Malaria knowledge, preventive actions, and treatment-seeking behavior among ethnic minorities in Ratanakiri Province, Cambodia: a community-based cross-sectional survey

Malaria incidence and mortality have steadily declined in the Greater Mekong sub-region over the past decades [1]. In Cambodia, a marked decrease in Plasmodium falciparum (P. falciparum) has been observed since 2009; this coincides with the scale-up of the distribution of insecticide treated bed nets and the strengthening of Village Malaria Workers (VMWs) who provide early diagnosis and treatment of malaria for their villagers. The number of malaria cases has significantly decreased especially in western Cambodia (i.e. Pailin and Battambang Provinces) [2]. Given the successes observed in the malaria program, the Cambodian government is aiming to eliminate malaria by 2025 [3].

Although malaria transmission in Cambodia has been low recently, 21 of the 25 provinces are still endemic, and more than half of the population (approximately 8.6 million people) is still at risk of malaria. The risk of malaria transmission is the highest in the forest or forest fringe areas in the northeastern part of the country [4]. Although, four provinces in northeastern Cambodia (Rattanakiri, Mondulkiri, Steung Treng, and Kratie Provinces) have only 3% of the national population, they were reported to have had 43% and 25% of falciparum and vivax malaria cases in the country in 2013, respectively [2].

Ratanakiri Province consists mainly of rural, agricultural villages and has one of the lowest socioeconomic standards in the country [4]. Several studies have been conducted to examine the complexity of malaria epidemiology in the Province [5–7]. However, few studies have addressed the behavioral and social aspects of malaria. Therefore, little is known about malaria knowledge, preventive actions, and treatment-seeking behavior of ethnic minorities living in the Province.

In the elimination phase of malaria, complete termination of local transmission is required, especially in areas with foci of malaria transmission, which can remain as a source of infection [5]. In addition to appropriate infrastructure and sufficient resources at primary health care facilities, active engagement of communities to recognize malaria symptoms and take prompt treatment-seeking actions is crucial to reduce the reservoir of malaria parasites [8]. Therefore, communities need to be encouraged to take immediate actions to treat fever in order to promote early diagnosis and appropriate treatment of malaria [9]. Moreover, community knowledge, attitudes and practices should be examined to further promote proper malaria preventive actions and to discourage risk behavior among different ethnic minority groups.

The aim of this study was to examine malaria knowledge, preventive actions, and treatment-seeking behavior among ten ethnic minority groups and the Khmer group in Ratanakiri Province. Additionally, we aimed to identify risk factors for fever during the most recent pregnancy and among children under two. Results from this study will contribute to the development and implementation of future intervention programs to achieve malaria elimination in Cambodia.

Sixty-five mothers (17.2%) reported that their child had fever during the last 3 months. Nearly 60% of them sought treatment from public providers: about half (47.0%) from health centers and 12.1% from VMWs. About one out of seven mothers (13.6%) did not seek treatment. Out of mothers who sought treatment (n = 57), majority of them did so on the same day (31.6%) or the next day (49.1%) from the onset of fever. Less than half (42.2%) were sure that their child received blood test, and three reported positive results (two P. falciparum cases and one P. vivax case).

To our knowledge, this is the first study which reported knowledge and preventive actions for malaria and treatment-seeking behavior for fever among mothers with children under two in Ratanakiri Province, Cambodia. Although the majority of mothers used bed nets at home and wore long-sleeved clothes to prevent malaria, more than half of the mothers were not aware of malaria symptoms and transmission route. Staying overnight at a farm hut was significantly associated with having fever during pregnancy as well as a child having fever during the last 3 months. Mothers’ undertaking a variety of malaria preventive actions was found to be a protective factor against the development of fever among children. Our findings suggest that improvements are needed in treatment-seeking behavior as nearly half of the mothers did not seek treatment for fever during pregnancy, and one in seven mothers did not seek treatment when their child had a fever.

Major preventive actions against malaria were already undertaken among mothers in our study sites due to continuous efforts by the National Center for Parasitology, Entomology and Malaria Control, Cambodia. For example, the majority of mothers used bed nets at home and wore long-sleeved clothes to protect themselves from malaria. Also, two-thirds of the mothers reported that they brought hammock nets with them when they worked in the forests. However, actions to reduce the number of mosquitoes around their living environments were taken by fewer mothers. More than half of the mothers were not aware of malaria symptoms and its transmission route, and more than two-thirds of the mothers did not know where mosquitoes breed. To ensure malaria elimination, community engagement is the key; elements to increase community engagement include increased knowledge regarding the disease at individual and community levels [18]. Strengthening knowledge about malaria epidemiology and mosquito ecology can boost mothers’ actions for malaria prevention and vector control.

Staying overnight at a farm hut was identified as a significant risk factor for developing a fever during pregnancy and for children under two. It has been widely reported that agricultural practices are strongly associated with community health, including malaria incidence [19–21]. Recent studies using PCR diagnosis indicated that an overnight stay in a farm hut is a risk factor for malaria in Ratanakiri Province [5, 6]. Although this study focused on fever episodes, one of the symptoms of malaria, it is possible that malaria risk was increased by mothers’ staying overnight at a farm hut. This study provides behavioral and socio-cultural information regarding malaria as it includes data on different aspects of household characteristics and lifestyles, including educational background, ethnicity, wealth, family structure, travel distance to the nearest health centers, malaria knowledge, preventive actions, hygiene practices, and agricultural practices. These factors have not been considered in previous studies. Based on our findings, combined with those from two previous studies, it is strongly recommended that future educational and public health interventions for malaria elimination put high priority on farmers who stay overnight at farm huts in Ratanakiri Province.

In other Asian countries, the association between staying at farm huts and fever or malaria incidence is inconsistent. An entomological and epidemiological study conducted in north-west Thailand reported that malaria infection among farmers engaged in agricultural activities, including movement away from residences to farm huts for ploughing and planting for rice cultivation, was three times that of people living in the residential villages [22]. In Laos, staying overnight in a farm hut was not associated with an increased risk of malaria infection in a rural district, where farmers slept under insecticide-treated bed nets in the huts [23]. Therefore, local malaria epidemiology and a wide variety of socio-demographic and socio-cultural characteristics need to be taken into consideration when examining risk factors for local malaria incidence and/or fever.

We also found that pregnant women and mothers with small children in Ratanakiri Province need to be encouraged to take prompt actions to seek treatment for fevers. A striking finding was that nearly half of the mothers who had fever during pregnancy did not seek treatment. Low use of VMW services could be partly due to temporary suspension of the services in 2015 caused by the delay in payment process (Ly P., personal communication, December 21, 2017). The majority of pregnant woman and children, who received a blood test after running a fever, tested negative for malaria. However, it has been widely reported that pregnant women have an increased susceptibility to infection by Plasmodia spp. [24], and that malaria during pregnancy increases the risk of maternal death, miscarriage, stillbirth, and neonatal death [25, 26]. Additionally, early diagnosis and effective treatment of malaria is critical for communities to prevent further transmission and spread of the disease [27]. To ensure the elimination of malaria, communities at risk of becoming infected with malaria need to be aware of the importance of treatment-seeking behavior for fever in order to achieve complete and continuous termination of local transmission. Additionally, non-malarial fevers need either specific treatments or referral to health facilities for appropriate care to reduce maternal and child morbidity and mortality [28]. Therefore, it is strongly recommended that future educational interventions stress the importance of treatment-seeking behavior for fevers. Thus, the presence of community health workers such as VMWs in Cambodia, with whom villagers can promptly consult regarding their health issues, is important for malaria elimination.

The limitations of the study must be considered when interpreting these study findings. First, mothers were asked about health issues they experienced and the timing and frequency of their use of health services during their most recent pregnancy, which might have introduced recall bias. It could have resulted in underreporting of health issues and/or inaccurate reports of the number and timing of their health service use. Second, to examine preventive actions and treatment-seeking behavior, mothers’ self-reported data were used, which could have introduced courtesy bias. To minimize these biases and improve the accuracy of mothers’ self-reports about the timing and frequency of their use of health services, data were confirmed with a maternal health record book and a child health record book whenever available. Moreover, interviews were conducted by experienced and trained interviewers with on-site supervision. Third, sample size or power computation was not possible to be performed because of the lack of information on the number of mothers with children under two. Therefore, recruitment was performed to include as many mothers as possible in the study. However, we conducted a post hoc power analysis, and the sample size was determined to easily achieve a power of greater than 80% using the effect size we observed.

To our knowledge, this was the first study to examine malaria knowledge, preventive actions, and treatment-seeking behavior of mothers who belonged to ten ethnic minority groups and the Khmer ethnic group, both of whom reside in remote, agricultural villages in Ratanakiri Province, Cambodia. This study revealed that even though the majority of mothers took malaria preventive measures, knowledge of malaria epidemiology and vector ecology and treatment-seeking behavior for fever were limited. The result indicates the importance of spreading accurate knowledge and promoting prompt treatment-seeking behavior to strengthen community engagement for malaria elimination. Furthermore, staying overnight at a farm hut, regardless of the differences in socio-demographic and socio-cultural characteristics, was strongly associated with fever during pregnancy and childhood. The findings of this study will contribute to the development and implementation of an effective community-based educational program to improve malaria knowledge, raise awareness of health risks related to agricultural practices, and promote treatment-seeking behavior, which will contribute to the achievement of malaria elimination in the region.