Incidence of clinical malaria, acute respiratory illness, and diarrhoea in children in southern Malawi: a prospective cohort study

Worldwide, the three most common causes of death of under-five children are pneumonia, diarrhoea and malaria [1]. In sub-Saharan Africa, an estimated 2.5 million children under-five years old die each year from one of these three preventable and treatable diseases [2]. High population density, poor housing and sanitation infrastructure, inadequate knowledge, malnutrition, and poor access to quality health, education, and employment all contribute to the risk of pneumonia, diarrhoea and malaria in this region [3]. Addressing these contributing factors may reduce the occurrence and therefore, morbidity and mortality associated with these diseases in rural communities. Reducing morbidity and mortality from infectious diseases has potential economic benefits: there is no income loss due to treatment and death costs, and there is an increase in productivity because healthy people are better able to participate in economic activities that will improve their lives [4, 5]. In addition, scaling up interventions such as vaccines (pneumococcal and rotavirus), insecticide-treated mosquito nets (ITNs), structural improvements of houses, and nutritional supplementation may further reduce the incidence of these diseases [6‐9].

Uptake and effectiveness of existing efficacious health interventions, in all communities, requires community buy-in. To improve the uptake and effectiveness of health interventions and promote positive health behaviour in communities, several strategies have been used in the past. These strategies include active community engagement and participation, information, education, and communication (IEC), communication for behavioural-impact (COMBI), behaviour change communication (BCC), and school-based health education [10‐15]. In any setting, health promotion and community engagement strategies should consider factors such as culture, pre-existing beliefs, community structures, literacy levels and education, communication language, transportation options, and availability of resources for personnel support and effective programme operation [16].

A community participation strategy was utilized to increase awareness and participation in malaria prevention and control in Chikwawa, Southern Malawi [17]. This community participation strategy included activities such as leadership training for local volunteers, known as health animators, who were tasked with conducting IEC on malaria and involving the local community in the implementation of malaria control strategies using existing community structures. The goal of this community-based approach was to influence a change in mind-set of the entire community to promote self-reliance through understanding malaria and its control [18, 19]. The approach was led and integrated into the community by building capacity within the communities in malaria prevention and control using regular community malaria workshops and sensitization [17]. Complementary malaria control strategies, such as house improvement (HI) and larval source management (LSM), were introduced in the community in addition to the government-recommended ITNs to achieve a significant reduction in malaria burden [20].

This study aimed to determine the trend of incidence of clinical malaria, acute respiratory infections (ARIs), and acute diarrhoea in under-five children residing in rural communities where a community-based malaria prevention strategy, using HI and LSM, was implemented in addition to government-recommended ITNs. The study also identified risk factors associated with clinical malaria, ARI and diarrhoea. Investigation of these conditions within the under-five population in a rural setting could provide a more holistic approach to the treatment and control of all conditions, especially in sites that are remote and hard to reach.

A total of 281 children were recruited for the study out of 325 children who were from sampled households. There were 29 children aged > 48 months and 15 guardians refused consent. Seven of the 281 children had incomplete data and were excluded from the analysis; the remaining 274 children from 254 households had complete data and were included (Fig. 1); 250 (91.2%) children completed all the 12 months of follow-up. Overall, participants who were lost to follow-up were 24 (8.8%), some of whom had withdrawn consent and others had relocated out of the study catchment area.

The median age of the children was 25 months with 50.7% of the children aged between 24 and 48 months (IQR 16.0–35.0); 52.6% were female (Table 1). The mean Hb was 11.0 g/dl and 47.7% of the children had Hb above 11.0 g/dl. There was a high prevalence of stunting (33.5%). A total of 12 children tested positive for malaria by RDT at recruitment; 58.7% of households reported owning at least one ITN for sleeping.

In this prospective cohort study in a rural area with a community engagement initiative on malaria prevention, overall incidence rates of clinical malaria, ARIs and diarrhoea were found to be fewer than 1 case per child-years at risk. There were significant temporal variations in disease incidence of all the diseases. The results of temporal measurements show when cases of malaria, ARIs and diarrhoea occurred. Previous estimates of malaria incidence within the same area are available, however no previous reliable estimates of ARIs and diarrhoea incidence from cohort studies in this area, or in Malawi in general, are available. Estimates of incidence based on LMIC data suggest an impending epidemic of malaria and pneumonia in low and middle-income countries (LMIC) [28, 30] and reports of deficiencies in care raise the concept that young lives are being unnecessarily lost.

Findings in this study suggest the incidence of clinical malaria was 2.5 times (1.2 cases per child-years at risk vs 0.5 cases per child-years at risk) lower than a previous study conducted between 2015 and 2016 in the same area [31]. Before 2016, household ownership of ITNs in the study site was at 29% with a majority having damaged or no ITNs at all [25]. During the current study, a mass ITN distribution campaign, community participation and mobilization, and community-led malaria interventions had been implemented [19, 32]. Studies from Kenya, Rwanda and Uganda have shown the significance of community participation in the formulation of appropriate measures towards malaria control [33, 34]. These interventions may have contributed to the decrease in malaria incidence. However, it should be noted that household ownership of at least one ITN in this community (58.7%) during the time of the study was lower than the 2017 national aggregated estimate (82%) [35].

Despite this disparity, these figures demonstrate a remarkable improvement in ITN ownership at both the community and national levels as evidenced by the previous study [31]. Between 2015 and 2018, the incidence of malaria in Malawi has stagnated between 214 to 217 per 1,000 population at risk despite an increase in the national ITN coverage [30]. Despite this intervention's enormous success, residual malaria transmission cannot be addressed by ITNs or indoor residual spraying (IRS) alone, even at very high coverage [36, 37]. Their long-term viability is jeopardized by a widespread increase in insecticide resistance in the target species [38, 39], which could be the case in the Malawian scenario, where despite an increase in ITN coverage there has been stagnation in clinical malaria incidence. The combined effect of existing interventions with novel strategies involving environmental management such as LSM and socio-economic development through house improvement provides a non-insecticidal, complementary approach to increasing protection against mosquito bites [40, 41]. These supplementary interventions could help to halt malaria transmission by reducing and preventing human-vector contact within residential areas. District-level malaria incidence estimates are however unavailable but could have been better if available for comparison sake.

Malnutrition is one of the possible risk factors for malaria infection among the under-five population in this rural area. At recruitment, some children in the area were discovered to be stunted, underweight and wasted. These conditions increase the children's susceptibility to infection. The presence of anaemia could be another contributing factor. The common causes of anaemia in Malawian under-five children include malaria, deficiencies in iron, folate, and other micronutrients, intestinal worms, and sickle cell disease. Though not specific to malaria as elaborated above, trends in anaemia prevalence can reflect malaria morbidity and have been illustrated in how they respond to changes in the coverage of malaria interventions [42]. Few cases of anaemia were recorded in this study compared to the previous study [31], with a majority being mild anaemia cases, suggesting that the intervention coverage that began in 2016 may have had an impact on malaria, resulting in reduced anaemia cases.

Clinical malaria was unevenly distributed among children within different age-groups and within the focal areas and intervention arms. There were 194 (70.8%) children that did not experience any clinical malaria, 80 children had a total of 110 clinical malaria infections, with some experiencing a repetition of infections. The majority of these cases occurred in focal area B. There are different reasons why focal area B had more malaria cases and repeated infections than other focal areas. Firstly, the Shire River, being the largest river in Malawi, flows through the district, including focal area B. This promotes Anopheles mosquito proliferation [43]. Furthermore, due to other agricultural and economic activities being conducted in the area, there is a development of multiple mosquito larval habitats which include cattle hoof prints, rice paddies, brick-pits, and wells [43]. A high number and repeated malaria infections in focal area B suggests the presence of a hotspot with higher malaria transmission than the surrounding areas [44]. A malaria transmission hotspot is defined as a geographical area within a malaria transmission focus where transmission intensity surpasses the average level [44]. Hotspots serve as foci for malaria transmission in most areas, particularly those undergoing malaria elimination, highlighting the importance of establishing targeted control within these areas. In this area, this could be accomplished through targeted interventions to reduce the human infectious reservoir, such as reactive screening and treatment of individuals diagnosed with malaria at health facilities [45], proactive case detection, which involves screening people in hotspots at regular intervals [45], and mass drug administration where a full therapeutic dose of drugs are administered to a population without prior screening. Targeted vector control activities, such as increasing ITN and IRS coverage [46, 47] and larviciding [48] of mosquito breeding sites, can be carried out; however, these interventions are laborious and costly in this context. In this study, 29% of households reported having open eaves, a decline of 10% from the previous study. Closing open eaves in houses has been shown in previous studies to reduce mosquito entry and anaemia in children [9, 49, 50]. A reduction of incidence in this area generally could be attributed to the fact that most houses in the area had closed eaves during the community-led implementation, limiting infectious bites from malaria mosquitoes.

In this study, malaria incidence in southern Malawi was observed peaking during and immediately after the hot-wet season. An increase in the number of mosquito breeding sites during and post rainfall is likely the main contributor to an increase in incidence in this area. A similar pattern of malaria incidence was observed in a previous study in the same area [31]. There are various studies supporting the significance of weather factors in malaria. Studies previously done have shown that temperature and humidity affect mosquito breeding and larval and malaria parasite development [51].

Considering that the current study area is rural, the findings of this study show that the incidence of clinical malaria, ARIs and diarrhoea was low. It was found that, compared to a pre-intervention malaria incidence study, malaria incidence in under-five children had significantly declined. Integrating malaria control strategies as a community participatory initiative has the potential to reduce malaria incidence as it allows a better understanding of the disease and the way it can be controlled, as shown in this community. However, there is a need for community engagement to be as comprehensive as possible by incorporating other major causes of childhood morbidity, such as ARIs and diarrhoea, to improve understanding concerning all diseases. This study highlights the need for more comprehensive studies on malaria, ARIs and diarrhoea to develop more effective interventional strategies to prevent and treat these conditions that impose a significant public health and socio-economic burden in resource-limited countries. Monitoring the epidemiology of these conditions in children may aid in the planning of local service expansion, educational programmes, and preventive measures.