S. Sirilak, P. Singhasivanon, N. White, and F. Nosten conceptualized and designed the project. V. I. Carrara coordinated the cross-sectional surveys, interpreted the results of all study components, and drafted the manuscript, with F. Nosten. J. Thonglairuam managed the data entry of the public health data and conducted preliminary results of morbidity and mortality. A. Brockman coordinated and analysed the in vitro studies, E. A. Ashley the in vivo studies, and R. McGready the studies related to pregnant women. S. Proux supervised and controlled the quality of all laboratory results. V. Gilbos developed and supervised the Information Education Communication programme. C. Rojanawatsirivet, S. Krudsood, S. Leemingsawat, and S. Looareesuwan coordinated and analysed part of the entomological data. All authors revised the manuscript for intellectual content.
NW is chairman of the World Health Organization malaria treatment guidelines committee and is on the editorial board of
Early diagnosis and treatment with artesunate-mefloquine combination therapy (MAS) have reduced the transmission of falciparum malaria dramatically and halted the progression of mefloquine resistance in camps for displaced persons along the Thai-Burmese border, an area of low and seasonal transmission of multidrug-resistant
Existing health structures were reinforced. Village volunteers were trained to use rapid diagnostic tests and to treat positive cases with MAS. Cases of malaria, hospitalizations, and malaria-related deaths were recorded in the 6 y before, during, and after the Tak Malaria Initiative (TMI) intervention. Cross-sectional surveys were conducted before and during the TMI period.
In the remote malarious north western border area of Thailand, the early detection of malaria by trained village volunteers, using rapid diagnostic tests and treatment with mefloquine-artesunate was feasible and reduced the morbidity and mortality of multidrug-resistant
Artesunate-mefloquine therapy and rapid diagnosis were introduced in five border districts in northwestern Thailand. Cases, hospitalizations, and malaria-related deaths were recorded before, during, and after. Reductions were achieved in morbidity and mortality from multidrug-resistant
Malaria kills about a million people worldwide every year. Most of these deaths are in children. One of the most serious problems in the battle against malaria is that the parasites that cause the disease are able to change and become resistant to the drugs used to treat it. Widespread drug resistance in the parasite that causes the most serious type of malaria,
Along the Thai-Burmese border there are a number of camps for displaced people (refugees). In a malaria programme for people in the camps, efforts were made to diagnose cases early, and people found to have malaria were then treated with a combination pill, containing artesunate (a derivative of artemisinin) and another drug, mefloquine. A study of this programme found that it made a big reduction in the number of cases of the most serious form of malaria, and suppressed the advance of resistance to mefloquine.
It was decided to introduce the same malaria strategy for all the estimated 450,000 people living in five border districts of the remote Tak province in northwestern Thailand. It was necessary for the impact of this larger programme to be studied to see whether this new approach to malaria control was effective in a remote area—not just in the particular circumstances of the camps, but on a wider scale.
Village volunteers were trained to identify cases early, using rapid diagnostic tests. Health facilities were improved and increased in number, so that the combined treatment could be given to those who needed it. The researchers conducted surveys of malaria cases before, during, and after the programme. They recorded the type of malaria test used and the result, and the treatment each person received, whether they were treated at health centres, at village health posts, or by mobile teams. They found that cases of malaria caused by
The reduction in the number of deaths shows that the new approach to treatment is more effective, and the drop in the number of cases shows that the transmission of malaria was also reduced. This means that, even in a remote area, a programme that involves both improved diagnosis and the combined artesunate-mefloquine treatment can have major benefits for the people living there. This is important not just for Thailand but also for other parts of the world where malaria is a problem.
Please access these Web sites via the online version of this summary at
•
• The US Centers for Disease Control and Prevention has
• MedlinePlus brings together authoritative information from the US National Library of Medicine, National Institutes of Health, and other government agencies and health-related organizations;
• Thailand's Ministry of Public Health has a Web site that includes
• The organization that spearheads the research described in this article is the
The development and spread of multidrug-resistant
The first focus of chloroquine resistance in
The effects of deploying the MAS3 combination on malaria incidence was analysed among Karen displaced people living in refugee camps in Tak province from 1986 to 1997. A 6-fold reduction in
In order to determine whether similar results (obtained in the well-controlled environment of camps for displaced persons) could be obtained in an open setting, a collaborative project between the Thai Ministry of Public Health, the Faculty of Tropical Medicine, Mahidol University, and the Shoklo Malaria Research Unit (SMRU, part of the Wellcome Trust–Mahidol University–Oxford Tropical Medicine Research Programme) was developed in the five border districts of Tak province (
The study area is shown in grey.
Tak province shares over 500 km of border with Myanmar and is divided into nine districts. The five districts selected for the TMI project are composed of rugged, hilly, and mostly forested terrain: Mae Ramat (MRM), Mae Sot (MS), Phob Phra (PP), Tha Song Yang (TSY), and Umphang (UMP). The remaining four districts of the province have virtually no malaria. The climate is tropical with a 6-mo rainy season (from May to early October). Mean annual rainfall varies between 1,400 mm in the southern and central areas and 2,300 mm in the northern district. The mean temperature ranges from 20.2 °C in December to 29.3 °C in April, and the annual relative humidity is above 75%.
The population at risk of acquiring malaria and targeted by the TMI project can be divided into three groups: (1) Thai citizens (approximately 300,000, of which half are Thai and the other half belong to ethnic minorities); (2) foreign nationals (FNs) (approximately 150,000, mainly migrant workers from Myanmar). The Thai National Ministry of Welfare and Social Labour provides reliable estimates of the total annual number of FNs living in the province. It includes registered workers with work permits, a mainly stable population estimated at 50,000 annually for the past 5–6 y. The families of registered workers are usually not registered but relatively stable. The remaining FNs are considered illegal, and are mostly temporary workers; and (3) displaced persons (65,000 mainly Karen displaced people living in semi-open camps in TSY, PP, and UMP (Thai Burma Border Consortium, unpublished data). Health care in the camps is provided by medical International Non-Governmental Organizations and food is supplied by a consortium of charities.
Most of the detailed information on entomology, epidemiology, and impact of ACT deployment in this area before the TMI intervention derives from observations in the camps for displaced persons [
The Thai National Malaria Control Programme (Malaria Division) is a vertical programme and functions in the five border districts of Tak Province through a network of 37 malaria clinics. Light microscopy of Giemsa-stained thick and thin blood films are used routinely in all malaria clinics for the diagnosis of malaria. Vector control consists of indoor-chemical residual spraying (DDT until 1997, then subsequently 5% deltamethrin) once or twice a year, and impregnation of bed nets with 10% permethrin 30 mg/m2 every 6 mo. Patients with a malaria smear positive for
In the Karen camps, the control of malaria is done by medical International Non-Governmental Organizations assisted by SMRU. The Malaria Division provided indoor-residual spraying in May 1998, March 2000, and April 2001. The majority of the displaced population uses bed nets (mostly non-impregnated) provided by the Thai Burma Border Consortium. The World Health Organization recommended a 3-d regimen of artesunate (4 mg/kg/d) combined with mefloquine (none on the first day of treatment, 15 mg/kg on the second day, and 10 mg/kg on the third day) is used for the treatment of uncomplicated
The basis of the TMI intervention was to provide ED and treatment with an ACT to all exposed inhabitants of the five border districts. The estimated target population was 450,000 (excluding the refugee population). The people living in camps (population 65,000) already had access to ED and treatment with an ACT since 1994, so the efforts concentrated on the villages and the population of migrant workers. In August 2001, 120 medical personnel from the health centres and health posts received a refresher course in microscopic diagnosis and were trained to perform rapid diagnostic tests for malaria (using both Paracheck and Optimal). All 59 health centres and 33 health posts from the conventional health system were provided with diagnostic tools and antimalarial drugs for uncomplicated malaria. In addition, 100 malaria posts (MPs) were created in villages where health services were not readily accessible and in migrant worker settlements (
Patients presenting with symptoms of malaria were offered a rapid diagnostic test. Those with a positive test were treated immediately according to the national protocol described previously (MAS2).
Messages and posters advertising the location of the MPs and the identity of the MP workers were added during the second half of 2002 in all villages, and information campaigns using leaflets and radio messages in several languages (Thai, Burmese, and Karen) were conducted among villagers and migrant workers. All other activities usually run by the Malaria Control Programme (bed net distribution and indoor insecticide spraying) continued unchanged.
Despite the withdrawal of financial support for the TMI project in September 2002, all health centres continued to provide malaria diagnosis and treatment. Most of the MPs located in villages of TSY, UMP, and PP districts remained operational, offering diagnosis and treatment mainly during the two malaria peak seasons, as did the Mae Ramat mobile team.
Treatment protocols in the Karen refugee camps remained unchanged, and indoor-residual spraying with 5% deltamethrin was done once during the TMI period (March 2002), but not the year post-TMI.
Demographic characteristics, type of malaria test and its result, and treatment received were recorded for each person visiting a conventional health facility (health centres, health posts, malaria clinics), an MP in the villages, or a mobile service. Information on patients seeking care in hospitals was computerized by hospital clerks and then categorized as inpatient or outpatient. In all structures, a Thai citizen with ID card was recorded as Thai, all others were recorded as FN. A patient presenting twice for illness with the same malaria species within 1 mo was considered a treatment failure rather than a new infection because of the low incidence of malaria in the region.
Mid-year population estimates of Thai citizens by district were obtained from the provincial statistics office and provided the denominators for measuring the crude malaria incidence per 1,000 inhabitants. Malaria morbidity was represented as the annual malaria incidence per 1,000 persons and was calculated as the total number of malaria cases reported in a year (or as a yearly average of the pre-TMI period) and divided by the mid-year population of the same period. Results were recorded by malaria species and by district.
In the Karen camps (since 1986), weekly antenatal-clinic consultations have been available for all pregnant women. A thick and a thin malaria smear were done at each consultation. Annual incidence rates were calculated for
To detect relative changes in the transmission of the two species, cross-sectional surveys in villages and migrant communities (Burmese workers, with or without their family, living in settlements on Thai soil) were conducted during the rainy season in 2002 using microscopy, to calculate the ratio
The efficacy of mefloquine and artesunate given as MAS3 for the treatment of patients with uncomplicated
Antimalarial susceptibility of
Entomological surveys were conducted in two Thai villages and in Maela refugee camp. Indoor catching was done in a small one-room house similar in structure to most of the Karen houses. In each site, four men (two indoors and two outdoors) collected the mosquitoes that landed on their exposed legs. Mosquito collections were done monthly, 4 d/mo in the Thai villages, and weekly, 5 d/wk in Maela camp. Mosquito species were identified on site and the heads of Anopheline mosquitoes were dissected and analysed for sporozoite carriage. The presence of
Monthly climatic conditions (rainfall, mean temperature, and humidity) were obtained from MS and UMP meteorological stations and from the TSY unit for the period 1996–2003. Those data provided annual trends and seasonal variations.
The TMI project was approved by the Thai Ministry of Public Health. The cross-sectional surveys and the in vitro drug susceptibility testing were approved by the Ethical Committee of the Faculty of Tropical Medicine at Mahidol University, Bangkok. All drugs studies conducted by the SMRU have been approved by the Ethics Committee of the Faculty of Tropical Medicine, Mahidol University, the Karen Refugee Committee, and the Oxford Tropical Medicine Research Ethics Committee. Participation in all studies and surveys was entirely voluntary and participants could withdraw from the study/survey at any time.
Data were analysed using SPSS 10.0 software for Windows (SPSS, Chicago, Illinois, United States). Proportions were compared by
The Thai Citizen population of the five districts increased from 258,529 in mid-year 1996 to 300,396 in 2003 with an annual population growth of 2%. The total FN population was stable and estimated at 150,000. The refugee population fluctuated because of armed attacks until October 1999, at which time only three camps remained open, and was estimated at 61,000 at mid-year 2000, increasing to 65,000 during the TMI period, and 63,400 the following year.
Between 1996 and 2001 (pre-TMI period), the public health services reported an average of 67,113 malaria cases per year, with little year-to-year variation. Overall, twice as many
Malaria Annual Rates, Incidences, and Deaths in Thai and FN Populations
During the TMI period (October 2001–September 2002), the public services diagnosed 45,313 confirmed malaria cases, of which 25,159 were
In the year following the intervention (October 2002-September 2003), there was a further 29.9% (95% CI, 29.4–30.5) overall reduction in
The number of hospital admissions of all causes was reliably recorded only in Mae Sot hospital and was stable throughout the three time-periods. By contrast, there was a marked decrease in the number of admissions due to malaria (
In the Karen displaced population a significant decrease in malaria incidence also occurred, although it was less marked than that observed in the Thai and FN populations. In the pregnant women cohort, the incidence of
Malaria Annual Rates, Incidences, and Deaths in Karen-Displaced Populations
In all three population groups,
Distribution of Malaria Cases by Age and Sex, in Thais, FN, and in Maela Camp (TMI Period)
Between June and October 2002, 43 sites were surveyed. Overall participation was 80% of the expected population of villagers and 90% of the FN communities. An additional 760 Burmese persons living in Burmese villages opposite four of the screening sites were included in the results, and are further referred to as Burmese villagers. The prevalence and ratios of
The annual mean humidity, temperature, and rainfall did not change significantly in the area during the study period (October 1995–September 2003). Surveys in the villages took place between September 2001 and August 2002, and in Maela camp between June 2002 and February 2003.
Between October 2002 and February 2003, 17,246 mosquitoes collected in Maela camp and 5,575 caught in the villages were analysed by ELISA to detect sporozoite carriage. Sporozoite rates were low in all sites (0.21% in Thai villages and 0.23% in Maela camp for
Efficacy of the combination of mefloquine and artesunate (as MAS3) prior to the TMI period was 96.2% (95% CI, 93.6–97.8). During the TMI period, the cure rate assessed at 42 d in 79 patients was also 96.2% (95% CI, 89.4–98.7) and remained unchanged the year after the TMI period, at 96.1%, (95% CI, 93.4–98.8), in a study involving 210 patients.
Temporal trends in the geometric mean IC50 (nM/l) for mefloquine and artesunate in isolates from primary
Isolates from primary infections were collected at SMRU clinics between 1996 and 2003 and assayed for sensitivity to artesunate and mefloquine, IC50 geometric means given as nM/l (95% CI).
The beneficial effects of the strategy of ED and the 3-d regimen of ACT implemented in 1994 in the Karen refugee camps and described previously, have persisted for over 10 y. The incidence of
In this project, we assessed the implementation of the same strategy of rapid diagnosis and treatment with ACT in a much (seven times) larger population living adjacent to the Burmese border in Tak province.
During the TMI project, the overall number of consultations increased only slightly despite an increased diagnostic capacity, probably because of the transfer of consultations from the malaria clinics to the newly created village-based MPs. Changes in diagnostic tools from microscopy to a rapid test (that detects
In the refugee camp population, where the ED and treatment with ACT were deployed in 1994, we expected the impact of the TMI intervention to be less noticeable. Indeed, the rate of infective bites per person for
The large-scale deployment of the MAS has not modified the efficacy of the treatment, which remains extremely high, and there is no in vitro change in susceptibility to either drug, although there was a shift of the mefloquine IC50 in 2003 back to the values of 1996. This is of concern and could be related to the use of unprotected mefloquine in the area, as well as to the less effective regimen (using 2 d of artesunate, MAS2) deployed by the Thai Ministry of Health [
During the 2 y of this project we were unable to collect information on adverse effects of the treatment. However we have conducted studies in thousands of patients in this area and consistently found that the MAS3 regimen was well tolerated [
The decrease in malaria morbidity observed during, and sustained 1 y after the TMI project, the persistence of high cure rates of the combination therapy 10 y after its first introduction in the province, the persistently low sporozoite rate, and the rapid acceptance of newly established MPs, all indicate that it is feasible to extend early malaria diagnosis and provide adequate treatment even to remote communities. The impact of this intervention was seen rapidly (within a year of introduction) in this area of low and seasonal transmission. We were not in a position to perform a cost-benefit analysis of the TMI project, but the public health benefits were evident given the large reductions in morbidity and mortality observed. These results give further support to the large-scale use of ED and prompt treatment with ACT, a strategy that has shown similar results in KwaZulu-Natal [
(22 KB DOC)
(189 KB JPEG)
We thank the staff of the SMRU, the Thai Ministry of Public Health and of the Malaria Division for their help and support. We thank also the Meteorological Stations and the Provincial Town Hall for the climatic data and the population census of Tak province and the Ministry of Interior for authorizing the work of SMRU in the Karen Camps. We thank the villagers, the workers and their families, and the patients in SMRU clinics for their participation to the studies and surveys. We acknowledge the help of Karen Barnes, who critically reviewed the manuscript.
artemisinin-based combination therapy
confidence interval
double-site enzyme-linked pLDH immunodetection
early diagnosis
foreign national
50% inhibitory concentration
mefloquine-artesunate combination therapy
[number]-d MAS
malaria post
Mae Ramat district
Mae Sot district
Phob Phra district
Shoklo Malaria Research Unit
Tak Malaria Initiative
Tha Song Yang district
Umphang district