Risk factors of malaria infection
The low altitude leads to hot temperature and adequate precipitation and then abundant breeding sites of anopheline mosquitoes in the China-Myanmar border area. All results of former investigations documented that people in low altitude areas were at a high risk factor of malaria infection. A retrospective case-control study suggested that independent risk factors associated with malaria infection were overnight in the lowland, foothill and half-hill areas, and near anopheline mosquito breeding sites in the China-Myanmar border area. [14]. A cross-sectional study reported that independent risk factors for slide positivity were age, lower altitude, lack of knowledge about malaria transmission and symptoms, inaction against mosquito bites and delayed treatment-seeking in the Salween river valley of Shan Special Region II (SR2), northern Myanmar [16]. In this study, the comparison of the multivariate LRA results between the subjects recruited in China and Myanmar indicated that the risk factors associated with malaria infection were mainly overnight out of home in the one month prior to illness, staying in rural lowland and foothill, staying at altitude <500m and streamlets≤100m in the border areas of Myanmar. For the subsample enrolled in China, the overnight in Myanmar was a part of overnight out of home. For the subsample enrolled in Myanmar, the Laiza city and nearby areas was altitude <500m and streamlets≤100m, staying in the Laiza city and nearby areas was therefore at a high risk of malaria infection. The LRA result of the overall sample indicated that no knowledge of malaria transmission was a risk factor of malaria infection due to the increased sample size. The proportion of case-patients with knowledge of malaria transmission (67.3%) was lower than that of controls (77.6%) (P=0.0055). This indicates that health education on malaria should be necessary for people in both China and Myanmar. The public health in China should maintain people’s malaria knowledge and vigilance, and remind people using personal protection against malaria infection when they are in the endemic areas of other countries. Cross-border workers should be educated on preventive measures for malaria through effective behavior change communication [25].
Malaria along China-Myanmar border and other parts in Southeast Asia
In most areas of the Southeast Asia, the year-round high rainfall and temperatures, and abundant malaria vectors lead to persistent and intense malaria transmission. Most parts of the GMS are low latitude and altitude, the year-round high temperatures and rainfall that are suitable for fertility of malaria vectors and persistent malaria transmission [32]. Forests are traditionally considered as a major determinant of malaria risk in the GMS [33, 34]. In Vietnam a high proportion of malaria cases and deaths were reported in the central mountainous and forested areas [35]. In Myanmar, most of malaria cases were reported to occur in forest or forest fringe areas, and loggers and gem miners were at high risk of malaria [36, 37]. However, the China-Myanmar border areas with altitude from 200m to 5128m are mostly mountainous. The low altitude areas have high temperatures, more mosquito breeding sites and malaria vectors [14]. After malaria was effectively controlled along the China-Myanmar border [10], the major malaria hot spots are the Laiza and its nearby areas of the KR2 [18-19], the Salween river valley of the SR2 [15, 16] and the small golden triangle at China-Myanmar-Laos border in the Mekong River Valley [4, 11] in the northern Myanmar. All the three malaria hot spots are at low altitude. The two former investigations [14, 16] and this study did not identify malaria in association with forest in the northern Myanmar. This suggests that the heterogeneity and complexity of malaria should be recognized and considered in planning control and elimination programmes in the GMS [38].
Malaria epidemiological characteristics along China-Myanmar border
In eliminating settings, malaria cases are increasingly male, adult, clustered geographically, imported among migrant and other hard-to reach groups, and caused by P. vivax [12, 14]. All of 152 case-patients recruited in China were imported among migrant, and most of them were male, adult and P. vivax malaria. In the chi square test, the proportions of male and aged ≥16 year old case-patients recruited in China were significantly higher than the proportions of these case-patients recruited in Myanmar (male, x2=7.4081, P=0.0065; age, x2=12.1295, P=0.0060). However, the proportions of P. vivax between two subsamples were not significant (x2=1.0111, P=0.3146). When the high percentage of P. vivax malaria indicated the success in control of P. falciparum malaria [10], it documents the difficulty in control of P. vivax malaria [18]. P. vivax malaria is one of challenges for malaria elimination in the GMS.
Implications
The WHO certification of malaria elimination requires applicant countries to provide evidence that 1) local malaria transmission has been fully interrupted, resulting in zero indigenous human malaria cases for at least the past three consecutive years (36 months), and 2) an adequate program for preventing reintroduction of malaria transmission is fully functional throughout the country [39, 40]. The last indigenous malaria case in China was reported from Yingjiang County in the Yunnan border area in April 2016 [41]. Malaria elimination in the Yunnan border area has strongly contributed to the remarkable achievement that the WHO certified China malaria-free status on June 30, 2021 [5, 11]. The results of this study helped public health decision-makers planning cost-effective strategies of malaria elimination gearing to the high risk location and populations [11]. However, the Yunnan border area is still challenged by reintroduction of malaria transmission. The threat of imported malaria from the border areas of Myanmar will continuously exist for a long time. With understanding the local risk factors of malaria infection, the ‘‘3+1’’ strategy for intensive surveillance, rapid response and border collaboration for malaria elimination was developed to reduce the threat of imported malaria for malaria elimination and prevention of malaria transmission reintroduction in the Yunnan border area [11]. Reduced border collaboration of malaria had ever led to slightly resurgent malaria in some border areas of Myanmar since 2014 [11, 15, 18]. The Laiza and nearby areas of the KR2 with a population of approximately 30 thousand persons is one of the malaria hotspot areas along China-Myanmar border due to the risk factors presented in this paper and the weak health services [11]. The number of malaria cases increased from 518 in 2013 to 2,367 in 2016 in the Laiza and nearby areas. The strengthened collaborative interventions between China and Myanmar during 2017-2019 reduced the number of malaria cases to 274 in 2019 [11]. However, reduced collaborative interventions due to the coronavirus disease 2019 pandemic led to malaria resurgence again. A total of 1,532 cases were reported in the Laiza and nearby areas of KR2 in 2021. This led to most of malaria cases in Yunnan were imported from the Laiza and nearby areas in 2021. The Yunnan borders Myanmar’s five special regions that are out of the central governmental management of Myanmar Union. The military conflicts and the weak health services cannot effectively control malaria in the five special regions. In cross-border collaboration of malaria control, the risk factors presented in this paper can be considered to plan cost-effective strategies.
Limitations
This study has certain weaknesses, which might be the limitations of the case-control study itself too. 1) To mitigate the confounding effect, controls were matched with age, sex, healthy status and health facilities. The matching criteria make the study lose the chance to identify whether they are an independent risk or confounding factors of malaria infection. 2) Because the Laiza and nearby areas are the main malaria hot spot that exported most of malaria cases (>80%) detected in Yunnan Province, this study only recruited the subjects from the Laiza city hospital in Myanmar. Due to malaria scarceness, the subjects were enrolled from 37 health facilities in China. The large difference in the number of health facilities involved in this study between China and Myanmar is one of weakness. However, the case-patients recruited in China were imported from most parts of the border areas of Myanmar. This would not lead to selection bias. 3) Owing to a few of non-malaria febrile patients with overnight history in Myanmar, among the 304 controls recruited in China, 259 (85.2%) had no overnight history in Myanmar in the one month before attendance of health facilities. The high proportion of non-malaria febrile patients without overnight history in Myanmar might influence assessment of other potential risk factors. However, this can document that overnight in Myanmar is an obvious risk factor for Chinese people. 4) The subjects were only recruited in the health facilities. If some patients just bought anti-malarial drugs from the drug stores for malaria treatment and the self-medication worked, they might not seek diagnosis and treatment in the health facilities. This occasion might lead to exclusion of them from the study and selection bias. However, with reduction of malaria burden, anti-malarial drugs are disappearing from drug stores because of losing chance of making money. This leads to anti-malarial drugs mainly available in the public health facilities in China. The chance of the selection bias should be small. 5) Some participants declined to answer certain questions that they thought of as sensitive, this might cause responding bias. 6) To avoid recall bias, most of variables were defined to collect information in the one month prior to attendance of health facilities. This may miss to collect enough valuable information.