Skip to main content
NCBI home page
Elsevier Sponsored Documents logoLink to Elsevier Sponsored Documents
. 2022 Jan;22(1):136–149. doi: 10.1016/S1473-3099(21)00090-6

Effect of preventive chemotherapy with praziquantel on schistosomiasis among school-aged children in sub-Saharan Africa: a spatiotemporal modelling study

Christos Kokaliaris a,b, Amadou Garba c, Martin Matuska a,b, Rachel N Bronzan d, Daniel G Colley e, Ameyo M Dorkenoo f, Uwem F Ekpo g, Fiona M Fleming h, Michael D French i, Achille Kabore j, Jean B Mbonigaba k, Nicholas Midzi l, Pauline N M Mwinzi m,n, Eliézer K N'Goran o,p, Maria Rebollo Polo n, Moussa Sacko q, Louis-Albert Tchuem Tchuenté r,s, Edridah M Tukahebwa t, Pitchouna A Uvon u, Guojing Yang a,b, Lisa Wiesner a,b, Yaobi Zhang v, Jürg Utzinger a,b, Penelope Vounatsou a,b,*
PMCID: PMC8695385  PMID: 34863336

Summary

Background

Over the past 20 years, schistosomiasis control has been scaled up. Preventive chemotherapy with praziquantel is the main intervention. We aimed to assess the effect of preventive chemotherapy on schistosomiasis prevalence in sub-Saharan Africa, comparing 2000–10 with 2011–14 and 2015–19.

Methods

In this spatiotemporal modelling study, we analysed survey data from school-aged children (aged 5–14 years) in 44 countries across sub-Saharan Africa. The data were extracted from the Global Neglected Tropical Diseases database and augmented by 2018 and 2019 survey data obtained from disease control programmes. Bayesian geostatistical models were fitted to Schistosoma haematobium and Schistosoma mansoni survey data. The models included data on climatic predictors obtained from satellites and other open-source environmental databases and socioeconomic predictors obtained from various household surveys. Temporal changes in Schistosoma species prevalence were estimated by a categorical variable with values corresponding to the three time periods (2000–10, 2011–14, and 2015–19) during which preventive chemotherapy interventions were scaled up.

Findings

We identified 781 references with relevant geolocated schistosomiasis survey data for 2000–19. There were 19 166 unique survey locations for S haematobium and 23 861 for S mansoni, of which 77% (14 757 locations for S haematobium and 18 372 locations for S mansoni) corresponded to 2011–19. Schistosomiasis prevalence among school-aged children in sub-Saharan Africa decreased from 23·0% (95% Bayesian credible interval 22·1–24·1) in 2000–10 to 9·6% (9·1–10·2) in 2015–19, an overall reduction of 58·3%. The reduction of S haematobium was 67·9% (64·6–71·1) and that of S mansoni 53·6% (45·2–58·3) when comparing 2000–10 with 2015–19.

Interpretation

Our model-based estimates suggest that schistosomiasis prevalence in sub-Saharan Africa has decreased considerably, most likely explained by the scale-up of preventive chemotherapy. There is a need to consolidate gains in the control of schistosomiasis by means of preventive chemotherapy, coupled with other interventions to interrupt disease transmission.

Funding

European Research Council and WHO.

Introduction

Schistosomiasis is a water-based disease caused by parasitic trematode worms of the genus Schistosoma.1 In 2017, the global burden of schistosomiasis was estimated at 1·4 million disability-adjusted life-years.2 WHO reported that 229 million people were affected by schistosomiasis in 2015, with more than 90% of them living in sub-Saharan Africa, 54% of whom were school-aged (aged 5–14 years) children.3 The two main Schistosoma species affecting people in Africa are Schistosoma haematobium (causing urogenital schistosomiasis) and Schistosoma mansoni (causing intestinal schistosomiasis). In 2001, the World Health Assembly endorsed a resolution that emphasised morbidity control through preventive chemotherapy with praziquantel as the global strategy and set a target of regularly treating at least 75% of school-aged children by the year 2010. Revitalising schistosomiasis control efforts in 2012, WHO put forward the 2020 targets (a series of action points, main targets, and milestones for accelerating work to overcome the global effect of neglected tropical diseases, including schistosomiasis, the ultimate goal being disease eradication) and established a roadmap to get there.4 Endemic countries were urged to scale-up schistosomiasis control interventions and strengthen surveillance, improve the environment to decrease disease transmission, and ensure access to praziquantel.4

Before the early 2000s, preventive chemotherapy using praziquantel was not widespread in sub-Saharan Africa, which changed in 2005 when 250 million praziquantel doses were pledged to be provided every year by Merck. From 2002 onwards, preventive chemotherapy efforts for schistosomiasis were scaled up in the form of mass drug administration programmes. A WHO report revealed that approximately 76·2 million school-aged children and 19·1 million adults were treated with praziquantel in 2018, corresponding to 61·2% treatment coverage for children and 18·2% treatment coverage for adults.3 By contrast, in 2006, only 7 million individuals were treated. In 2017, 17 African countries had achieved the 75% treatment coverage target for school-aged children.3 However, difficulties accessing praziquantel for at-risk adult populations and preschool-aged children remain a key issue in further scaling up control efforts. Over the past 20 years, praziquantel has been targeted at school-aged children and made available through WHO to ministries of health free of charge.5

Research in context.

Evidence before this study

Schistosomiasis is a neglected tropical disease that is mainly concentrated in sub-Saharan Africa. In 2012, WHO published a roadmap for the control and elimination of the disease and set targets to be reached by 2020. Over the past 20 years, control efforts against schistosomiasis have been scaled up, with preventive chemotherapy using praziquantel being the main intervention. There is a need to evaluate the effect of these increased control efforts on the distribution of schistosomiasis and to estimate progress made towards the WHO targets. In 2015, Lai and colleagues published maps of schistosomiasis prevalence across sub-Saharan Africa during 2000–12 and estimated that 122 million doses of praziquantel were required for 228 million school-aged children in 2012 using survey data from 1980 to 2012.

Added value of this study

We provide updated maps of schistosomiasis prevalence in sub-Saharan Africa at high resolution covering 2000–19, by including the latest available epidemiological data that correspond with periods in which preventive chemotherapy was scaled up. In our analysis, we assess the effect of preventive chemotherapy using praziquantel on schistosomiasis prevalence in school-aged children, overall and by species. We provide estimates of schistosomiasis prevalence reduction by country during the scaling up of praziquantel by comparing data from 2015–19 with 2000–10. We also update the estimates of schistosomiasis prevalence and treatment needs for each sub-Saharan African country using population data from 2020, to compare them with the 2020 WHO goals. Our modelling accounts for spatial confounding in the covariates, which enables an accurate estimation of the effect of each risk factor.

Implications of all the available evidence

Our research should assist policy makers to plan their future schistosomiasis control strategies according to the prevalence trends from the past 5–10 years; provide a measure of programme assessment and evaluation by using prevalence estimates before, during, and after preventive chemotherapy; and provide evidence on the effect of socioeconomic and environmental factors (eg, improved sanitation and distance from freshwater bodies) to be considered for future supplementary control projects.

Complementary control interventions include access to clean water and improved sanitation,6, 7 snail control, and behaviour change.8, 9 These approaches have not been applied widely in sub-Saharan Africa because there are insufficient human and financial resources and challenges in identifying water bodies containing infected intermediate-host snails.5

In this study, we estimate the effect of preventive chemotherapy on schistosomiasis prevalence across sub-Saharan Africa. We aimed to assess the changes in the geographical distribution of schistosomiasis, comparing 2000–10 (the early stages of scaling up the control programmes) with 2011–14 (intermediate period) and 2015–19 (after substantial scale-up efforts), and provide updated estimates of treatment needs per country.

Methods

Prevalence data

Cross-sectional survey data pertaining to S haematobium and S mansoni infection prevalence in the 2000–19 period were extracted from the Global Neglected Tropical Diseases database.10 Notably, Global Neglected Tropical Diseases compiles survey data through systematic reviews of published research, coupled with grey literature from country programmes and data from the Expanded Special Project for Elimination of Neglected Tropical Diseases (ESPEN).121

To supplement data available in the Global Neglected Tropical Diseases database, we did a systematic review following PRISMA guidelines, with strict inclusion and extraction criteria to ensure that high-quality data were included in the analysis.12 We searched PubMed, ISI Web of Science, and African Journals Online from Jan 1, 2000, to May 29, 2020, without language restrictions, for surveys that reported schistosomiasis prevalence data for countries in sub-Saharan Africa. The search string included the following criteria: “schisto* (OR mansoni, OR bilhar*, OR haema*) AND sub-Saharan Africa (OR Angola, OR Benin, OR Botswana, OR Burkina Faso, OR Burundi, OR Cameroon, OR Central African Republic, OR Chad, OR Congo*, OR Côte d'Ivoire, OR Democratic Republic of the Congo, OR Djibouti, OR Equatorial Guinea, OR Eritrea, OR Eswatini, OR Ethiopia, OR Gabon, OR Gambia, OR Ghana, OR Guinea, OR Guinea-Bissau, OR Kenya, OR Lesotho, OR Liberia, OR Madagascar, OR Malawi, OR Mali, OR Mauritania, OR Mozambique, OR Namibia, OR Niger, OR Nigeria, OR Rwanda, OR Senegal, OR Sierra Leone, OR Somalia, OR South Africa, OR South Sudan, OR Sudan, OR Tanzania, OR Togo, OR Uganda, OR Zambia, OR Zimbabwe).

We excluded case reports, in-vitro studies, non-human studies, or those that did not report on schistosomiasis. We additionally excluded studies without prevalence data, those done in specific groups of patients (eg, patients in hospital, people living with HIV) or clearly defined population groups (ie, travellers, military personnel, expatriates, nomads, and displaced or migrating populations, pregnant women, neonates) not representative of the general population, studies that used either indirect diagnostic techniques (because such tests distinguish between active and cleared infection) or direct stool smear (because of low diagnostic sensitivity), reports of case-control studies, clinical trials, pharmacological studies (except control groups without anthelmintic intervention), intervention studies (except for baseline data or control groups), studies that reported on species other than S haematobium and S mansoni, and surveys done before 2000, that were not community based or school based, or were done in places where population deworming had been done within 1 year, or study findings reported aggregated within regions (ie, administrative division of level one).

The search strategy and selection criteria are described in detail in the appendix (pp 1–2) and in a previous publication.13 Quality control was applied for each country on approximately 30% of the data, which were selected at random and embedded in the GNTD database as a function written in Javascript Survey. Locations with missing coordinates were geolocated using Google Maps and georeferenced school databases such as the Humanitarian Data Exchange. Relevant survey data were extracted and entered in the Global Neglected Tropical Diseases database.

Data on covariates

Socioeconomic data on improvements in drinking water, sanitation, and infant mortality rates during 2000–19 were obtained from Demographic and Health Surveys, Multiple Indicator Cluster Surveys, and World Health Surveys. Socioeconomic survey data were not available at yearly basis. Therefore, socioeconomic data from these surveys were aggregated in the 2000–10, 2011–14, and 2015–19 time periods as per schistosomiasis prevalence data and were aligned to the appropriate countries.

Environmental proxies (including land surface temperature [LST] during the day [LSTD] and at night [LSTN], normalised difference vegetation index [NDVI], rainfall, bioclimatic variables, agro-ecological zones and distance to freshwater bodies) were downloaded from remote-sensing satellite sources and model-based gridded surfaces (appendix p 3). Population data for 2010–19 were extracted from the WorldPop database. The national preventive chemotherapy coverage for schistosomiasis in sub-Saharan Africa for each country and year was obtained from a publicly available WHO database.14 A description of the raw data and their sources is provided in the appendix (p 3).

Statistical modelling

We used Bayesian restricted, geostatistical, hierarchical models to predict the prevalence of S haematobium and S mansoni across sub-Saharan Africa (appendix p 5), relating cross-sectional disease survey data with socioeconomic and environmental predictors, accounting for spatial confounding in the covariates.15 The models included a temporal indicator variable for the periods of 2000–10 (baseline), 2011–14, and 2015–19. The socioeconomic predictors were linked with the prevalence data as areal exposures, aggregated at the first-level administrative unit because of misalignment. The environmental predictors LST, NDVI, and rainfall were summarised by yearly averages and linked to the survey data according to the corresponding year. The bioclimatic predictors were obtained as long-term averages between 1970 and 2000 and linked to the survey data according to the geographical location. Potential socioeconomic and environmental predictors resulted in a large number of models. All models were fitted to the environmental and socioeconomic predictors and the models with the best predictive ability (minimum log conditional predictive ordinate score)16, 17 for S haematobium and S mansoni prevalence were used for inference and predictions. Disease heterogeneity can vary between ecological zones.18 Hence, we validated the assumption of non-stationarity (ie, implying that spatial correlation varies in space) by introducing the agroecological zone covariates (ie, humid, arid, highlands, semiarid, and subhumid) in the covariance structure of the spatial Gaussian process. Furthermore, we examined the hypothesis of disease prevalence distribution depending on the geographical and temporal variation of the data, by implementing spatiotemporal geostatistical models.19, 20 We obtained two separate models, one for each species, differing by the associated covariates and spatial heterogeneity. Parameter estimates were summarised using posterior medians and the corresponding 95% Bayesian credible intervals (BCI) obtained from the 0·025 and 0·975 quantiles of the posterior distribution of the parameter. The effect of a predictor was considered to be statistically important if the 95% BCI did not include a 0. Model validation was done through out-of-sample predictions, leaving out 10% of the data repeated 20 times.

To estimate the effect of preventive chemotherapy on schistosomiasis prevalence for a given country and year, we calculated the mean national coverage of preventive chemotherapy over the past 3 years and used it as a covariate in the models. The number of school-aged children infected with Schistosoma species was estimated by overlaying predicted Schistosoma species prevalence surfaces with gridded surfaces of population counts at high spatial resolution (100 m × 100 m). Preventive chemotherapy needs for each country were calculated in accordance with WHO guidelines (appendix p 8).21 We used R (version 3.3.3) for the statistical analysis. Modelling details are presented in the appendix (pp 5–6, 8).

Role of the funding source

The funders of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the paper.

Results

We identified 781 relevant references from the systematic review and national control programmes, including data provided by WHO after consultation with ministries of health and the ESPEN portal. A flowchart of the total number of articles searched and those deemed relevant after applying our inclusion and exclusion criteria is presented in the appendix (p 2). There were 19 166 unique survey locations for S haematobium and 23 861 unique survey locations for S mansoni, of which 77% (14 757 locations for S haematobium and 18 372 locations for S mansoni) corresponded to 2011–19. From the 26 780 unique locations, 15 344 (57·3%) were obtained from peer-reviewed studies, 7739 (28·9%) from control programmes, and 3695 (13·8%) from ESPEN records. The main diagnostic method for S haematobium was urine filtration (done in 340 [77%] of 444 surveys). The Kato-Katz method was the predominant approach for S mansoni (421 [96%] of 437 surveys). The raw prevalence data by Schistosoma species and survey period are presented in figure 1. Most of the surveys were school-based: 174 (97%) of 180 schistosomiasis surveys were school-based during 2000–10, and 594 (99%) of 601 were school-based in 2011–19. A detailed description of the data, stratified by country and survey period, is provided in the appendix (p 4).

Figure 1.

Figure 1

Open in a new tab

Observed prevalence of Schistosoma species in sub-Saharan Africa

(A–C) Schistosoma haematobium. (D–F) Schistosoma mansoni.

Risk factor analysis showed that high praziquantel coverage during a 3-year period and higher proportion of households with access to improved sanitation facilities were related to lower risk of infection for S haematobium and S mansoni (table 1). Close proximity to freshwater bodies was associated with higher risk of Schistosoma species infection. S haematobium risk was associated positively with LSTN, NDVI, and the mean diurnal range. Humid agroecological zones, altitude, and the mean temperature of the driest quarter showed negative associations with S haematobium risk. S mansoni infection risk was associated positively with isothermality, precipitation, and humid agroecological zones (table 1).

Table 1.

Estimates for covariates obtained from Bayesian geostatistical stationary and non-stationary models for Schistosoma haematobium and Schistosoma mansoni in sub-Saharan Africa

S haematobium S mansoni
Stationary geostatistical model
Risk factors
Year period
2000–10 1·00 (ref) 1·00 (ref)
2011–14 −0·98 (−1·05 to −0·91) −0·09 (−0·18 to 0·01)
2015–19 −1·39 (−1·48 to −1·29) −1·00 (−1·12 to −0·88)
Mean preventive chemotherapy coverage in the past 3 years (%) −0·13 (−0·17 to −0·09) −0·16 (−0·22 to −0·11)
Distance from freshwater bodies
≥500 m 1·00 (ref) 1·00 (ref)
<500 m 0·42 (0·31 to 0·54) 1·15 (1·03 to 1·27)
Isothermality* NA 0·30 (0·22 to 0·38)
Precipitation (mm) NA 0·24 (0·18 to 0·30)
Improved sanitation (%)† −0·21 (−0·25 to −0·18) −0·22 (−0·27 to −0·16)
Altitude −0·64 (−0·72 to −0·57) NA
Land surface temperature at night (°C) 0·17 (0·10 to 0·23) NA
Normalised difference vegetation index 0·28 (0·24 to 0·33) NA
Mean diurnal range (°C) 0·12 (0·07 to 0·17) NA
Mean temperature of driest quarter (°C) −0·32 (−0·39 to −0·25) NA
Precipitation of wettest month −0·18 (−0·22 to −0·13) NA
Agroecological zone
Humid 1·00 (ref) 1·00 (ref)
Arid 0·48 (0·24 to 0·72) −1·54 (−2·05 to −1·02)
Highlands 1·65 (1·48 to 1·82) −0·50 (−0·63 to −0·37)
Semiarid 2·16 (2·02 to 2·29) −0·77 (−0·99 to −0·54)
Subhumid 1·30 (1·19 to 1·40) −0·31 (−0·45 to −0·17)
Geographical variation parameters
Range (km) 133·5 (122·1 to 146·2) 222·1 (200·1 to 252·3)
Non-spatial variance (σe2) 1·44 (1·41 to 1·46) 1·42 (1·39 to 1·45)
Spatial variance (σ2) 2·20 (2·09 to 2·31) 4·40 (4·07 to 4·84)
Non-stationary geostatistical model‡
Spatial variance (σ2)§
Agroecological zone
Humid 1·00 (ref) 1·00 (ref)
Arid 2·50 (2·20 to 2·87) 2·37 (1·96 to 2·89)
Highlands 2·67 (2·43 to 2·93) 3·20 (2·99 to 3·43)
Semiarid 2·15 (1·98 to 2·32) 3·02 (2·78 to 3·28)
Subhumid 1·38 (1·35 to 1·41) 1·42 (1·38 to 1·46)
Open in a new tab

Data are posterior medians (95% Bayesian credible intervals) unless otherwise stated. NA=not applicable.

*

Isothermality is calculated by the ratio of the mean diurnal range (difference between minimum and maximum daily temperature) to the annual temperature range.

†

Improved sanitation refers to connection to a public sewer or to a septic system, pour-flush latrine, simple pit latrine, or ventilated improved pit latrine.

‡

Risk factors, range, and non-stationary variance are similar to the non-stationary geostatistical model and were omitted.

§

Spatial variance was modelled as a function of the agroecological zone categories.

Our geostatistical analysis showed that the infection prevalence decreased over time compared with 2000–10 (table 1). Population-adjusted estimates of the prevalence reduction (ie, relative to prevalence during 2000–10) and the prevalence ratio (table 2) between different time periods confirmed that there was a statistically important (determined by 95% BCI) drop in prevalence for both Schistosoma species across sub-Saharan Africa (table 2). For the 44 countries in sub-Saharan Africa included in our analysis, we estimated a 67·9% overall relative prevalence reduction for S haematobium and 53·6% reduction for S mansoni in 2015–19 compared with 2000–10 (table 2).

Table 2.

Estimated relative prevalence reduction (percentage decrease of schistosomiasis prevalence) by species and country, comparing 2011–14 and 2015–19 with 2000–10 across sub-Saharan Africa

Relative prevalence reduction (2011–14)
 Relative prevalence reduction (2015–19)
Schistosomiasis S haematobium S mansoni Schistosomiasis S haematobium S mansoni
Angola 39·2% 43·2% 21·9% 58·2% 61·7% 53·6%
Benin 42·8% 44·9% 29·8% 65·4% 65·6% 69·7%
Botswana 41·5% 47·8% 16·5% 59·1% 63·7% 44·6%
Burkina Faso 67·2% 69·5% 50·1% 77·9% 78·5% 70·7%
Burundi 43·7% 67·4% 34·0% 69·7% 80·1% 64·0%
Cameroon 41·1% 53·3% 19·9% 68·1% 73·1% 60·8%
Central African Republic 24·6% 43·6% 14·0% 36·4% 56·7% 23·7%
Chad 34·2% 37·6% 14·8% 50·9% 52·6% 47·1%
Congo 27·3% 46·5% 6·9% 57·4% 67·8% 48·8%
Côte d'Ivoire 36·4% 54·6% 17·8% 54·9% 67·0% 44·4%
Democratic Republic of the Congo 50·1% 53·0% 24·2% 71·6% 73·3% 59·2%
Djibouti 54·0% 60·5% 37·2% 66·0% 68·5% 45·6%
Equatorial Guinea 44·6% 67·2% 21·8% 60·7% 74·5% 48·3%
Eritrea 19·9% 54·8% 2·7% 68·1% 78·4% 61·1%
Eswatini 44·1% 63·5% 24·5% 66·5% 78·3% 54·7%
Ethiopia 32·7% 53·0% 9·9% 59·7% 69·4% 49·3%
Gabon 34·1% 54·2% 8·7% 55·3% 69·0% 35·4%
Gambia 51·0% 52·5% 9·5% 71·6% 72·0% 65·4%
Ghana 46·7% 54·0% 21·9% 64·6% 68·0% 57·5%
Guinea 15·2% 43·8% 7·0% 29·0% 58·6% 20·4%
Guinea-Bissau 45·0% 55·3% 33·0% 64·3% 71·6% 52·9%
Kenya 39·2% 51·0% 18·2% 61·7% 67·4% 55·3%
Lesotho 49·2% 61·0% 8·5% 67·9% 69·7% 61·1%
Liberia 30·1% 51·6% 19·3% 43·1% 58·2% 38·8%
Madagascar* .. .. .. .. .. ..
Malawi 45·7% 51·2% 28·7% 69·4% 70·9% 69·8%
Mali 42·9% 46·6% 35·4% 56·3% 59·3% 55·6%
Mauritania 41·6% 44·4% 25·0% 56·9% 58·7% 48·9%
Mozambique 31·8% 40·4% 17·3% 47·5% 53·0% 43·6%
Namibia 38·8% 49·4% 8·8% 59·4% 64·4% 43·7%
Niger 57·5% 58·7% 42·7% 70·5% 71·3% 63·9%
Nigeria 44·9% 48·6% 21·1% 67·1% 68·7% 59·9%
Rwanda 28·2% 60·3% 20·5% 61·9% 78·6% 57·8%
Senegal 47·5% 52·8% 30·9% 64·5% 67·7% 55·1%
Sierra Leone 44·2% 69·2% 37·8% 51·3% 79·4% 42·5%
Somalia 39·6% 45·3% 14·1% 55·9% 59·8% 49·1%
South Africa 37·3% 46·8% 6·8% 55·5% 60·3% 40·7%
South Sudan 22·3% 48·8% −1·6% 46·3% 64·3% 32·4%
Sudan 27·9% 45·1% 10·3% 48·7% 62·2% 38·2%
Tanzania 31·9% 46·2% 6·7% 51·6% 59·8% 37·8%
Togo 68·4% 69·7% 65·0% 73·6% 74·7% 69·9%
Uganda 24·9% 53·3% 8·9% 62·0% 71·5% 55·9%
Zambia 31·8% 45·9% 15·5% 56·3% 63·8% 53·9%
Zimbabwe 47·0% 57·9% 22·8% 72·0% 75·8% 65·1%
Total (95% Bayesian credible interval) 39·4% (34·3–43·6) 51·2% (46·9–56·4) 19·5% (9·2–27·3) 60·5% (55·7–64·6) 67·9% (64·6–71·1) 53·6% (45·2–58·3)
Open in a new tab

Relative prevalence reduction was calculated as the prevalence in 2000–10 minus the prevalence in 2011–14 or 2015–19 divided by the prevalence in 2000–10. For example, the 39·2% relative prevalence reduction of schistosomiasis in Angola during the 2011–14 period suggests that in Angola the schistosomiasis prevalence during 2011–14 decreased by 39·2% compared with the baseline prevalence during 2000–10.

*

Madagascar only had data available for 2015–19, therefore reduction prevalences could not be calculated.

The predictive prevalence maps (figure 2) show a decline in prevalence for S haematobium and S mansoni from 2000–10 to 2011–14 and from 2011–14 to 2015–19. Prediction uncertainty was high in areas with sparse data for both periods (2000–10 to 2011–14 and 2011–14 to 2015–19; appendix p 4). The population-adjusted prevalence of schistosomiasis in sub-Saharan Africa was estimated at 23·0% (95% BCI 22·1–24·1) during 2000–10, which declined to 9·6% (9·1–10·2) by 2015–19 (table 3). In sub-Saharan Africa, the population-adjusted prevalence in 2010 was estimated at 17·4% (16·5–18·5) for S haematobium and 7·1% (6·5–7·6) for S mansoni (table 3). Estimates based on population data from 2019 suggest a prevalence of 6·2% (5·7–6·7) for S haematobium and 3·7% (3·4–4·0) for S mansoni (table 3). The countries with the highest prevalence of infection (ie, >40%) in 2000–10 were the Central African Republic, Chad, Guinea, Liberia, and Mozambique. During the same period, Burundi, Equatorial Guinea, Eswatini, Lesotho, and Rwanda had the lowest prevalence (<5%). During 2015–19, we observe that all countries had lower prevalence than the 2000–10 period; however, Guinea remained the country with the highest prevalence (32·6%%).

Figure 2.

Figure 2

Open in a new tab

Schistosomiasis prevalence estimates across sub-Saharan Africa

Data are posterior predictive median. (A–C) Schistosoma haematobium. (D–F) Schistosoma mansoni.

Table 3.

Total number of school-aged children and population-adjusted prevalence for each infectious species in sub-Saharan Africa by year

2010
 2014
 2019
Population (in 1000s) Schistosomiasis (%) S haematobium (%) S mansoni(%) Population (in 1000s) Schistosomiasis (%) S haematobium (%) S mansoni(%) Population (in 1000s) Schistosomiasis (%) S haematobium (%) S mansoni(%)
Angola 5321 32·2% (21·4–41·3) 26·6% (16·4–36·4) 6·0% (4·0–16·9) 6488 19·6% (12·8–31·9) 14·9% (8·7–26·8) 4·8% (3·0–10·5) 7721 12·9% (7·5–23·1) 9·7% (5·2–18·6) 2·8% (1·8–8·8)
Benin 2366 36·7% (33·1–41·8) 34·3% (30·8–39·2) 3·4% (2·6–5·1) 2689 21·0% (18·6–24·7) 19·1% (16·5–22·8) 2·4% (1·8–3·3) 3037 12·6% (10·9–15·7) 11·7% (10·0–14·7) 1·0% (0·7–1·5)
Botswana 407 12·5% (7·7–19·8) 10·6% (5·9–17·4) 2·1% (1·0–4·3) 435 7·2% (4·6–12·8) 5·5% (2·8–10·4) 1·8% (0·9–3·5) 474 5·0% (2·6–9·8) 3·7% (1·7–8·4) 1·2% (0·6–2·9)
Burkina Faso 4444 17·5% (14·2–21·3) 15·7% (12·6–19·7) 1·9% (1·2–3·2) 5178 5·7% (4·5–7·7) 4·8% (3·7–6·6) 1·0% (0·6–1·8) 5900 4·0% (3·0–5·5) 3·4% (2·6–4·7) 0·5% (0·3–1·2)
Burundi 2256 4·8% (3·3–7·8) 1·7% (0·6–4·8) 3·1% (2·3–4·4) 2669 2·7% (1·9–4·3) 0·5% (0·2–1·8) 2·0% (1·5–3·0) 3211 1·5% (1·0–2·2) 0·3% (0·1–1·2) 1·1% (0·8–1·6)
Cameroon 5191 15·5% (13·6–17·8) 10·6% (9·3–12·6) 5·4% (4·4–6·9) 5977 9·1% (7·8–10·7) 4·9% (4·2–6·2) 4·2% (3·5–5·5) 6746 5·0% (4·3–6·1) 2·9% (2·3–3·5) 2·1% (1·6–2·9)
Central African Republic 1204 44·6% (34·1–57) 25·7% (17·1–38·8) 24·3% (17·1–35·2) 1304 33·4% (24·9–45·7) 14·4% (8·0–25·2) 21·0% (14·9–32·4) 1369 28·2% (19·9–40·7) 11·2% (5·3–22·4) 18·5% (12·7–29·1)
Chad 3361 40·9% (37·0–45·3) 37·9% (34·0–42·4) 4·9% (3·6–7·1) 3949 27·1% (24·3–31·1) 23·8% (20·9–27·3) 4·3% (3·0–6·2) 4514 20·1% (17·8–22·5) 18·0% (15·7–20·4) 2·6% (1·8–3·9)
Congo 1008 7·8% (6·1–10·5) 6·9% (5·3–9·9) 0·8% (0·5–1·4) 1186 3·8% (2·9–5·6) 3·2% (2·3–4·8) 0·6% (0·3–1·4) 1362 2·2% (1·6–3·2) 1·8% (1·3–2·9) 0·3% (0·2–0·7)
Côte d'Ivoire 5228 21·5% (20–23·4) 12·5% (11·0–14·1) 10·3% (9·2–11·8) 5747 13·7% (12·6–14·9) 5·7% (4·9–6·5) 8·5% (7·6–9·4) 6341 9·6% (8·7–10·8) 4·1% (3·4–5) 5·7% (5–6·7)
Democratic Republic of the Congo 18 004 24·6% (22·4–26·8) 15·3% (13·5–17·4) 11·2% (9·7–13·2) 21 646 17·9% (16·1–19·7) 8·2% (7·3–9·5) 10·4% (9·1–12·2) 25 684 10·3% (9·3–11·8) 4·9% (4·3–5·8) 5·7% (4·7–6·7)
Djibouti 165 13·9% (1·3–56·7) 10·8% (1·0–54·5) 0·9% (0·0–20·8) 163 6·0% (0·5–43·1) 4·0% (0·2–35·9) 0·6% (0·0–19·4) 169 4·0% (0·4–32·7) 3·0% (0·2–30·1) 0·4% (0–9·9)
Equatorial Guinea 281 3·7% (1·9–8·3) 2·0% (1·0–4·5) 1·6% (0·5–4·9) 344 2·0% (1·0–4·3) 0·7% (0·3–1·6) 1·3% (0·5–3·2) 414 1·5% (0·6–3·3) 0·5% (0·2–1·4) 0·8% (0·3–2·7)
Eritrea 1184 6·9% (4·7–9·7) 2·4% (0·9–5·6) 4·4% (3·3–6·2) 1486 5·4% (3·9–7·8) 1·0% (0·3–3·4) 4·4% (3·2–5·6) 1572 2·2% (1·4–3·9) 0·5% (0·1–1·8) 1·6% (1·2–2·7)
Eswatini 269 4·2% (3·0–5·9) 2·2% (1·3–3·4) 1·9% (1·2–3·1) 279 2·3% (1·6–3·3) 0·8% (0·5–1·3) 1·5% (0·9–2·5) 283 1·3% (1·0–2·1) 0·5% (0·3–0·8) 0·9% (0·6–1·6)
Ethiopia 22 766 16·4% (11·9–24·1) 10·3% (5·4–17·4) 6·8% (5·7–8·7) 24 497 11·0% (7·9–16·3) 5·1% (2·2–9·9) 6·3% (5·0–7·8) 26 358 6·5% (4·8–10·7) 3·1% (1·6–7·3) 3·4% (2·5–4·7)
Gabon 372 18·5% (12·1–33·7) 12·6% (8·3–17·3) 6·4% (1·7–25·7) 422 11·4% (6·6–34·6) 5·9% (3·4–9·2) 5·7% (1·5–30·2) 506 8·0% (4·1–27·3) 3·7% (2·3–6·3) 4·2% (0·9–24·6)
Gambia 457 10·0% (8·0–12·5) 9·7% (7·7–12·3) 0·3% (0·1–0·9) 523 4·9% (3·7–6·5) 4·6% (3·5–6·0) 0·2% (0·1–0·6) 606 2·8% (2·1–3·7) 2·7% (2·0–3·6) 0·1% (0·0–0·3)
Ghana 6035 24·1% (20·4–27·8) 19·7% (16·6–23·2) 5·3% (3·2–7·9) 6562 12·7% (10·6–16·4) 9·1% (7·3–11·2) 4·0% (2·6–6·8) 7234 8·5% (6·9–10·9) 6·2% (4·9–8·0) 2·2% (1·4–4·1)
Guinea 3053 45·8% (38·8–51·6) 22·8% (15·8–29·4) 31·8% (25·7–39·6) 3366 38·5% (32·2–47·6) 12·7% (8·5–18·4) 29·6% (23·0–37·8) 3697 32·6% (27·7–38·5) 9·5% (5·8–14·3) 25·2% (19·9–31·2)
Guinea-Bissau 420 15·0% (8·5–38·0) 8·9% (6·4–13·6) 6·4% (0·8–30·0) 478 9·0% (4·0–26·6) 4·0% (2·8–6·3) 4·7% (0·5–23·0) 544 5·3% (2·6–18·4) 2·6% (1·6–4·0) 2·8% (0·4–16·7)
Kenya 10 978 15·7% (11·3–27·0) 10·7% (6·4–22·3) 5·6% (4·4–7·0) 12 327 9·7% (7·2–15·0) 5·3% (3·0–11·6) 4·5% (3·7–5·7) 13 192 5·9% (4·0–11·6) 3·4% (1·7–9·4) 2·5% (2·0–3·9)
Lesotho 422 4·4% (0·8–21·1) 3·1% (0·4–20·3) 0·6% (0·1–6·5) 420 2·2% (0·4–11·7) 1·3% (0·1–10·2) 0·4% (0·1–3·5) 414 1·3% (0·3–9·5) 0·9% (0·1–9·5) 0·2% (0·0–1·8)
Liberia 982 42·1% (36·4–48·5) 21·7% (15·7–29·6) 26·1% (21·8–32·4) 1130 29·6% (25·4–34·5) 10·9% (6·9–16·3) 20·9% (17·6–25·7) 1237 24·0% (19·6–29·9) 9·0% (5·6–14·7) 16·0% (13·5–19·6)
Madagascar 5674 .. .. .. 6258 .. .. .. 6838 29·4% (20·0–41·7) 12·2% (8·5–16·5) 18·5% (11·1–29·6)
Malawi 4200 36·0% (33·6–38·4) 31·0% (28·7–33·3) 7·3% (6·1–8·9) 4882 19·5% (18·1–21·3) 15·1% (13·7–16·8) 5·1% (4·4–6·3) 5421 11·0% (10·0–12·0) 9·0% (8·1–10·1) 2·2% (1·8–2·8)
Mali 3918 39·8% (36·5–42·9) 36·1% (32·8–39·5) 6·7% (5·2–8·5) 4729 22·8% (19·9–26·0) 19·4% (16·4–22·2) 4·2% (3·1–5·9) 5492 17·2% (14·9–19·9) 14·6% (12·3–17·2) 3·0% (2·2–4·0)
Mauritania 915 30·3% (25·2–37) 27·7% (23·1–34·7) 3·5% (2·0–6·2) 1040 18·0% (14·7–22·2) 15·5% (12·6–19·8) 2·6% (1·4–4·6) 1189 13·2% (10·3–16·6) 11·6% (8·9–14·8) 1·7% (0·8–3·2)
Mozambique 6544 42·9% (33·1–50·8) 34·3% (25·7–43·4) 11·7% (6·6–20·3) 7599 28·6% (21·0–36·5) 20·8% (14·6–28·4) 9·6% (4·9–17·3) 8562 22·1% (16·3–28·6) 16·1% (11·6–23·1) 6·7% (3·1–13·2)
Namibia 542 24·1% (19·0–32·5) 19·2% (15·0–28·0) 5·4% (3·7–13·7) 571 14·7% (11·2–23·1) 9·9% (7·3–17·7) 5·0% (3·4–11·2) 644 10·0% (7·7–18·2) 7·1% (5·0–14·9) 2·8% (1·9–6·1)
Niger 4509 21·2% (18·8–23·7) 20·2% (17·9–22·6) 1·2% (0·6–2·5) 5574 9·0% (7·8–10·5) 8·3% (7·2–9·9) 0·7% (0·3–1·6) 6778 6·3% (5·3–7·7) 5·9% (5·0–7·1) 0·4% (0·2–1·0)
Nigeria 40 014 23·5% (21·9–25·8) 21·5% (19·8–23·7) 2·7% (2·2–3·6) 46 384 13·1% (11·9–14·2) 11·1% (10·1–12·2) 2·1% (1·7–2·9) 52 944 7·7% (6·7–8·9) 6·7% (5·9–7·9) 1·1% (0·8–1·7)
Rwanda 2225 2·8% (2·0–3·6) 0·5% (0·3–1·0) 2·2% (1·6–2·9) 2510 2·0% (1·5–2·8) 0·2% (0·1–0·5) 1·8% (1·3–2·6) 2770 1·1% (0·7–1·4) 0·1% (0·1–0·3) 0·9% (0·6–1·3)
Senegal 3426 21·2% (18·9–23·9) 19·1% (17·2–21·8) 3·1% (2·1–5·8) 3958 11·0% (9·3–13·5) 9·1% (7·5–10·4) 2·2% (1·4–4·5) 4592 7·6% (6·3–9·2) 6·1% (5·2–7·3) 1·4% (0·9–2·9)
Sierra Leone 1540 30·1% (25·3–36·5) 13·5% (7·0–21·4) 21·5% (17·7–25·9) 1717 16·9% (13·2–21·5) 3·8% (1·9–7·9) 13·3% (10·3–17·1) 1858 14·9% (11·3–19·5) 2·7% (1·2–5·5) 12·3% (9·4–17·0)
Somalia 2561 23·0% (17·4–32·5) 19·8% (14·0–26·0) 4·0% (1·6–10·8) 2911 14·1% (9·7–21·7) 10·7% (7·3–15·3) 3·4% (1·2–9·6) 3265 10·0% (6·8–15·2) 8·0% (5·3–11·6) 2·1% (0·6–6·5)
South Africa 9471 14·5% (10·8–19·4) 11·6% (8·4–16·5) 3·1% (1·8–6·6) 10 170 9·0% (6·4–13·5) 6·1% (4·4–10·3) 2·8% (1·6–6·2) 11 076 6·4% (4·7–11·4) 4·6% (3·1–8·3) 1·8% (1·0–5·0)
South Sudan 2235 19·0% (15·9–22) 9·8% (7·1–12·9) 10·4% (8·6–12·5) 2480 15·0% (12·9–17·4) 4·8% (3·2–7·1) 10·5% (8·5–12·4) 2553 10·2% (8·3–13·2) 3·4% (2·2–6·0) 7·0% (5·6–8·6)
Sudan 8285 31·3% (24·7–38·5) 20·1% (14·4–28·2) 14·4% (9·3–20·3) 9127 22·4% (16·3–29·8) 10·8% (7·2–17·3) 12·6% (7·7–19·0) 9906 15·6% (11·4–22·0) 7·5% (4·5–13·1) 8·5% (5·3–13·1)
Tanzania 12 124 25·0% (19·8–30·1) 17·6% (13·7–21·8) 8·3% (5·9–11·6) 14 381 16·6% (13·4–20·9) 9·5% (7·2–12·4) 7·6% (5·5–11·5) 16 448 11·9% (9·4–15·4) 7·1% (5·2–10·1) 5·0% (3·5–7·4)
Togo 1578 23·0% (20·6–25·8) 20·8% (18·3–23·4) 2·9% (2·3–3·7) 1818 7·3% (6·4–8·7) 6·3% (5·4–7·7) 1·0% (0·8–1·5) 2011 6·1% (5·4–7·2) 5·2% (4·6–6·3) 0·9% (0·7–1·1)
Uganda 9930 19·0% (14·2–28·5) 8·2% (2·9–17·5) 11·6% (10·2–13·3) 11 642 14·1% (11·4–20·8) 4·1v (1·2–11·3) 10·6% (9·3–11·9) 13 633 7·3% (5·5–11·8) 2·2% (0·6–6·6) 5·1% (4·5–5·8)
Zambia 3798 36·3% (32·6–39·4) 27·3% (23·6–30·3) 14·0% (11·1–17·0) 4406 24·8% (21·6–28·9) 14·5% (12·4–18·0) 11·7% (9·5–14·6) 4950 15·6% (13·4–18·6) 9·9% (8·1–12·6) 6·4% (4·9–8·3)
Zimbabwe 3249 22·5% (20·2–24·6) 17·7% (15·6–20·0) 6·1% (5·1–7·3) 3578 11·9% (10·4–13·3) 7·5% (6·3–8·8) 4·8% (3·8–6·0) 4085 6·3% (5·5–7·4) 4·2% (3·6–5·2) 2·1% (1·6–2·8)
Total 222 910 23·0% (22·1–24·1) 17·4% (16·5–18·5) 7·1% (6·5–7·6) 255 001 14·5% (13·8–15·3) 9·1% (8·6–9·9) 6·0% (5·7–6·6) 287 597 9·6% (9·1–10·2) 6·2% (5·7–6·7) 3·7% (3·4–4·0)
Open in a new tab

Among the 223 million school-aged children in 2010, 51 million (23·0%, 95% BCI 22·1–24·1) were infected with either species (table 3), and approximately 111 million children (109–113) were in need of treatment according to WHO praziquantel guidelines (table 4). In 2019, of 288 million school-aged children, 28 million (9·6%, 9·1–10·2) were infected with either S haematobium or S mansoni (table 3), and 112 million (110–113) required praziquantel in accordance with WHO guidelines (table 4).

Table 4.

Population at risk of schistosomiasis and treatment needs for school-aged children in sub-Saharan Africa

2010
 2019
Population (in 1000s) Prevalence <10% Prevalence 10–50% Prevalence >50% Number of treatments (in 1000s) Population (in 1000s) Prevalence <10% Prevalence 10–50% Prevalence >50% Number of treatments (in 1000s)
Angola 5321 1681 (1340–2451) 2087 (1551–2567) 1463 (649–2148) 3078 (2555–3441) 7721 4749 (3813–6052) 2220 (1285–2803) 472 (175–1570) 3226 (2891–3908)
Benin 2366 448 (324–596) 1168 (982–1327) 727 (600–960) 1465 (1391–1592) 3037 1941 (1659–2118) 913 (746–1175) 162 (111–254) 1268 (1224–1339)
Botswana 407 280 (216–334) 98 (57–139) 29 (14–59) 171 (156–195) 474 414 (365–447) 49 (24–88) 8 (2–28) 172 (164–189)
Burkina Faso 4444 2350 (1987–2745) 1662 (1369–1913) 425 (290–656) 2037 (1910–2203) 5900 5324 (4964–5494) 533 (378–857) 40 (20–100) 2084 (2046–2152)
Burundi 2256 1972 (1781–2074) 253 (161–414) 30 (11–87) 816 (792–869) 3211 3114 (3021–3158) 94 (49–187) 4 (0–15) 1089 (1080–1106)
Cameroon 5191 3149 (2849–3424) 1544 (1283–1870) 472 (364–646) 2303 (2217–2405) 6746 5841 (5570–6016) 803 (633–1073) 84 (48–153) 2434 (2397–2500)
Central African Republic 1204 267 (145–425) 417 (312–538) 517 (354–714) 819 (717–933) 1369 543 (369–754) 486 (359–597) 321 (196–532) 752 (659–880)
Chad 3361 816 (603–994) 1238 (1079–1412) 1305 (1120–1507) 2196 (2081–2315) 4514 2360 (2144–2542) 1526 (1377–1684) 637 (504–771) 2181 (2089–2276)
Congo 1008 802 (710–864) 161 (97–251) 34 (20–70) 382 (367–408) 1362 1286 (1231–1310) 55 (37–108) 6 (2–15) 464 (457–475)
Côte d'Ivoire 5228 2429 (2119–2675) 2027 (1783–2338) 747 (655–886) 2569 (2502–2671) 6341 4726 (4490–4907) 1297 (1147–1511) 281 (224–351) 2510 (2460–2569)
Democratic Republic of the Congo 18 004 8232 (7512–8911) 6340 (5767–6918) 3433 (2985–3967) 9361 (9019–9730) 25 684 18 645 (17 781–19 445) 5746 (5037–6409) 1298 (1013–1667) 10 369 (10 131–10 700)
Djibouti 165 95 (19–154) 42 (3–98) 8 (0–102) 68 (53–126) 169 146 (51–162) 15 (0–90) 1 (0–58) 57 (54–99)
Equatorial Guinea 281 239 (202–256) 25 (10–55) 1 (0–12) 94 (91–105) 414 382 (356–391) 11 (2–33) 0 (0–5) 133 (131–140)
Eritrea 1184 965 (879–1039) 187 (128–256) 26 (11–54) 443 (423–470) 1572 1490 (1429–1528) 71 (36–113) 5 (1–21) 537 (529–558)
Eswatini 269 241 (223–253) 26 (14–44) 1 (0–3) 95 (92–98) 283 278 (270–281) 5 (2–13) 0 (0–1) 95 (95–97)
Ethiopia 22 766 13 873 (10 920–16 084) 6418 (5058–8112) 2456 (1511–4306) 10 303 (9504–11 765) 26 358 21 953 (19 279–23 125) 3602 (2739–5375) 738 (416–1615) 9889 (9540–10 712)
Gabon 372 189 (96–260) 136 (66–220) 41 (17–118) 171 (150–225) 506 390 (220–453) 84 (34–232) 14 (2–130) 190 (175–277)
Gambia 457 331 (276–352) 83 (65–138) 26 (16–36) 178 (172–189) 606 540 (524–554) 41 (30–55) 3 (1–10) 204 (201–209)
Ghana 6035 2276 (1836–2699) 2613 (2267–2985) 988 (742–1280) 3051 (2865–3249) 7234 5306 (4761–5717) 1555 (1179–1988) 177 (110–348) 2737 (2640–2885)
Guinea 3053 793 (600–1010) 802 (598–1031) 1382 (1136–1620) 2053 (1893–2182) 3697 1379 (1041–1680) 1165 (960–1408) 1068 (811–1317) 2112 (1956–2266)
Guinea-Bissau 420 262 (102–318) 110 (72–214) 36 (14–157) 182 (162–269) 544 461 (295–502) 66 (29–200) 9 (2–96) 195 (185–256)
Kenya 10 978 6689 (4435–7841) 3188 (2312–4384) 1024 (667–2319) 4868 (4483–5905) 13 192 11 121 (9482–11 814) 1716 (1157–2886) 283 (141–1037) 4849 (4659–5448)
Lesotho 422 375 (227–418) 42 (4–132) 4 (0–64) 151 (142–204) 414 404 (318–413) 10 (1–80) 0 (0–21) 140 (138–165)
Liberia 982 163 (136–216) 383 (310–494) 401 (285–487) 650 (587–696) 1237 483 (357–614) 510 (387–651) 205 (144–306) 623 (577–687)
Madagascar 5674 .. .. .. .. 6838 2411 (1595–3423) 1930 (1250–2310) 2073 (1336–3062) 3904 (2684–4905)
Malawi 4200 758 (638–910) 2203 (2070–2360) 1231 (1076–1381) 2590 (2502–2676) 5421 3643 (3443–3817) 1591 (1436–1785) 181 (127–243) 2194 (2146–2251)
Mali 3918 1039 (913–1184) 1425 (1262–1648) 1456 (1240–1649) 2517 (2398–2623) 5490 3195 (2881–3471) 1688 (1419–1941) 604 (441–831) 2519 (2394–2660)
Mauritania 915 325 (220–432) 347 (269–441) 232 (176–307) 515 (477–572) 1185 795 (704–870) 288 (237–377) 89 (59–135) 500 (476–534)
Mozambique 6544 1425 (847–2081) 2273 (1930–2689) 2696 (1881–3364) 4296 (3787–4715) 8562 4046 (3187–4926) 2901 (2397–3460) 1384 (872–2090) 4191 (3806–4644)
Namibia 542 193 (154–268) 235 (195–283) 83 (56–151) 275 (252–313) 644 450 (369–499) 149 (113–223) 22 (12–90) 252 (240–296)
Niger 4509 2151 (1944–2429) 1705 (1490–1905) 642 (518–816) 2217 (2125–2329) 6775 5614 (5318–5852) 1035 (813–1312) 117 (80–197) 2512 (2459–2589)
Nigeria 40 014 17 584 (16 276–18 843) 15 480 (14 585–16 371) 6777 (6004–7980) 20 376 (19 828–21 185) 52 943 41 520 (40 188–42 811) 9734 (8692–10 646) 1394 (1067–2162) 20 129 (19 772–20 722)
Rwanda 2225 2082 (2019–2134) 137 (90–202) 5 (0–19) 768 (757–785) 2770 2729 (2686–2753) 41 (16–84) 0 (0–6) 931 (926–938)
Senegal 3426 1677 (1398–1857) 1031 (892–1332) 555 (480–657) 1638 (1571–1718) 4592 3507 (3304–3653) 717 (604–895) 155 (108–217) 1685 (1643–1751)
Sierra Leone 1540 542 (340–652) 470 (367–632) 418 (323–542) 830 (774–910) 1858 1085 (861–1205) 449 (373–649) 177 (102–284) 768 (718–841)
Somalia 2561 1164 (878–1442) 861 (639–1049) 447 (278–737) 1269 (1148–1464) 3265 2328 (1938–2626) 668 (455–983) 166 (71–352) 1283 (1193–1427)
South Africa 9471 6237 (5635–6907) 2164 (1750–2513) 970 (573–1501) 4135 (3834–4478) 11 075 9163 (8295–9661) 1471 (1101–1886) 318 (184–940) 4120 (3968–4522)
South Sudan 2235 1223 (1120–1353) 709 (632–783) 290 (208–377) 1060 (1000–1112) 2553 1882 (1731–1980) 539 (452–634) 132 (82–205) 1028 (990–1088)
Sudan 8285 3231 (2550–3959) 2744 (2355–3184) 2276 (1635–3133) 4740 (4326–5242) 9906 6256 (5329–7130) 2568 (1968–3041) 1044 (587–1735) 4431 (4102–4881)
Tanzania 12 124 5119 (4313–6159) 4530 (4006–5029) 2264 (1536–3072) 6296 (5750–6797) 16 448 11180 (9998–12220) 4110 (3321–4909) 992 (644–1671) 6794 (6455–7303)
Togo 1578 640 (505–749) 668 (574–808) 242 (196–302) 794 (760–832) 2011 1643 (1567–1699) 316 (256–388) 26 (15–52) 732 (719–753)
Uganda 9930 5023 (3598–6191) 3731 (2967–4461) 1133 (720–2089) 4688 (4315–5430) 13633 10 923 (9425–11 673) 2347 (1738–3370) 321 (195–898) 5160 (4994–5672)
Zambia 3798 970 (828–1103) 1617 (1420–1794) 1212 (1021–1397) 2345 (2243–2460) 4950 2841 (2577–3125) 1722 (1447–1981) 374 (239–596) 2187 (2082–2322)
Zimbabwe 3249 1409 (1249–1597) 1362 (1202–1507) 478 (391–570) 1627 (1568–1691) 4084 3374 (3239–3473) 649 (558–778) 61 (43–87) 1511 (1488–1548)
Total 222 910 101 488 (97 243–105 275) 74 985 (72 128–77 284) 39 296 (36 942–42 021) 110 637 (109 062–112 729) 287 585 209 010 (205 180–212 177) 55 987 (53 475–58 957) 13 981 (12 495–16 102) 111 632 (110 455–113 185)
Open in a new tab

Model validation suggested that our models were able to correctly estimate the prevalence within a 95% BCI in 78% of S haematobium locations and 86% of S mansoni locations on average. The mean absolute error was 8% for S haematobium and 5% for S mansoni, with a low percentage of prevalence underestimation of around 2% (appendix p 7).

Discussion

We assessed the effect of large-scale preventive chemotherapy on reducing the prevalence of S haematobium and S mansoni by comparing 2000–10 data (before most countries in sub-Saharan Africa had scaled up their schistosomiasis control programmes) with 2011–14 and 2015–19 data. Additionally, we provide updated estimates of the disease prevalence at high spatial resolution, the number of infected school-aged children, and treatment needs in sub-Saharan Africa for schistosomiasis. Our results should enable disease control programme managers to deliver spatially targeted treatment, according to WHO guidelines, and prioritise disease control in a cost-effective manner. Furthermore, the maps identify regions with high heterogeneity in the geographical distribution of the disease, data sparsity, or large uncertainty in the prevalence estimates and, therefore, could assist control programmes in the design of follow-up surveys for disease monitoring and evaluation.

We estimated that, by comparison with 2000–10, the prevalence of S haematobium was reduced by 67·9% and S mansoni by 53·6% in 2015–19. However, reductions were not uniform, with variations by country and species. Differences in relative reduction rates between species could be due to the higher effectiveness of single-dose praziquantel against S haematobium than S mansoni.22, 23

Countries that had multiple rounds of preventive chemotherapy showed a considerable decline in schistosomiasis prevalence. Our analysis suggests that maintaining high preventive chemotherapy coverage over a 3-year period is associated with reductions in Schistosoma species prevalence. Burkina Faso, Burundi, Gambia, Malawi, Niger, and Togo reported national treatment coverage rates above 75% during 2015–19; therefore, further reduction can be expected in these countries in the coming years. Similarly, Cameroon, Congo, and Zimbabwe with reported national preventive chemotherapy coverage of more than 50% during 2015–19 (prevalence >10% in 2000–10) are expected to result in further reductions of schistosoma prevalence. In 2010, the highest estimates of schistosomiasis population-adjusted prevalence (>40%) were observed in the Central African Republic, Chad, Guinea, Liberia, and Mozambique, with prevalence reduction rates of less than 50% when comparing 2000–10 with 2015–19. In 2019, these same countries had the highest prevalence in sub-Saharan Africa, with all of them reporting low national treatment coverage that were far from the WHO roadmap target of 75%, apart from Liberia, where it was only reached in 2019. Notably, the countries where prevalence was highest did not necessarily correspond to those countries where the number of treatment needs was highest, as treatment needs depend on the population at risk (ie, all population living in an area). Treatment needs in 2019 for the Democratic Republic of the Congo, Ethiopia, Nigeria, and Tanzania amounted to 52 million, with Nigeria requiring approximately 18% (20 million) of the total treatment needs for sub-Saharan Africa.

Preventive chemotherapy campaigns mostly focus on school-aged children, as this age group is considered to be at highest risk of infection and associated morbidity, and schools are a suitable setting for treatment campaigns. Our estimates are based on school-based survey data; there were very few community-based surveys in the Global Neglected Tropical Diseases and other databases. To avoid bias, we did not consider community-based surveys in our analysis. According to our estimates, in 2010, approximately 111 million doses of praziquantel were required for a total of 223 million school-aged children living in sub-Saharan Africa. The required number of tablets varies for each child, as praziquantel is administered according to a child's weight (or height), with a recommended dose of 40 mg/kg of bodyweight. These estimates increased in 2019 to 112 million doses of praziquantel for 288 million school-aged children. The need of praziquantel in the study is based on the treatment of the estimated number of infected people through targeted treatment. However, from an implementation perspective, the need of praziquantel is much higher given that preventive chemotherapy is recommended for the treament of all at-risk children (infected or not) leaving in the adminstrative unit target for treatment. Schistosomiasis prevalence across sub-Saharan Africa has decreased from 23·0% in 2000–10 to 9·6% in 2015–19; however, the population of school-aged children has grown considerably, leading to a slight increase in preventive chemotherapy needs when considering existing WHO guidelines.21 The 2015 estimates by Lai and colleagues13 were 122 million doses for a population of 228 million school-aged children in 2012. Considering the population growth and the schistosomiasis prevalence, Lai and colleagues'13 estimates are consistent with the estimates presented in this Article.

The geostatistical models suggested an increased S haematobium and S mansoni risk in areas located in close proximity to freshwater bodies. Such habitats are required to complete the schistosomiasis lifecycle, which involves an intermediate host snail. The parasites require optimal environmental conditions to survive; extreme humidity or high temperatures are detrimental for survival. In our study, we had a smaller number of statistically significant climatic predictors than Lai and colleagues,13 most likely because of the effect of interventions, which blur the effect of climatic factors on disease transmission.

Improved water and sanitation in the general population are associated with a lower risk of schistosomiasis and are considered as supplementary strategies in control and elimination programmes.24, 25 In our study, high coverage of improved sanitation was associated with lower risk of infection for S haematobium and S mansoni. Reports from WHO suggested that sub-Saharan African countries in 2015 had a coverage of basic sanitation services below 50%, apart from South Africa and Botswana (50–75% coverage).26 Improvements in sanitation comparing the situation in 2000–10 with 2011–19 are relatively modest. In fact, almost half of the countries had improvement rates below 20% at the national level. In Ethiopia, the proportion of households with improved sanitation has increased by almost 80%, from 15% in 2005 to 28% in 2015, although admittedly from a low starting point. Angola, Benin, Burkina Faso, Guinea, Guinea-Bissau, Mauritania, and Niger follow, with improvement rates varying from 42% (Niger) to 31% (Guinea). These countries also showed high Schistosoma species prevalence reductions. However, in Gambia, Nigeria, South Sudan, and Zimbabwe, the proportion of households with improved sanitation decreased at the national level when comparing prevalence in 2000–10 with 2011–14 and 2015–19, yet we observed high Schistosoma species reduction rates in these countries.

Our model-based estimates of the geographical patterns of S haematobium prevalence in 2000–10 showed similarities to those reported by Lai and colleagues.13 Countries such as Guinea and Namibia initiated cross-sectional surveys after 2011, which enabled us to obtain more accurate estimates. For 2011–19, the predictions for S haematobium prevalence showed a change in the geographical distribution across all of the included sub-Saharan African countries due to a major reduction of existing disease clusters in most of these countries. The only remaining areas with prevalence above 50% were settings in Senegal (around Kénièba near to the border with Mali), in Guinea (around Touba near the Guinea-Bissauan and Senegalese border, and around Banora near to the Malian border), and in the Democratic Republic of the Congo in the southwestern-most bank of the Congo River. By comparison with prevalence estimates for 2000–10, areas considered at high risk are much smaller in size, with prevalence at moderate levels (between 10% and 50%). The majority of high prevalence (>50%) settings can be found in the vicinity of large rivers (eg, the Niger River) or lakes (eg, Lake Victoria). The geographical pattern of S mansoni for 2000–10 is in accordance with the pattern observed by Lai and colleagues,13 with the exception of Djibouti, Mozambique, and South Africa, for which we had no or very sparse data. During 2015–19, estimates showed a decline in countries that were previously known to be endemic. The reduction is most noticeable in Cameroon, the Democratic Republic of the Congo, and Ethiopia, where high-risk settings identified in 2000–10 have considerably decreased in size. Existing high-risk areas in west Africa still persist, although they have slightly decreased in size. In central and east Africa, S mansoni has decreased considerably across all countries; a few high-risk areas with infection prevalence among school-aged children above 50% are located in close proximity to major rivers or lakes. In Angola, a major high-risk area is noticeable in the northwestern region in a large area around Uige.

The main limitation in our study is the scarcity of more granular data before 2010, which might have been caused by purposeful selection of sites for surveys in known endemic areas at that time. The prevalence during 2000–10 could have been overestimated because of this bias, leading to an overestimation of the effect of preventive chemotherapy. The blank areas (scarce or no observed data with low predicted prevalence) might have been of low or no prevalence historically. The large number of surveys included in the analysis, obtained from heterogeneous sampling designs, made assessing their risk of bias difficult. Prevalence values were based mainly on urine filtration and reagent strip testing for S haematobium and on Kato-Katz stool examinations for S mansoni. Although most of the surveys used the same diagnostic technique, bias might have been introduced by differences in the sampling efforts (eg, single strip for urine sample versus multiple samples). Unfortunately, the sampling effort is not always reported; therefore, it is not possible to adjust the estimates for this source of heterogeneity. In most cases urine filtration reagent strip testing or Kato-Katz thick smear examinations were done on one specimen; however, some data sources used two or three specimens from different days and calculated the mean prevalence.

Our analysis is based on survey data obtained through school-based sampling, however, information on school attendance was not available to adjust the models for possible selection bias in locations where school attendance was low. Most of the data were aggregated over different age groups for school-aged children, thus we could not obtain age-specific risk estimates. Bias might occur when the age distribution in the survey population differs across locations as different age groups might have different infection risks. Furthermore, long-term data are scarce as most countries started doing national surveys only after 2011. This limitation did not allow for a full spatiotemporal assessment of disease prevalence similar to the works of Blangiardo and colleagues,19 and Chammartin and colleagues,20 who analysed periodic data in time over the same set of locations. The most prominent examples in our study were Chad, Mozambique, and Sudan, which had available data only after 2011, showing a high prevalence of schistosomiasis. The spatiotemporal models for these countries predicted higher disease prevalence during 2015–19 than 2000–10. This outcome might be an artefact of the imposed temporal structure, where observed data from the period 2015–19 only partially inform the predictions during 2000–10. Data were scarce in Central African Republic, Djibouti, and Equatorial Guinea; therefore, estimates for these countries might not be accurate. Our estimates of schistosomiasis prevalence assume that the probability of infection with one species does not affect the infection probability of the other species, an assumption also made in previous studies.13, 27

In accordance with WHO guidelines, many countries in sub-Saharan Africa scaled up preventive chemotherapy in an effort to control and eliminate schistosomiasis. Our model-based predictions confirm that schistosomiasis decreased significantly during the period of intensified control and several countries (eg, Burundi, Eritrea, Eswatini, Gambia, Lesotho, and Rwanda) could already be in a position to start considering elimination strategies. Our observations suggest that it is feasible to get to a low prevalence of, and perhaps eliminate, schistosomiasis as a public health problem, as policy makers and WHO and the national programmes further amplify their efforts for disease control. Large-scale preventive chemotherapy targeting all at-risk groups coupled with social and economic development (eg, improvements in sanitation), snail control and information, education, and communication strategies will enable further decline in disease transmission and contribute to elimination. Unfortunately, the COVID-19 pandemic has delayed or disrupted altogether preventive chemotherapy treatment in 2020. Hence, there is considerable concern that the progress made in schistosomiasis control over the past several years is reversing,28, 29 which requires close monitoring and surveillance.

This online publication has been corrected. The corrected version first appeared at thelancet.com/infection on December 22, 2021

Data sharing

The survey data used in this study are available via the Global Neglected Tropical Diseases database. Model-based estimates will also be shared via a web-based application that can be accessed from the Global Neglected Tropical Diseases database. Estimates can also be obtained by contacting the corresponding author.

Declaration of interests

We declare no competing interests.

Acknowledgments

Acknowledgments

The authors are grateful to the Demographic and Health Surveys Program for making the data available and all collaborating institutions whose contributions have enabled this study. This study received funding by the European Research Council (advanced grant project number 323180). The study has also received funding from WHO.

Editorial note: the Lancet Group takes a neutral position with respect to territorial claims in published maps and institutional affiliations.

Contributors

CK processed and analysed the data, interpreted the results, and wrote the first draft of the manuscript. CK, MM, GY, LW, and PV contributed to the systematic review and data extraction. PV extracted the water, sanitation, and climatic data. CK, JU, and PV developed the protocol and search strategy for the systematic review. RNB, DGC, AMD, UFE, FMF, MDF, AK, JBM, NM, PNMM, EKN, MRP, MS, L-ATT, EMT, and PAU provided substantial data. PV formulated research goals and objectives; planned, coordinated, and executed research; and spearheaded study methodology development and manuscript writing. AG, JU, and PV conceptualised the study and revised the manuscript. AG, DGC, AMD, UFE, FMF, MDF, AK, JBM, NM, PNMM, EKN, MRP, MS, L-ATT, EMT, PAU, YZ, JU, and PV provided important intellectual content. PV and MM accessed and verified all the data in the study. All authors approved the final version of the paper before submission. The authors had full access to all the data in the study and had final responsibility for the decision to submit for publication.

Supplementary Material

Supplementary appendix
mmc1.pdf (1.6MB, pdf)

References

  • 1.McManus DP, Dunne DW, Sacko M, Utzinger J, Vennervald BJ, Zhou XN. Schistosomiasis. Nat Rev Dis Primers. 2018;4:13. doi: 10.1038/s41572-018-0013-8. [DOI] [PubMed] [Google Scholar]
  • 2.Kyu HH, Abate D, Abate KH, et al. Global, regional, and national disability-adjusted life-years (DALYs) for 359 diseases and injuries and healthy life expectancy (HALE) for 195 countries and territories, 1990–2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet. 2018;392:1859–1922. doi: 10.1016/S0140-6736(18)32335-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.WHO Schistosomiasis and soil-transmitted helminthiases: numbers of people treated in 2018. Wkly Epidemiol Rec. 2018;94:601–612. [Google Scholar]
  • 4.WHO . WHO; Geneva: 2012. Accelerating work to overcome the global impact of neglected tropical diseases: a roadmap for implementation: executive summary. [Google Scholar]
  • 5.Tchuem Tchuenté LA, Rollinson D, Stothard JR, Molyneux D. Moving from control to elimination of schistosomiasis in sub-Saharan Africa: time to change and adapt strategies. Infect Dis Poverty. 2017;6:42. doi: 10.1186/s40249-017-0256-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Grimes JET, Croll D, Harrison WE, Utzinger J, Freeman MC, Templeton MR. The relationship between water, sanitation and schistosomiasis: a systematic review and meta-analysis. PLoS Negl Trop Dis. 2014;8 doi: 10.1371/journal.pntd.0003296. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Lo NC, Addiss DG, Hotez PJ, et al. A call to strengthen the global strategy against schistosomiasis and soil-transmitted helminthiasis: the time is now. Lancet Infect Dis. 2017;17:e64–e69. doi: 10.1016/S1473-3099(16)30535-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.King CH, Sutherland LJ, Bertsch D. Systematic review and meta-analysis of the impact of chemical-based mollusciciding for control of Schistosoma mansoni and S. haematobium transmission. PLoS Negl Trop Dis. 2015;9 doi: 10.1371/journal.pntd.0004290. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Sokolow SH, Wood CL, Jones IJ, et al. Global assessment of schistosomiasis control over the past century shows targeting the snail intermediate host works best. PLoS Negl Trop Dis. 2016;10 doi: 10.1371/journal.pntd.0004794. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Lai YS, Biedermann P, Ekpo UF, et al. Spatial distribution of schistosomiasis and treatment needs in sub-Saharan Africa: a systematic review and geostatistical analysis. Lancet Infect Dis. 2015;15:927–940. doi: 10.1016/S1473-3099(15)00066-3. [DOI] [PubMed] [Google Scholar]
  • 12.WHO Expanded special project for elimination of neglected tropical diseases. 2017. https://www.afro.who.int/health-topics/expanded-special-project-elimination-neglected-tropical-disease#:~:text=The%20Expanded%20Special%20Project%20for,to%20help%20them%20accelerate%20the
  • 13.Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med. 2009;151:264–269. doi: 10.7326/0003-4819-151-4-200908180-00135. [DOI] [PubMed] [Google Scholar]
  • 14.Hanks EM, Schliep EM, Hooten MB, Hoeting JA. Restricted spatial regression in practice: geostatistical models, confounding, and robustness under model misspecification. Environmetrics. 2015;26:243–254. [Google Scholar]
  • 15.Giardina F, Sogoba N, Vounatsou P. In: Handbook of spatial epidemiology. Lawson AB, Banerjee S, Haining RP, Ugarte MD, editors. CRC Press; Boca Raton: 2016. Bayesian variable selection in semi-parametric and nonstationary geostatistical models: an application to mapping malaria risk in Mali; pp. 399–416. [Google Scholar]
  • 16.Blangiardo M, Cameletti M, Baio G, Rue H. Spatial and spatio-temporal models with R-INLA. Spat Spatio-Temporal Epidemiol. 2013;4:33–49. doi: 10.1016/j.sste.2012.12.001. [DOI] [PubMed] [Google Scholar]
  • 17.Chammartin F, Guimarães LH, Scholte RGC, Bavia ME, Utzinger J, Vounatsou P. Spatio-temporal distribution of soil-transmitted helminth infections in Brazil. Parasit Vectors. 2014;7:440. doi: 10.1186/1756-3305-7-440. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Gelman A, Hwang J, Vehtari A. Understanding predictive information criteria for Bayesian models. Stat Comput. 2014;24:997–1016. [Google Scholar]
  • 19.WHO . WHO; Geneva: 2006. Preventive chemotherapy in human helminthiasis: coordinated use of anthelminthic drugs in control interventions: a manual for health professionals and programme managers. [Google Scholar]
  • 20.WHO Preventive chemotherapy (PC) by country. 2018. https://www.who.int/neglected_diseases/preventive_chemotherapy/sch/en/
  • 21.Held L, Schrödle B, Rue H. In: Statistical modelling and regression structures. Kneib T, Tutz G, editors. Springer; Berlin: 2010. Posterior and cross-validatory predictive checks: a comparison of MCMC and INLA; pp. 91–110. [Google Scholar]
  • 22.Coulibaly JT, N'Gbesso YK, Knopp S, Keiser J, N'Goran EK, Utzinger J. Efficacy and safety of praziquantel in preschool-aged children in an area co-endemic for Schistosoma mansoni and S. haematobium. PLoS Negl Trop Dis. 2012;6 doi: 10.1371/journal.pntd.0001917. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Garba A, Lamine MS, Barkiré N, et al. Efficacy and safety of two closely spaced doses of praziquantel against Schistosoma haematobium and S. mansoni and re-infection patterns in school-aged children in Niger. Acta Trop. 2013;128:334–344. doi: 10.1016/j.actatropica.2012.08.008. [DOI] [PubMed] [Google Scholar]
  • 24.Grimes JET, Croll D, Harrison WE, Utzinger J, Freeman MC, Templeton MR. The roles of water, sanitation and hygiene in reducing schistosomiasis: a review. Parasit Vectors. 2015;8:156. doi: 10.1186/s13071-015-0766-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.French MD, Evans D, Fleming FM, et al. Schistosomiasis in Africa: improving strategies for long-term and sustainable morbidity control. PLoS Negl Trop Dis. 2018;12 doi: 10.1371/journal.pntd.0006484. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.UN UN-Water global analysis and assessment of sanitation and drinking water (GLAAS) 2017 report: financing universal water, sanitation and hygiene under the Sustainable Development Goals. April 12, 2017. https://www.unwater.org/publications/un-water-glaas-2017-financing-universal-water-sanitation-hygiene-sustainable-development-goals/
  • 27.Chammartin F, Houngbedji CA, Hürlimann E, et al. Bayesian risk mapping and model-based estimation of Schistosoma haematobium-Schistosoma mansoni co-distribution in Côte d'Ivoire. PLoS Negl Trop Dis. 2014;8 doi: 10.1371/journal.pntd.0003407. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Ehrenberg JP, Zhou XN, Fontes G, Rocha EMM, Tanner M, Utzinger J. Strategies supporting the prevention and control of neglected tropical diseases during and beyond the COVID-19 pandemic. Infect Dis Poverty. 2020;9:86. doi: 10.1186/s40249-020-00701-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Toor J, Adams ER, Aliee M, et al. Predicted impact of COVID-19 on neglected tropical disease programs and the opportunity for innovation. Clin Infect Dis. 2021;72:1463–1466. doi: 10.1093/cid/ciaa933. [DOI] [PMC free article] [PubMed] [Google Scholar]

Uncited References

  • 11.Hürlimann E, Schur N, Boutsika K, et al. Toward an open-access global database for mapping, control, and surveillance of neglected tropical diseases. PLoS Negl Trop Dis. 2011;5 doi: 10.1371/journal.pntd.0001404. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.International Food Policy Research Institute HarvestChoice. 2011. http://harvestchoice.org/node/4996
  • 31.Lindgren F, Rue H, Lindström J. An explicit link between Gaussian fields and Gaussian Markov random fields: the stochastic partial differential equation approach. J R Stat Soc Series B Stat Methodol. 2011;73:423–498. [Google Scholar]
  • 32.Willmott CJ, Matsuura K. Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance. Clim Res. 2005;30:79–82. [Google Scholar]
  • 33.Rue H, Martino S, Chopin N. Approximate Bayesian inference for latent Gaussian models by using integrated nested Laplace approximations. J R Stat Soc Series B Stat Methodol. 2009;71:319–392. [Google Scholar]
  • 34.Gorelick N, Hancherb M, Dixon M, et al. Google Earth Engine: planetary-scale geospatial analysis for everyone. Remote Sens Environ. 2017;202:18–27. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary appendix
mmc1.pdf (1.6MB, pdf)

Data Availability Statement

The survey data used in this study are available via the Global Neglected Tropical Diseases database. Model-based estimates will also be shared via a web-based application that can be accessed from the Global Neglected Tropical Diseases database. Estimates can also be obtained by contacting the corresponding author.

RESOURCES