Introduction

Schistosomiasis (SCH) and soil-transmitted helminthiasis (STH) are among neglected tropical diseases (NTDs), a group of diseases that affect the poorest and have received little attention in terms of control and research and development (R&D). Soil-transmitted helminths (STH) are a group of intestinal worm parasites (Ascaris lumbricoides, Ancylostoma duodenale, Necator americanus and Trichuris trichiura) that infect an estimated 1.45 billion people globally, resulting in the loss of an estimated 1.9 million disability adjusted life years (DALYs)1. Schistosomiasis is a waterborne disease, caused by Schistosoma parasites, which require fresh water snails (intermediate hosts) for their development. The disease affects over 250 million people globally, resulting in an estimated 1.6 million DALYs and over 200,000 deaths1. Of those affected, 95% are from sub-Saharan Africa (SSA) which bears the highest burden of disease2. The most prevalent species of schistosomes in Africa are Schistosoma haematobium and S. mansoni.

Chronic schistosomiasis and STH infections are associated with increased risk of malnutrition, iron deficiency anemia, cognitive delays, internal organ dysfunction and other morbidities, particularly in children. Control programs in several SSA countries have mostly relied on preventive chemotherapy or mass administration of anti-helminthic drugs (Praziquantel and Albendazole/Mebendazole for SCH and STH respectively) to school children and adults at highest risk with the aim to prevent morbidity3,4. Deworming STH and schistosomiasis has proven cost effective with a cumulative gain of about 26 million DALYs between 2000 and 20191. However, eliminating these infections as public health issues, as set out in the most recent WHO roadmap, has not yet been achieved in many nations5. A review on the prospect of STH elimination stresses that although deworming may achieve elimination (with high coverage), improvement in water, sanitation and hygiene (WaSH) is paramount6. Moreover, mathematical modelling indicates that WaSH interventions can sustain gains made by deworming, potentially leading to interruption of mass deworming on the long run7. For SCH, the 2022 WHO guideline includes a recommendation on WaSH, environmental interventions (including snail control) as well as behavioral change as essential measures to reduce the transmission of schistosomiasis8. Epidemiological studies in different countries indicate a significant impact of preventive chemotherapy but suggest that the interruption of transmission or elimination targets will only be achieved if deworming strategies are tailored to small administrative boundaries such as sub-Districts or sub-counties9,10 and/or coupled with WaSH interventions9,11 in a One Health approach12.

Rwanda aligns with the global goal to eliminate STH and SCH as public health issues. The national mass drug administration (MDA) campaigns are planned based on cross-sectional and school-based STH and SCH prevalence surveys that leave out other demographic groups, equally at risk. Thus, the country has persistent and newly identified endemic areas where the prevalence of these diseases remains high (2020 community-based mapping survey, in prep) despite more than a decade of preventive chemotherapy as evidenced by previous studies13,14,15. Water, Sanitation and Hygiene (WaSH) interventions have received little attention in terms of their impact on STH and schistosomiasis control and current sanitation and water development efforts are disconnected with disease control efforts16. Given the increasing evidence of a strong impact of access to drinking water and improved sanitation on the disease burden7,17,18, we sought to document the current WaSH status and its effect on STH and SCH prevalence in endemic areas of Rwanda. The ultimate goal is to highlight priority interventions that could be designed to supplement deworming efforts, taking into account the community contexts, for effective elimination of STH and SCH as public health problems in the country.

Results

Prevalence and infection intensity

A total of 2499 preschool aged children (PSAC), school aged children (SAC) and adults from 10 villages distributed in 7 districts were screened for schistosomiasis and STH. Overall, the prevalence of STH was 12.6% [95% Confidence interval (CI 11.3–14.0). The median [interquartile range (IQR)] infection intensity (among infected people) was highest for Ascaris lumbricoides, 132 (IQR 24–516) eggs per gramme (EPG) and lowest for Trichuris trichiura 24 (IQR 12–66) EPG.

The prevalence of SCH was 2.4% (95% CI 1.9–3.1) on KK, 4.9% (CI 4–6) on CCA when ‘trace’ results were considered negative and 7.2% (CI 6–8) when ‘trace’ results were considered positive.

There was no heavy infection intensity in all species but 0.2% of the population had moderate intensity infection of SCH and Ascaris.

Infection distribution according to participants’ age, gender and village

There were differences in prevalence (and intensity) of infection between age groups (Tables 1 and 2). School aged children (5–15 years) had significantly higher prevalence and intensity of infection with Ascaris and SCH (X2 = 49.7, p < 0.0001 and X2 = 11.4, p = 0.003 respectively) while participants aged ≥ 16 years had significantly higher prevalence of infection with Hookworm (X2 = 12.4, p = 0.002). T. trichiura infection was comparable between SAC and adults.

Table 1 Prevalence and intensity of STH infection per age and gender.
Full size table
Table 2 Prevalence and intensity of SCH infection per age and gender.
Full size table

Compared to females, male participants had significantly higher prevalence; 3.5%(n = 42) versus 1.5% (n = 19) of SCH (KK) infection (p = 0.001) while STH infections were comparable between males and females (p = 0.219; p = 0.941; p = 0.795 respectively for A. lumbricoides, T trichiura and Hookworm).

The prevalence and distribution of STH infection per village is illustrated on Fig. 1. Overall, there were heterogeneities in STH infection distribution, the highest infection prevalence being found in Gisagara, Gatsibo, Nyanza and Karongi Districts.

SCH distribution per village is illustrated on Fig. 2 where different diagnostic technics reveal differing levels of infection. S. mansoni cercaria shedding from snails occurred in villages with the highest SCH infection levels in humans (Fig. 3).

Fig. 1
Fig. 1
Full size image

Visualization of STH distribution per study sites. These are ArcGIS generated maps from geographical coordinates recorded at each collection site.

Fig. 2
Fig. 2
Full size image

Visualization of Schistosomiasis distribution. Human schistosomiasis prevalence based on the KK diagnostic, CCA trace considered negative and CCA trace considered positive.

Fig. 3
Fig. 3
Full size image

Visualization of snail distribution and cercariae shedding in study sites. Low possibility: presence of snails but no Schistosoma cercaria shedding; Significant possibility: Schistosoma cercaria shed from processed snails.

WASH risk factors

Logistic regression analyses (binary and multivariable) were carried out to assess associations between WaSH and sociodemographic variables and infection status (Table 3). Overall, there were higher odds of infection among people with limited access to clean water and a poor quality household toilet (a toilet with no slab, door, lid and/or roof was considered poor quality). Specifically, in bivariate logistic regression analysis, ready access to clean water and having a good quality house toilet were significantly associated with lower odds of infection, OR 0.72 (95% CI 0.55–0.95) and OR 0.55(95% CI0.40–0.76) respectively.

Table 3 Bi-variate and multivariable logistic regression analysis of sociodemographic and WASH risk factors of any parasite infection.
Full size table

The use of human excreta as agricultural fertilizer; OR:1.43 (95% CI 1.09–1.87) and host age > 5 years were significantly associated with twice the odds of infection with any parasite.

However, in multivariable logistic regression analysis, incorporating variables with p < 0.5 in bivariate analysis, only history of use of human excreta as fertiliser; adjusted OR (aOR) 6.4 (95% CI 1.97–20.8) and age > 15 years; aOR 5.52 (95% CI 1.13–27.1) remained significantly associated with infection.

Next, we analysed our data per helminth species (Supplementary Tables S1A–D).

STH: Hookworm infection was negatively associated with the availability of clean water cOR:0.43(95% CI 0.27–0.68) in bivariate analysis and aOR: 0.40(95% CI 0.22–0.70) in multivariable analysis. On the other hand, hookworm infection was positively associated with host age, with age ≥ 15years associated with higher odds of infection cOR: 9.21 (95% CI 2.91–29.2) in bivariate logistic regression analysis and aOR:14.7 (95% CI 3.6–60.9) in multivariable analysis (Suppl. Table S1A). A significant association was found between Trichuris trichiura infection and distance to a clean water source, OR: 1.94 (95% CI 1.02–3.7) with increased odds observed at greater distances in bivariate logistic regression analysis (Suppl. Table S1B). A. lumbricoides infection was associated with poor quality household toilet (no slab, lid, roof, door) while primary school educational attainment was relatively protective (Suppl. Table S1C).

SCH: The availability of clean water was protective for SCH, while the use of human faeces as agricultural fertilizer was a significant risk factor (Suppl. Table S1D).

Malacology

A total of 1031 Biomphalaria and 233 Bulinus snails were collected, out of which 795 and 91 Biomphalaria and Bulinus respectively were processed for shedding. Snail abundance per water body type is illustrated in Supplementary table S2 while Fig. 2 illustrates the percentage of snails that shed Schistosoma cercaria in different study villages. S. mansoni cercaria were found in villages with the highest prevalence of human infection.

Discussion

Schistosomiasis and STH infections are still of public health concern in Rwanda despite more than a decade of MDA interventions. The current study shows infection levels ranging from moderate to light, with Hookworm and Trichuriasis exclusively classified as light infections, according to WHO categorisation. We confirm that school aged children (SAC) exhibit the highest infection levels, with the exception of Hookworm, which primarily affects adults. The transmission route and the behaviour may partly explain this infection distribution: while SAC may be exposed to STH through poor feeding hygiene (mainly Ascaris sp and Trichiuris sp that are transmitted through the oral route), adults may be more prone to these infections as a result of walking or working on farms barefoot. This is consistent with the observed higher prevalence of Hookworm which is transmitted through skin penetration. Consistently, higher prevalence of Hookworm in adults walking barefoot were previously reported elsewhere19.

In this survey, despite the fact that more than 95% of the surveyed households had a domestic toilet, majority of the toilets were of inadequate quality (with no slab, lid, roof and/or door). Logistic regression analysis revealed a significant decrease (50%) in the odds of having helminth infections in households with an adequate toilet (with a slab, lid, roof and door). This suggests a strong impact of sanitation on geo-helminthiasis prevalence as previous studies reported7,18,20,21.

The lack of access to clean water was another important contributing factor to STH infection and schistosomiasis in the logistic regression model of household data. Furthermore, villages with the highest infection prevalence lack or have inadequate water supply systems as per the National Institute of Statistics Rwanda (NISR)’ data on national water coverage (Statistical Year Book 2023 | National Institute of Statistics Rwanda) and information obtained from WASAC Utility Ltd (unpublished data). STH transmission is indeed closely linked to poor hygiene and the lack of potable water is a main driver18,22. Moreover, populations in areas without access to clean water tend to use open water sources such as lakes, rivers and dams for their domestic and personal hygiene, which may expose them to the skin penetrating Schistosoma sp23. This is consistent with our observation that S. mansoni cercariae were shedding from freshwater snails in study areas with the highest schistosomiasis prevalence. Indeed, cercaria shedding has been suggested as a good proxy for transmission estimation since it coincides with infection levels in humans24,25.

In the current study, around 12% of participants used human excreta as agricultural fertiliser. Logistic regression analysis revealed that this practice increases the odds of being infected, and this was more robust with SCH. The result suggests that these fertilisers may contain viable eggs of parasites when they are applied to agricultural land. Irrigation and wetland exploitation would allow for Schistosoma eggs maturation into miracidia, consistent with the fact that 95% of Schistosoma cercaria shedding occured in wetlands dedicated to rice crop farming in the current study, and that schistosomiasis hotspots in Rwanda are often linked with rice paddies24.

Previous reports highlighted the risk of helminth transmission through the use of human excreta in farming26,27 and our results align with these findings. However, adequate treatment of human excreta before use in agriculture is warranted given its economic potential26. Specifically, this evidence calls for the development and enforcement of national policies in Rwanda that regulate the safe use of human waste in farming. Policies should promote safe sanitation practices, proper waste treatment technologies, and community education to mitigate the health risks while allowing for the economic benefits of using human excreta as fertilizer.

The level of education was another factor associated with infection since people with a basic level of education were protected against Hookworm and Ascaris. This suggests that disease awareness may increase with the level of education and the findings are in line with previous reports21,28. It is worth noting that socio-economic development goes hand in hand with improved education and it may be difficult to determine the relative contribution of each on general health.

Male participants were more at risk of SCH infection compared to female participants in the current study, consistent with previous reports29,30. Notwithstanding the potential contribution of hormonal differences in the development of protective immunity, the results indicate a high level of gender inequality in disease exposure that may be worth exploring. Indeed, male behavior (SAC swimming or playing in water) and occupation (adults fishing or farming in wetlands) may pose a higher risk of SCH on males than females and control efforts should be context tailored to leave no one behind.

In conclusion, STH and SCH distribution largely vary between different geographical locations in Rwanda. The prevalence is closely linked to various WaSH factors, particularly poor household sanitation, the lack of clean water for domestic use, the use of human excreta in agriculture and the presence (and infectivity) of snail intermediate hosts for Schistosoma parasites. For SCH and STH to be eliminated as public health issues in Rwanda, control interventions should incorporate health education and inter-sectorial (health-agriculture-environment-infrastructure) collaboration strategies to improve WaSH infrastructure and human behavior. Importantly, strategies for (1) promoting the safe use of human excreta in agriculture and (2) controlling snail abundance in SCH hotspots should be explored in the local context.

Materials and methods

Study design

This was a ‘cross-sectional’ study design where participants were recruited in November 2023 from communities where SCH and STH are endemic using the 2020 community-based mapping survey results. The study assessed the prevalence and intensity of SCH and STH infections as well as WaSH status in the selected sentinel sites (SS). This was coupled with a malacological survey to assess the abundance of snail intermediate hosts and their level of infectivity and associated risk of transmission.

Sample size Estimation and sampling technique

A stratified sampling method was used for the selection of sentinel sites, first by ecological zones from which Districts were selected. Within Districts, villages were grouped based on their previous STH and SCH infection levels (as per the 2020 mapping survey). Next, a random selection of villages in each category was performed with adjustments where a village was not endemic for both SCH and STH.

Within villages, households were listed and assigned numbers to allow for a random (computed) selection. In selected households, all members were eligible for sample collection and testing.

Study area and population

Spatially explicit sampling targeted communities in endemic areas (10 administrative villages selected). The sampling was stratified, aiming to include villages with different co-endemicity (STH and SCH) levels (based on the 2020 mapping results), SCH being the primary indicator to guide sampling. The number of PSAC (12–59 months old), SAC (5–15 years old) and adults (16 years and above) sampled was proportional to the villages’ population structure (60% adults, 27% SAC and 13% PSAC) as per the latest population census’ report (Rwanda 5th Population and Housing Census, NISR 2022 accessible at https://www.statistics.gov.rw/datasource/fifth-population-and-housing-census-2022). The target population is from rural settings where the major socio-economic activity is subsistence farming. The study spanned on a wide geographical area, covering heterogeneous environmental conditions within a tropical temperate climate that is typical throughout the country.

Parasitological assessment

The parasitological assessment was carried out using the Kato Katz (KK) technique for STH and SCH (N = 2495) and the Circulating Cathodic Antigen (CCA) rapid test for SCH (N = 2442). For KK (Lot #XDXFBC, FG-L11, Sterlitech Corporation, WA), duplicate slides of fecal material were prepared for each participant and examined (by blinded technicians) using light microscopy31. Egg counting was done for each individual helminth species to capture infection intensity. As Hookworm eggs can disintegrate and become undetectable after the standard waiting time (40–60 min) for glycerin to react, the reading of slides was done within 20 min but slides were retained for a second examination to screen other parasites’ eggs after 60 min. Quality control was performed by the National Reference Laboratory (ISO 15189 accredited). For CCA, urine samples were run on a lateral flow device (Batch no 230515042, IDT Diagnostics CC t/a ICT International, South Africa) and readings were recorded after 20 min as per the manufacturer’s instructions. The thickness of the positive band was used to determine the 4 categories of intensity: +++, ++, + or trace as previously validated32.

Malacological survey

Snail sampling points were selected using existing spatial data within main rivers, wetlands, lakes and/or multipurpose water dams. Snails were collected using the scooping technique, and transported in the nearest laboratory for shedding using standard protocols33,34. Briefly, snails are alternatively exposed to light and darkness to stimulate shedding and cercaria are observed and counted under a microscope.

Geographical coordinates were recorded for all water bodies and sample collection sites were visualized on maps generated using ArcGIS 10.4 software.

WASH and demographic assessment

Water, sanitation and hygiene status in different households as well as demographic data were assessed through a questionnaire (Supplementary material). The questionnaire was adapted from the WHO/UNICEF Joint Monitoring Programme for Water, Sanitation and Hygiene households core questions35 and was validated by the National NTD working group. Since the assessment was household based, the head of family or their representative responded to the questions on behalf of all family members. Data were collected electronically using a questionnaire designed in RedCap and was administered by trained data collectors through face- to- face interviews coupled with direct observations where applicable as a means of verification for the self-reported answers.

Data analysis

Statistical analyses were done using SPSS version 21 and STATA version 16 (StataCorp, College Station, TX) statistical softwares. Descriptive statistical and spatial analyses were performed to determine the prevalence and infection intensity and their distribution by village, age and gender, as well as snail distribution and infectivity. Logistic regression analyses were performed to assess the relationship between infection and different WASH risk factors. The variables in question were categorized into two types:

  1. 1.

    Continuous variables: Duration to water source (in minutes) and time spent farming (in hours) are presented as continuous variables.

  2. 2.

    Dichotomous variables: Availability of clean water, quality house toilet, history of use of faeces as fertilizer, and current use of faeces as fertilizer are dichotomous variables, with only two possible outcomes (e.g., yes/no, present/absent).

Reference Categories:

For the dichotomous variables, the reference categories used in logistic regression analysis are consistently defined as the negative outcome (e.g., clean water not available, poor quality house toilet, use of human faeces as fertilizer).

All variables were analyzed using bivariate logistic regression and potentially significant variables (cutoff p < 0.5) were included in the multivariable logistic regression.