Relationship between unimproved household sanitation facilities and malaria infection among under-five children in Nigeria: insights from Malaria Indicator Survey 2021

Abstract

Background

Malaria is a leading cause of illness and death among children under-five in sub-Saharan Africa, with Nigeria bearing the highest global malaria burden. Despite interventions such as insecticide-treated nets and indoor residual spraying, malaria prevalence remains high. While housing structure and sanitation are recognized as risk factors in other African countries, their relationship with malaria infection among under-five children in Nigeria remains underexplored. This study investigates this association using nationally representative data from the 2021 Nigeria Malaria Indicator Survey (NMIS).

Methods

This study analysed data from the 2021 NMIS, including 1833 children aged 5–59 months (weighted sample size: 1,784,805,486) tested for malaria using rapid tests. Data on malaria prevention practices, household characteristics, and children’s blood samples were collected. The primary outcome was malaria test results (rapid diagnostic test, RDT), with the type of toilet facility as the main predictor. Covariates included age, sex, wealth index, maternal education, residence type, household construction materials, drinking water sources, type of mosquito nets, and mosquito net usage. Descriptive statistics and logistic regression analyses were conducted to assess associations, reporting adjusted odds ratios (aORs), 95% confidence intervals (CIs), and p-values < 0.05.

Results

Children from the poorest (aOR = 3.412, 95% CI: 1.798–6.477, p = 0.0002) and poorer households (aOR = 3.103, 95% CI: 1.714–5.617, p = 0.0002) had significantly higher malaria risk. Rural residence (aOR = 1.898, 95% CI: 1.318–2.734, p = 0.0006) and no maternal education (aOR = 2.003, 95% CI: 1.153–3.480, p = 0.0139) were also associated with increased malaria prevalence. Additionally, unimproved wall materials (aOR = 1.604, 95% CI: 1.061–2.425, p = 0.025) increased malaria risk. However, unimproved sanitation facilities were not significantly associated with malaria risk (p = 0.166).

Conclusion

Malaria risk among under-five children in Nigeria is strongly associated with socioeconomic factors, rural residence, and maternal education, but not unimproved sanitation alone. Efforts to reduce malaria prevalence should target broader social determinants through health education and socioeconomic interventions in Nigeria and other endemic regions.

Background

Malaria infection, caused by Plasmodium species, is a parasite-borne illness spread by mosquitoes. It remains the leading cause of morbidity and mortality among under-five children in sub-Saharan Africa [1]. Globally, the number of malaria cases increased from 245 million in 2020 to a projected 247 million cases in 84 malaria-endemic nations in 2022, with most of these spikes coming from nations in the sub-Saharan Africa region. In sub-Saharan Africa, there were 627,000 malaria-related deaths among children in 2020, and an estimated 241 million new cases [2, 3]. Nigeria has the highest malaria burden in the world, with over 173 million people at risk of infection, 51 million cases, and 207,000 deaths recorded annually (about 30% of the entire malaria burden in Africa [4]. The largest prevalence is observed among children between the ages of 5 and 59 months (under-five) in various geographic areas of the country [5].

Several studies have established a relationship between housing structure and the risk of malaria infection in Nigeria [6,7,8]. Housing type/structure has been implicated as an important risk factor for malaria infection among under-five children in Nigeria [6,7,8]. The likelihood of contracting malaria is influenced by home design [9, 10]. Non-improved housing has been established as a predictor of malaria infection among under-five children [6, 8]. Association between drinking water and sanitation use in relation to malaria prevalence has been established across multiple countries in Africa [11]. Specifically, the association between unprotected drinking water and houses with no sanitation facilities in relation to malaria prevalence has been established in countries like Angola, Kenya, Liberia, Madagascar, Malawi, Rwanda, Tanzania, Togo, and Uganda [11]. However, no study has explicitly examined the association between household sanitation facilities (unimproved sanitation facilities) and malaria prevalence in Nigeria. It becomes imperative as a major issue of public health to look at this association. Malaria prevention strategies, including insecticide-treated nets and indoor spraying, have been effective over the years. Despite this, malaria prevalence is still increasing. The relationship between household sanitation facilities and malaria prevalence among under-five children in Nigeria is still under-explored as hygiene is one of the most universal and straightforward measures to prevent disease transmission. Consequently, the impact of this relationship would substantiate existing knowledge and policy for the prevention and control of malaria transmission in Nigeria. The purpose of this study was to examine the relationship between household sanitation facilities and malaria infection among under-five children in Nigeria. The objectives of this research are (1) to assess the influence of unimproved household sanitation facilities and the increased risk of malaria infection among under-five children in Nigeria compared to improved household sanitation facilities and (2) to identify multiple influences such as sociodemographic, educational status, mosquito bed net use, type of mosquito net, household construction materials, source of drinking water, and type of residence on unimproved household sanitation facilities and increased risk of malaria infection.

Methods

Data source

This study is a secondary data analysis utilizing data from the 2021 Nigeria Malaria Indicator Survey (NMIS), a nationally representative cross-sectional survey conducted as part of the Demographic and Health Survey (DHS) program. The DHS provides standardized population health and malaria-related data across developing countries. The NMIS offers insights into malaria prevalence and control measures at both national and subnational levels, covering all 36 states and the Federal Capital Territory (FCT). Nigeria, located in West Africa, bears the highest burden of malaria cases in sub-Saharan Africa. The dataset includes survey data from both urban and rural areas, ensuring broad geographical representation, as illustrated in Fig. 1.

Fig. 1

Showing urban and rural population distribution in Nigeria

Sampling technique

The study sample for this analysis consisted of children aged 5 to 59 months who underwent malaria rapid diagnostic testing (RDT) as part of the 2021 Nigeria Malaria Indicator Survey (NMIS). A total of 1833 children were included in the dataset, with a weighted sample size of 1,784,805,486.

Households were eligible for inclusion if they provided complete data on sanitation facilities, household construction materials, drinking water sources, type of mosquito nets, and mosquito net availability. Additionally, the dataset included key demographic and socioeconomic variables, such as maternal education, household wealth index, and place of residence, allowing for a comprehensive assessment of malaria risk factors.

To maintain the integrity of the analysis, participants were excluded if they were five years or older at the time of data collection. Cases with missing, refused, or ambiguous malaria rapid test results were also excluded. Furthermore, households with incomplete data on essential predictor variables, including sanitation facilities, drinking water sources, household construction materials, and mosquito net usage, were not included in the final analytical sample.

The 2021 NMIS used for this study was the third of its kind, following earlier surveys conducted in 2010 and 2015. The 2021 survey had several unique features, most notably its timing during the first year of implementing the current National Malaria Strategic Plan. A two-stage sampling method was employed to ensure national representativeness. In the first stage, 568 enumeration areas (EAs) were selected using probability proportional to size, where the EA size was determined by the number of households within it. This design ensured representation from all 36 states and the Federal Capital Territory (FCT). A total of 568 clusters were identified, comprising 195 urban and 373 rural areas [12]. Household listings within these clusters were conducted between August 26 and September 18, 2021, and these listings served as the sampling frame for the second stage of household selection. During this process, global positioning system (GPS) devices were used to record the coordinates of the clusters [12].

In the second stage, a systematic sampling technique was applied, with 25 households selected per cluster using equal probability sampling. Data collection for the 2021 NMIS utilized computer-assisted personal interviewing (CAPI) technology. The survey included women aged 15–49 years as participants, all of whom provided informed consent and were aware of their right to withdraw from the survey at any time [13].

Description of variables

This study explores the relationship between household sanitation facilities and malaria prevalence using key variables relevant to the research objectives. The primary predictor variable, household sanitation facilities, was categorized based on the type of toilet facility used. Originally, response options included flush toilets (to piped sewer systems, septic tanks, or pit latrines), pit latrines (with or without slabs, ventilated improved pit latrines), and other forms such as bucket toilets, composting toilets, or open defecation. For analysis, flush toilets were classified as improved sanitation, while pit latrines without slabs, open defecation, and other forms were categorized as unimproved sanitation [6].

The primary outcome variable, malaria infection, was assessed using malaria rapid diagnostic test (RDT) results. The original response options included negative, positive, not present, refused, or other. The variable was recoded into a binary outcome: negative for participants who tested negative and positive for those who tested positive, with all other responses excluded from the analysis. Sociodemographic variables included sex, wealth index, education level, and place of residence. Sex was recorded as male or female and remained unchanged. Wealth index was categorized into five levels: poorest, poorer, middle, richer, and richest, reflecting participants' relative income and socioeconomic status. Maternal education was classified as no education, primary, secondary, or higher education, with cases marked as "don’t know" excluded. The place of residence was categorized as urban or rural without modification. Children’s age was originally recorded as a continuous variable (0–59 months) and was recoded into five age groups: 5–12 months, 13–24 months, 25–36 months, 37–48 months, and 49–59 months [13].

Household construction materials were categorized into improved and unimproved based on quality and durability. Improved floor materials included parquet or polished wood, vinyl or asphalt strips, ceramic tiles, cement, and carpet, while unimproved floors consisted of natural materials such as earth, sand, and dung, as well as rudimentary materials like wood planks and palm or bamboo [6]. Wall materials were classified as improved if made of cement, stone with lime or cement, bricks, cement blocks, covered adobe, or wood planks/shingles. Unimproved walls included natural materials (no walls, cane, palm trunks, dirt) and rudimentary materials (bamboo with mud, stone with mud, uncovered adobe, plywood, cardboard, and reused wood) [6]. Roof materials were similarly categorized, with improved materials including metal/zinc, wood, calamine/cement fiber, ceramic tiles, cement, roofing shingles, and asbestos. Unimproved roofing consisted of natural materials (no roof, thatch, palm leaves, grass) and rudimentary materials (rustic mats, palm/bamboo, wood planks, cardboard) [6].

Drinking water sources were classified according to the WHO/UNICEF Joint Monitoring Programme (JMP) framework [14]: safely managed water (piped into dwelling), basic water (protected wells, boreholes, public taps), unimproved water (unprotected wells, tanker trucks), and surface water (rivers, lakes, canals).Mosquito net usage among children was categorized based on the type of net they slept under: insecticide-treated nets (ITNs), untreated nets, or did not sleep under net. Mosquito net availability was assessed based on whether the household had at least one designated sleeping net.

Statistical analysis

The outcome variable is the result of the malaria rapid test, and the main predictor is the type of toilet facility. Data was weighted according to DHS guidelines. Variables were presented as frequencies and percentages. Descriptive statistics, bivariate, and multivariate analyses of the predictors and outcome variables were conducted while controlling for some selected covariates including sex, wealth index, mothers’ highest educational level, age in months, type of place of residence, wall materials, floor materials, roof materials, type of mosquito nets, source of drinking water, having mosquito net for sleeping, and mothers’ usage of long-lasting insecticide-treated net (LLIN). Logistic regression analyses were conducted to examine the association between each predictor and the outcome variable while controlling for possible confounders. A bivariate assessment of all the predictor variables was conducted to examine the association between each predictor variable and the outcome variable (the result of rapid test). The backward selection approach was selected for the multivariate analysis, odd ratios and 95% confidence interval were reported with p-value < 0.05 All analyses were done with statistical software SAS version 9.4 (SAS Institute Inc., Cary, NC, USA).

Results

Descriptive analysis

This study analysed data from 1,784,805,486 weighted participants aged 5 to 59 months. Among these, 63.4% tested negative for malaria, while 36.6% tested positive. Regarding sanitation, 39.4% of children lived in households with improved toilet facilities, whereas 60.6% resided in households with unimproved sanitation. The sex distribution showed that 48.7% were female, while 51.3% were male. Socioeconomic disparities were evident in the sample. 25.1% of children were in the richest wealth category, while 16.7% were in the poorest wealth group. Maternal education levels varied, with 34.8% of children having mothers with no formal education, 13.1% having mothers with primary education, 38.0% having mothers with secondary education, and 14.2% having mothers with higher education.

Children living in rural areas accounted for 68.2%, while 31.8% were in urban areas. The highest proportion of malaria cases was recorded among children aged 49–59 months (40.4%), while the lowest proportion was among children aged 5–12 months (4.6%). Regarding mosquito net use by mothers, 66.1% of mothers did not sleep under a mosquito net, while 33.8% reported sleeping under a net. Among under-five who used nets, 35.8% slept under untreated nets, 1.07% used insecticide-treated nets (ITNs), and 63.1% did not use any net. Housing conditions varied, with 34.1% of children residing in homes with unimproved wall materials, 38.5% living in homes with unimproved floor materials, and 14.1% in homes with unimproved roofing materials. Access to safe drinking water was also assessed, with 64.0% of children living in households with basic water sources, 24.4% relying on unimproved water sources, 7.1% using surface water, and only 4.5% having access to safely managed drinking water sources. These results are summarized in Table 1.

Table 1 Frequency table of predictors and the outcome variable

Bivariate and multivariate analyses

Results of the bivariate analysis and multivariate analysis are detailed in Table 2. The final multiple logistic regression model identified significant predictors of malaria risk, including wealth index, maternal education, rural residence, type of mosquito net, and housing conditions as shown in Fig. 2.

Table 2 Bivariate and multivariate analysis showing the association between the predictors and outcome variable

Fig. 2

Showing the forest plot of predictors of malaria using adjusted odds ratio

Children from the poorest households (adjusted Odds Ratio (aOR) = 3.412, 95% CI: 1.798–6.477) and poorer households (aOR = 3.103, 95% CI: 1.714–5.617) had significantly higher odds of malaria infection compared to those in the richest wealth category. Similarly, children whose mothers had no formal education (aOR = 2.003, 95% CI: 1.153–3.480) faced greater malaria risk than those with mothers who attained higher education. Children residing in rural areas had nearly twice the odds of malaria infection (aOR = 1.898, 95% CI: 1.318–2.734) compared to their urban counterparts. Additionally, under-five children who did not sleep under a mosquito net (aOR = 2.704, 95% CI: 1.034–7.073) were at significantly higher risk compared to children who slept under an insecticide treated mosquito nets.

Housing materials were also influential, with unimproved wall materials associated with increased malaria risk (aOR = 1.604, 95% CI: 1.061–2.425). Conversely, unimproved roofing materials appeared protective, with lower odds of malaria infection (aOR = 0.621, 95% CI: 0.392–0.985). However, unimproved floor materials and drinking water sources did not show significant associations in the adjusted model. These findings emphasize the importance of socioeconomic status, maternal behaviours, rural residence, and housing conditions in determining malaria risk. Interventions should focus on economic empowerment, educational improvements, and promoting mosquito net use as essential malaria control measures.

Discussion

This study examined the relationship between household sanitation facilities and malaria prevalence among under-five children in Nigeria using data from the 2021 Nigeria Malaria Indicator Survey (NMIS). The findings revealed that unimproved sanitation facilities were not significantly associated with malaria infection after adjusting for socioeconomic and demographic factors. This contrasts with previous studies in sub-Saharan Africa, such as the study by Yang et al. [15], which found that poor sanitation and unsafe drinking water increased malaria risk among children across multiple countries. However, broad regional analyses may not fully capture country-specific variations in malaria transmission dynamics. The lack of a significant association in this study suggests that unimproved sanitation alone may not be a primary driver of malaria transmission in Nigeria. Instead, factors such as poverty, rural residence, and maternal education appear to play a more influential role.

The initial bivariate analysis suggested an association between unimproved sanitation and malaria infection, supporting the hypothesis that inadequate sanitation may contribute to increased mosquito breeding and exposure. However, this association lost significance in the adjusted model, indicating potential confounding by wealth index, maternal education, and rural residence. The strongest predictors of malaria risk were socioeconomic status, rural residence, and maternal education, supporting prior research that links malaria prevalence to broader social determinants rather than sanitation alone [16, 17].

Children from the poorest households had more than three times the odds of malaria infection compared to those from the richest households. This aligns with previous studies highlighting poverty as a key determinant of malaria risk due to limited access to healthcare, preventive measures, and adequate housing [18,19,20]. Similarly, children whose mothers had no formal education were at significantly higher risk than those whose mothers attained higher education. This underscores the role of health literacy in malaria prevention, as educated mothers are more likely to engage in preventive measures, such as ITN use and early treatment-seeking behaviour [21, 22].

Rural residence was also a strong predictor of malaria infection, with children in rural areas nearly twice as likely to test positive as those in urban areas. This finding aligns with prior studies showing that rural populations face a higher malaria burden due to environmental conditions favourable for mosquito breeding, limited access to healthcare services, and reduced ITN coverage [23]. Many rural communities lack proper drainage, have open water sources, and are surrounded by vegetation conducive to mosquito proliferation. Additionally, inadequate access to malaria control interventions such as ITN distribution and timely diagnosis contributes to the persistent malaria burden in rural Nigeria.

Housing conditions were also significant, particularly wall materials, with unimproved walls associated with an increased malaria risk. Studies have demonstrated that housing improvements—such as sealed walls, screened windows, and durable roofing—play a crucial role in reducing mosquito entry and limiting malaria transmission. These structural modifications act as physical barriers, preventing mosquitoes from entering indoor spaces where they can bite and transmit Plasmodium parasites [24,25,26]. However, unimproved roofing materials were unexpectedly associated with a lower malaria risk. This may be due to unmeasured factors such as ventilation, differences in indoor residual spraying (IRS) coverage, or behavioral variations in mosquito avoidance. Further research is needed to explore these interactions.

A strong association was observed between mosquito net usage and malaria risk. Children who did not sleep under an ITN had significantly higher odds of malaria infection, reinforcing the effectiveness of ITNs in malaria prevention [27, 28]. This aligns with findings by Barker et al. [28], which demonstrated that ITNs, particularly those with dual active ingredients, significantly reduce malaria transmission. Efforts to improve ITN coverage should focus on proper utilization, addressing barriers such as discomfort and low adherence, while also monitoring insecticide resistance to ensure sustained effectiveness.

To effectively reduce the malaria burden in Nigeria, interventions should address social determinants of health while strengthening malaria control programs. Expanding evidence-based strategies could enhance malaria prevention and treatment efforts. Socioeconomic empowerment is crucial, as malaria disproportionately affects low-income households. Expanding programs like Nigeria’s National Cash Transfer Programme (NCTP) to include malaria-specific incentives, such as free ITN distribution, could enhance prevention efforts [29]. Strengthening rural healthcare infrastructure by expanding mobile clinics and increasing access to rapid diagnostic tests (RDTs) and artemisinin-based combination therapy (ACT) would improve early detection and treatment [30].

Housing improvements and environmental management are also essential. Studies show that improved housing structures with sealed walls, floors, and roofs reduce malaria transmission by up to 50% [24]. Government-backed housing subsidies for insect-proof materials and expanded drainage and waste management initiatives, modelled after Kenya’s Community-Based Malaria Elimination Programme, could further mitigate malaria risk [31]. Despite ITN distribution efforts, ensuring proper and consistent use remains a challenge. Community-based education programmes, as seen in Ghana and Tanzania, have improved ITN retention and use by addressing barriers like discomfort and misuse [32, 33]. Similar initiatives in Nigeria could enhance ITN effectiveness. Additionally, scaling up Indoor Residual Spraying (IRS), as successfully implemented in Zambia and Mozambique, could further reduce malaria transmission in high-burden areas [34].

Finally, health education plays a crucial role. Community-led programmes, such as Ethiopia’s Health Extension Programme, have improved ITN use, malaria symptom recognition, and healthcare-seeking behaviours [35]. Implementing similar initiatives in Nigeria could enhance malaria prevention and early treatment, ultimately reducing the malaria burden.

This study has several notable strengths that enhance its credibility, reliability, and generalizability. First, it is the first to examine the association between household sanitation facilities and malaria infection risk in Nigeria, addressing a critical research gap. Leveraging the 2021 NMIS—a nationally representative dataset—ensures that the findings reflect malaria risk factors across diverse geographic regions, including both rural and urban areas. The sample size and rigorous sampling methodology further strengthen the study’s reliability.

The study applies robust statistical analyses, including bivariate and multivariate logistic regression models, to control potential confounders such as socioeconomic status, maternal education, mosquito net use, and housing conditions. By adjusting for these factors, the study provides a refined assessment of the association between unimproved sanitation and malaria risk, reducing the likelihood of spurious correlations. Additionally, its comprehensive approach to categorizing environmental and household factors strengthens its findings. Instead of focusing solely on sanitation, the study examines flooring, wall, and roof materials, drinking water sources, mosquito net availability, and socioeconomic factors, providing a holistic perspective on malaria risk.

Despite its strengths, this study has several limitations. It relied on self-reported household characteristics, such as sanitation facilities and mosquito net usage, which could lead to recall bias or misreporting. Participants may not have accurately reported their sanitation conditions or net use, potentially affecting the observed associations. Second, the cross-sectional design of the survey used limits causal inference. While associations between socioeconomic factors and malaria prevalence were identified, the inability to establish temporality prevents definitive conclusions about whether poor housing conditions or socioeconomic status directly contributed to malaria risk. A longitudinal study design would provide stronger evidence of causal relationships. Third, the study did not account for seasonal variations in malaria transmission, which fluctuates due to rainfall, temperature, and mosquito breeding patterns. Since data collection occurred at a single point in time, the findings may not fully capture the seasonal burden of malaria. Future research should incorporate seasonal adjustments for a more comprehensive analysis. Additionally, the study focused exclusively on children under five years old, limiting the generalizability of its findings to older children and adults. The exclusion of participants who tested negative via rapid diagnostic testing but subsequently tested positive via microscopy is another limitation, as it may have led to misclassification of malaria cases. Finally, there is potential for residual confounding. Although key covariates such as wealth index, maternal education, and residence type were adjusted for, other unmeasured factors—such as access to healthcare, migration patterns, and genetic susceptibility to malaria—could contribute to the observed malaria risk differences. Future studies should incorporate a more comprehensive set of confounders to strengthen the validity of findings.

Conclusion

This study provides critical insights into the determinants of malaria infection among under-five children in Nigeria. The findings reveal that socioeconomic factors, such as wealth index, maternal education, and rural residence, play a significant role in malaria risk. Children from the poorest households, those whose mothers had no formal education, and those living in rural areas were at significantly higher risk of malaria infection. These associations highlight the broader social determinants of malaria transmission, emphasizing that poverty and limited access to education contribute to increased vulnerability.

While unimproved household sanitation facilities were hypothesized to be a key risk factor for malaria, the study found no significant association after adjusting for other confounders. This suggests that sanitation alone may not directly influence malaria prevalence but rather interacts with other environmental and socioeconomic conditions. Proper housing structures, effective drainage systems, and waste management may be more critical in reducing mosquito breeding and malaria transmission than sanitation facilities alone.

The study further underscores the importance of mosquito prevention strategies, particularly the use of insecticide-treated nets (ITNs). Although ITN ownership is widespread, ensuring consistent and proper use remains a challenge. Addressing these gaps through targeted interventions can significantly reduce malaria transmission.

Recommendations

To effectively reduce malaria prevalence in Nigeria, a multi-faceted approach is required, incorporating health interventions, socioeconomic development, and environmental management. First, efforts should focus on improving housing conditions in rural and low-income communities. Constructing homes with durable walls, floors, and roofs, along with implementing proper drainage systems, will help minimize mosquito breeding sites. Infrastructure development should be integrated into malaria prevention programs to create healthier living environments.

Second, education plays a vital role in malaria control. Expanding educational opportunities, especially for women, can enhance knowledge of malaria prevention, encourage proper ITN use, and improve healthcare-seeking behavior. Public health campaigns should emphasize not only ITN distribution but also effective use and maintenance to maximize their protective benefits. Additionally, routine net replacement programs should be strengthened to ensure households have functional ITNs.

Third, targeted malaria interventions should be expanded in rural areas where the burden is highest. Community-based initiatives, such as distributing ITNs, increasing access to malaria testing and treatment, and implementing environmental management programs, can effectively reduce malaria transmission. Health education campaigns should focus on eliminating stagnant water sources, promoting indoor residual spraying, and reinforcing the proper use of ITNs.

Lastly, policy measures should integrate malaria control efforts with broader poverty alleviation programs. Addressing economic disparities by improving access to healthcare, clean water, and improved housing can have long-term benefits in reducing malaria risk. Strengthening collaborations between government agencies, non-governmental organizations, and local communities will be essential in sustaining malaria prevention efforts and working toward malaria elimination in Nigeria.