Household and maternal risk factors for malaria in pregnancy in a highly endemic area of Uganda: a prospective cohort study

Malaria remains a major preventable cause of maternal morbidity and adverse birth outcomes in Africa, where an estimated 12.4 million pregnant women were exposed to malaria in 2010 [1]. Although most malaria infections during pregnancy remain asymptomatic in endemic areas, these infections are associated with maternal anaemia and poor birth outcomes including preterm birth, low birth weight (LBW) and perinatal mortality [2, 3]. For prevention of malaria in pregnancy the World Health Organization (WHO) recommends the use of long-lasting insecticide-treated nets (LLINs), intermittent preventive treatment (IPTp) with sulfadoxine–pyrimethamine (SP) and prompt diagnosis and effective case management. However, widespread parasite resistance to SP and mosquito resistance to the pyrethroids used in LLINs has led to concern over reduced efficacy of these interventions [4, 5]. Therefore, additional approaches to prevent malaria in pregnancy and improve birth outcomes are needed.

Social and environmental factors such as wealth [6] and house design [7] can be important determinants of malaria risk that may inform supplementary approaches to malaria control [8], but there are only a few examples of studies examining these risk factors in relation to pregnant women [9, 10]. Indeed, most observational studies of malaria in pregnancy have explored factors associated with uptake of anti-malarial interventions [11‐16] and perceptions of malaria in pregnancy [17, 18], as well as maternal risk factors for malaria in pregnancy [10, 19, 20]. In Uganda, it has been observed that younger and less educated women are at greater risk of malaria in pregnancy [21], while IRS and ≥ 2 doses of SP during pregnancy may offer some protection against adverse birth and maternal outcomes [5, 22]. In this study, maternal and household risk factors for malaria were evaluated in a high malaria transmission setting in Busia, eastern Uganda. This study is one of the first to examine the association between household characteristics and malaria in pregnancy in Uganda.

This study investigated the association between maternal and household factors and malaria in pregnancy in a rural, high transmission setting in Uganda. Gravidity was consistently and strongly associated with malaria throughout pregnancy, with primigravid women having an 84% higher risk of microscopic parasitaemia at enrolment, more than double the risk of microscopic parasitaemia following the initiation of IPTp and nearly three times the risk of placental malaria, compared to multigravid women. Similarly, IPTp with SP was associated with nearly a 60-fold higher risk of malaria parasitaemia and doubled risk of placental malaria, compared to DP. Of the household and other maternal factors assessed, belonging to the poorest households, living in traditional houses and having only primary or no education were associated with an increased risk of malaria in pregnancy, compared to the least poor households, modern houses and having at least O-level education. To the knowledge of the authors, this is the first study to evaluate household risk factors for malaria in pregnancy in Uganda.

Gravidity is a well-established risk factor for malaria in pregnancy in high-transmission areas, where multigravid women have a lower risk of infection due to the acquisition of immunity through consecutive pregnancies [2]. In contrast, as transmission declines the burden of malaria shifts from paucigravidae towards all pregnant women. The present finding that primigravid women were at a higher risk of malaria parasitemia and placental malaria than multigravid women, despite high coverage with IPTp, is consistent with the gravidity-related pattern observed in other high-transmission settings [26]. A strong association between IPTp regimen and malaria infection in pregnancy and placental malaria was also observed, as well as a lower prevalence of malaria infection following the initiation of IPTp. Indeed, the parent study found that DP provided almost complete protection against microscopic parasitaemia [23] and an earlier randomized controlled trial in neighbouring Tororo found that IPTp with DP reduced malaria in pregnancy more than IPTp with SP, although DP was not associated with a lower risk of malaria in infancy [27]. IPTp with SP is a primary intervention recommended by the WHO for preventing malaria in pregnancy. However, among 33 African countries that reported on IPTp coverage levels in 2017, only an estimated 22% of eligible pregnant women received the recommended three or more doses [28]. The present study reiterates the need to maintain high effective coverage with IPTp to protect women in all pregnancies, as well as identifying and using the most efficacious regimen, such as DP, where possible.

Malaria in pregnancy remains a major public health challenge in Uganda and these findings suggest that within communities the risk is highest among the poorest women. A study in neighbouring Tororo district found an association between malaria in children and socioeconomic position, a relationship hypothesized to be explained by causal pathways that include access to and uptake of interventions and treatment-seeking behaviour, nutrition, housing conditions and education [29]. In the present study, modern housing, higher education levels and bed net ownership at enrolment were all associated with greater wealth. Importantly, an association between malaria and poverty was detected even after controlling for IPTp regimen, indicating that social and economic conditions may be relevant even where protective interventions are in place. This suggests a concerning cycle in which children born to less socioeconomically advantaged mothers may have worse health outcomes at birth and subsequently in later life, than those born to more advantaged mothers.

In this study, traditional housing and lower educational attainment were also associated with an increased risk of malaria infection during pregnancy compared to modern housing and more education. Housing quality can affect malaria risk through its effect on house entry on the primary malaria vector Anopheles gambiae, which mainly bites people indoors at night-time [30]. The relationship between house design and malaria in the general population has been well-studied [31], but few studies have evaluated its association with malaria in pregnancy. A recent meta-analysis, however, revealed a strong linear relationship between malaria infection in children and in pregnant women [26], so these groups plausibly have overlapping risk factors. Meanwhile, education may offer a protective effect against malaria through increased knowledge of the disease and use of preventive interventions such as LLINs. Studies elsewhere have shown women’s education status to be associated with the risk of malaria among children [32], including in Uganda [33]. Supplementary approaches to prevent malaria in pregnancy, such as housing improvements and education initiatives, may therefore merit consideration.

This study had several limitations. First, self-report was used to measure the variables used to calculate the wealth index and educational attainment. The wealth index is also an imperfect metric and influenced by the variables included. Indeed, the observed relationship between housing, education and malaria is likely to have been affected by residual confounding by household wealth. Second, since the study was a secondary analysis of data from a randomized controlled trial, it was not possible to assess a complete range of risk factors for malaria in pregnancy, such as maternal marital status and occupation, as well as certain household characteristics such as screening of airbricks. Third, the findings of this study may not be generalizable to other settings with different malaria transmission profiles. Fourth, maternal age was not included in the analysis, since this was highly co-linear with gravidity, but younger maternal age (especially adolescence) is a known risk factor for malaria in pregnancy due to the lack of age-associated immunity [2]. Finally, the observed associations between household and maternal factors and malaria risk are not evidence of causality [34]. For example, it was found that the poorest women had a 24% greater risk of microscopic or submicroscopic parasitaemia than the least poor, after controlling for factors including IPTp regimen. However, this may plausibly be explained by the direct and indirect costs of malaria contributing to poverty, lower educational attainment and the use of more affordable building materials. Nonetheless, to the knowledge of the authors, this study provides the first evaluation of household risk factors for malaria in pregnancy among Ugandan women.

This study provides evidence that the risk of malaria in pregnancy in Uganda is highest among primigravid women and those belonging to the poorest households, living in traditional homes and with the least education. Alongside efforts to maintain high coverage with IPTp, LLINs and prompt and effective diagnosis and treatment, housing improvements and education initiatives could be explored as supplementary approaches to reduce malaria in pregnancy.