on knowledge, attitude, beliefs and behaviour of HIV-positive and -negative women
The level of knowledge of HIV and use of condoms for prevention was high (Table 2). Women with HIV performed better on basic HIV knowledge (Table 2); STIs were less well understood and i.v. drug use was low. Husbands of HIV-positive women were more likely to accept the use of condoms and to have been outside the camp for work than those of HIV-negative women. Tattoos were significantly more common in older women. The proportion of women who ever heard of gay, 35% (36/102) and lesbian, 42% (43/102) people, was surprisingly low.
|
4, AOR 7.9, 95% CI 1.5–42.5). In the husbands, i.v. drug use could not be evaluated in the model since the numbers were too small. | DISCUSSION |
|---|
|
TopSummaryINTRODUCTIONMETHODSSTUDY DESIGNRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Monogamy was cited as a reason for low rates of HIV and STI, yet more than one marriage, even if monogamous, was still one of the highest risk factors for HIV. A serious gap in knowledge and practice was identified: while women were good at answering how HIV and STI were transmitted at interviews, the majority floundered during the focus group discussion when asked a less direct question about how Karen women in the camp who were HIV-positive may have acquired the infection. Senior counsellors suspect that women fail to personalize the fact that the sexual intercourse which led to their pregnancy puts them at risk of HIV, believing monogamy will protect them. Alternatively, women may not have wanted to comment about others in the camp or they may not understand the biology of HIV transmission. Nevertheless, this survey has led to a shift in the PMTCT programme within Mae La refugee camp in order to overcome this gap, to ‘opt-out’ of HIV testing, rather than to ‘opt-in’.
Migration, identified here as seeking work outside the camp, is a co-factor for HIV infection in south-east Asia9,11 and was identified both by the PMTCT counsellors and by the women themselves. Working outside the camp is a new context from the traditional, controlled and protective environment of the refugee camp or a rural village in Burma. The low acceptance rate of condoms in HIV-negative women makes them vulnerable. Clearly future programmes to prevent HIV and STI will need to include males. Incorrect beliefs regarding HIV transmission are a worrying and recurring theme. A survey in Mae Sot confirmed low levels of HIV knowledge and incorrect beliefs in Burmese migrant workers.21 While literacy levels remain low (<50% in pregnant women in cross-sectional surveys) belief in these falsehoods will probably persist.
Unfortunately, the model for risk factors for HIV seropositivity was affected by the small sample size and low number of events. Conclusions drawn from the model must be considered cautiously. Certainly, there were no surprises in which factors were associated with HIV as they have all been reported previously. The relative ORs, however, need confirmation by modelling with higher numbers of patients. The practice of tattooing is hundreds of years old among the Karen. In some ways it is surprising that HIV rates were very low, given the high percentage of people with tattoos and the apparent strong association they had with HIV. The history of where the people got their tattoos done was not detailed in this study. The counsellors reported that women would have got these tattoos in their villages in Burma and not in the refugee camp and this is reflected by the fact that tattoos are less common in younger women.
A recent multicentre trial concluded that serological screening should be performed early in pregnancy even in low prevalence populations and repeated in the third trimester.12 This was a highly cost-effective strategy at relatively low prevalence rates, e.g. 1% congenital syphilis.12,22 The cost to screen once for syphilis in a population of 1500 pregnant women (estimated camp pregnant population per year) is US $2450 (THB 92,310) or US $1.60 per women, while the cost of successfully treating woman, partner and neonate to prevent congenital syphilis is US $1.20 at a 1% incidence rate (26 US$ [THB 996] per single women, man, neonate unit). In 2006, the funds projected to be available per refugee on the border per year were US $161 (THB 6272 baht, at an exchange rate of 39 THB/US $) for basic shelter, food and sanitation.23
This study found lower rates of syphilis and HIV in pregnant refugees and migrants living on the Thai-Burmese border than reported from surrounding major centres. Programmatic changes including: a change to ‘opt-out’ system of HIV testing, routine syphilis testing for those diagnosed HIV-positive and routine STI education within the PMTCT programme, have been adopted. Pending funding syphilis and HIV testing should be conducted in all pregnant refugee and migrant women.
| ACKNOWLEDGEMENTS |
|---|
|
TopSummaryINTRODUCTIONMETHODSSTUDY DESIGNRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Financial support for this study was received from the University of Calgary, Faculty of Medicine, International Health Department, Calgary, Alberta, Canada. RM and FN are supported by the Wellcome Trust of Great Britain.
| REFERENCES |
|---|
|
TopSummaryINTRODUCTIONMETHODSSTUDY DESIGNRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
