Respondent-Driven Sampling on the Thailand-Cambodia Border: Can Malaria Cases be Contained in Mobile Migrant Workers?
Respondent driven sampling methodology, for hidden populations, is an effective strategy to study the migrant populations from Myanmar and Cambodia on the Thailand-Cambodia border, write Amnat Khamsiriwatchara, Piyaporn Wangroongsarb, Julie Thwing, James Eliades, Wichai Satimai, Charles Delacollette and Jaranit Kaewkungwal in the Malaria Journal.
There is substantial population movement across the Thai-Cambodian border that is largely driven by economics. Migrants from both Cambodia and Myanmar settle for varying periods of time in Thailand, often in search of work. The International Organization of Migration reported that Thailand has attracted increasing numbers of migrant workers, mostly from neighbouring countries with over one million registered migrant workers entering the country since 2004.
Channels for migration, in particular labour migration, are defined by the policy of the destination country, usually in response to the demand of domestic labour markets for foreign workers. When the supply through established channels does not match the demand, irregular migration dynamics develop, and migrants enter illegally and undocumented.
While various government ministries attempt to collect data on migrant workers, they usually have information on the number of registered migrants and those applying for work permits, but little information on the unregistered migrants. The true size of the migrant worker population in Thailand, in particular of irregular migrants, is notoriously difficult to quantify.
Rather than classifying migrant workers as documented or undocumented, the Thailand Ministry of Public Health defines migrants who have been in Thailand for more than six months as M1, and migrants who have been in Thailand for less than six months as M2.
Both M1 and M2 migrants are eligible to receive diagnosis and treatment for malaria free of charge at malaria clinics in border zones. Patients who cross the border for a day to seek treatment at the border clinics are counted among the M2. Migrants in Thailand account for a higher proportion of cases than Thai citizens, especially among the M2 migrants.
Little is known about migratory patterns along the Thailand-Cambodia border. However, given the hidden and mobile nature of these populations and the difficulty in generating a sampling frame, traditional sampling techniques such as cross sectional community-based surveys or time-location sampling are unable to produce an adequate and statistically valid sample.
Malaria surveillance in undocumented migrants is challenging, and ensuring treatment compliance and parasite clearance in an environment in which increased parasite clearance times need to be closely monitored is quite difficult in this population. While undocumented and highly mobile workers may receive diagnosis and treatment at malaria clinics along the border areas free of charge, they are often not followed up for compliance and parasite clearance, as is the norm in the provinces targeted by the containment project.
Innovative strategy with respondent driven sampling
Developing innovative strategies for follow-up is critical to the containment effort.
Little is known about migratory patterns along the Thailand-Cambodia border. However, given the hidden and mobile nature of these populations and the difficulty in generating a sampling frame, traditional sampling techniques such as cross sectional community-based surveys or time-location sampling are unable to produce an adequate and statistically valid sample. Respondent driven sampling (RDS) is a modified form of chain-referral or snowball sampling that has been used to sample hidden populations, and seeks to overcome the biases in traditional snowball sampling.
It uses a structured system of incentives to encourage recruitment by peers while limiting the number of individuals each participant can recruit, records the size of each participant’s network to weight the sample, and enables calculation of sampling error, thus allowing for inferences about the characteristics of the population from which the sample is drawn.
Pre-survey focus group discussions were held with migrants from Cambodia and Myanmar to assess the feasibility of using RDS. Discussions focused on migrants’ links to and relative size of social networks, willingness to participate, and potential barriers to participation such as limitation of local travel due to lack of transportation or poor roads, a burdensome work schedule, or fear of authority figures.
RDS starts with purposeful selection of a few members of the target population, or “seeds”, who are selected based on: (1) diversity of demographic and geographic factors; (2) diversity on key outcome variables; and (3) commitment to the goals of the study. Each participant receives an incentive for participating, and a preset number of coupons with which to recruit peers. The recruiter then receives a secondary incentive for each recruited peer who participates. The method continues as seeds recruit first-wave respondents, first wave respondents recruit second-wave respondents, and continues until the desired sample size is reached. Participants are encouraged to recruit randomly from their personal social networks. Data are collected on the size of the social networks, and the analysis is weighted based on the size of the social networks.
RDS has typically been used to study urban populations, among highly networked communities, such as injection drug users or commercial sex workers. While the migrant population faces similar issues of stigma, and those who are undocumented wish to remain undetected by authorities, questions were raised as to whether the networks in this population would be sufficient to support this methodology.
This analysis focuses on the understanding of mobility dynamics of different types of migrant workers in the border area as well as the application of RDS methodology to the migrant population; demographics, malaria prevention and treatment-seeking behaviour of this population.
Study area and population
Five study sites within three of seven provinces along the Thailand-Cambodia border were selected in areas where there were known to be many migrant workers and close by areas where P. falciparum resistance to artemisinins has been documented; three sites recruited for Cambodian migrants (one each in Chantaburi, Trat, and Sa Kaeo provinces) and another two sites for Myanmar migrants (both in Trat). The survey was conducted from June to September 2009.
Results – migration patterns
Migrants from Myanmar had a longer duration of residence in Thailand than migrants from Cambodia, with a median duration of residence of 87 months in M1 migrants from Myanmar compared to 62 months in M1 migrants from Cambodia. Among M2 migrants, those from Myanmar had been in Thailand 4.5 months, compared to 2.6 months for those from Cambodia.
Long-term migrants were more likely to have crossed the border by themselves (among Cambodians, 49% of M1 vs. 35% of M2, and among those from Myanmar 45% of M1 vs. 34% of M2). Long-term Cambodian migrants had been helped by relatives (19%), friends (16%), and employers (27%), while 57% of short-term Cambodian migrants were helped by employers.
Among migrants from Myanmar, 32% of M1 and 41% of M2 had used the services of a broker, compared to 1% of Cambodians. Migrants from Myanmar were also helped by relatives (19% of M1 and 26% of M2). Cambodian M1 migrants were least likely to have paid money upfront to come to Thailand (36%), compared to 53% of Cambodian M2, 58% of Myanmar M1, and 76% of Myanmar M2.
Among Cambodian migrants, 5% of M1 and 6% of M2 returned to Cambodia more frequently than once monthly. Among M2, 36% had not returned since their arrival, 32% returned every 2-3 months, and 26% returned every 6 months. Among M1, 21% returned every 2-3 months, 16% returned every six months, 23% returned once annually, and 29% had not returned since their arrival.
This pattern was very different for migrants from Myanmar; 85% of M1 and 100% of M2 had never returned to Myanmar. Only 2% reported returning as often as once per year. Of those who returned, the primary reason among long-term migrants from both Cambodia and Myanmar was to visit friends and family. Cambodian M2 migrants were more likely to return for traditional or national holidays.
Public transportation was not a popular option for Cambodians returning home; most travelled by personal or hired vehicle. While only 22% of M1 return at least every 2-3 months, 72% of M2 have been to Cambodia in the last three months.
In terms of future plans, of Cambodians, 32% of M1 and 68% of M2 planned to return to Cambodia. Only 4% of migrants from Myanmar planned to return. Among M1 migrants, 36% of Cambodians and 82% of those from Myanmar had no plans to move, and an additional 32% of Cambodians and 14% from Myanmar did not know. By comparison, among Cambodian M2, 22% had no plans to move and 9% did not know. The primary reason for an upcoming move was ‘work finished’ or ‘to live with family members.
Discussion and conclusion
The migration patterns demonstrated here have implications for containment of artemisinin resistance. There was a concern that migrants from Myanmar may carry the resistant parasite back to their country of origin, which is highly endemic, has not achieved the levels of malaria control that other Southeast Asian countries have, and has limited access by international agencies due to the political situation. However, those residing along the Thai-Cambodia border, primarily in Trat, have in fact largely settled in Thailand, and do not return often, if at all.
The Cambodian population in the three border provinces of Trat, Chantaburi, and Sa Kaeo differ in three important ways. First, there is considerably more frequent cross-border mobility, with even long-term migrants returning on a regular basis. Second, while migrants come primarily from western and central Cambodia, there has been migration from and back to almost all provinces in Cambodia, some with very low transmission levels and largely non-immune populations. Interestingly, no migrants claimed an origin in Pailin Province, which has typically been thought to be an epicenter of development of anti-malarial resistance.
Finally, while most migrants had been in Cambodia in the last three months, very few crossed the border more frequently than monthly. Whether this reflects the true reality of the situation or is due to sampling bias if the networks of those who cross more frequently did not intersect with those of the labourers in this sample, it seems as if the population resides in Thailand long enough to benefit from a treatment follow-up programme.
Frequent population movements, both across the Thailand-Cambodia border and from the border area across Cambodia, indicate the need for heightened surveillance for artemisinin tolerance outside what has been designated as the containment zone, as well as close cooperation amongst Thai and Cambodian authorities. This information on the mobility of cross-border migrants on the Thailand-Cambodia border will be valuable in planning effective malaria prevention and control for the border areas and help move forward the programme goals of containment and elimination of the disease.
This is a condensed version of the article that appeared in Malaria Journal 2011, 10:120 doi:10.1186/1475-2875-10-120
The full article can be downloaded from http://www.malariajournal.com/content/10/1/120