Estimating the burden of malaria and soil-transmitted helminth co-2 infection in sub-Saharan Africa: a geospatial study

27 Background : Limited understanding exists about the interactions between malaria and soil-28 transmitted helminths (STH), their potential geographical overlap and the factors driving it. This study 29 characterised the geographical and co-clustered distribution patterns of malaria and STH infections 30 among vulnerable populations in sub-Saharan Africa (SSA). 31 Methodology/Principal findings : We obtained continuous estimates of malaria prevalence from the 32 Malaria Atlas Project and STH prevalence surveys from the WHO-driven Expanded Special Project 33 for the Elimination of NTDs (ESPEN) covering 2000-2018 and used spatial autocorrelation methods 34 to identify statistically significant clusters for both diseases across SSA. We used the inverse distance 35 weighted kriging (interpolation) methods to estimate STH prevalence. We calculated the population-36 weighted prevalence of malaria and STH co-infection, and used the bivariate local indicator of spatial 37 association (LISA analysis) to explore potential co-clustering of both diseases at the implementation 38 unit levels. 39 Our analysis shows spatial variations in the estimates of the prevalence of Plasmodium falciparum 40 co-infections and identified hotspots across many countries in SSA with inter-and intra-country variations. High P. falciparum and high hookworm co-infections were more prevalent Central Africa, high P. falciparum , high Ascaris lumbricoides , high P . falciparum, and high other regions in Conclusions/Significance

Malaria and worms frequently co-exist together among children living in the poorest 55 countries of the world, but little is known about the specific locations of the combined 56 infections involving the two major parasitic diseases and how they interact and change over 57 the years. 58 We used open access data collected by two public registries, that is, the Malaria Atlas Project 59 and Expanded Special Project for the Elimination of NTDs, to understand the overlap of the 60 two diseases in different parts of Africa, where their burden are more predominant. 61 We found significant differences in the distributions of the combined diseases across different 62 parts of Africa, with large concentrations identified in Central and West Africa. For example, 63 double infections with malaria and hookworm were more common in West and Central 64 Africa, whereas malaria and roundworm, and malaria and whipworm were predominantly 65 found in Central Africa. A large collection of the dual infections was also found in some 66 localities within the countries which appeared to have low burden of the two diseases. 67 These findings provide a useful insight into the areas which could be serving as a reservoir to 68 propagating the transmission of the two diseases. The results of this study could also be used 69 to develop and implement integrated control programmes for malaria and worms, and this 70 could help to achieve the WHO NTD roadmap to ending the neglect to attain Sustainable 71 Development Goals by 2030. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint Introduction 77 78 Due to environmental and host factors which favour transmission of multiple parasitic infections, 79 malaria and soil-transmitted helminths (STH), including Ascaris lumbricoides, Trichuris trichiura, 80 and hookworms, co-exist in many parts of the world, predominantly in sub-Saharan Africa (SSA) [ 99 Empirical studies conducted across co-endemic countries have shown that variations in the prevalence 100 of malaria and STH co-infections differed significantly across geographic locations. [5][6][7][8][9] Limited 101 understanding exists about the spatial distribution of these co-infections in vulnerable populations.

102
Majority of the published studies targeting malaria-STH co-infection have focused on describing the . CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted March 15, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022 103 transmission and burden among affected populations, [5,6,9,10] and have paid little attention to 104 their concurrent spatial distribution and causes. Previous systematic reviews have also reported an 105 over-estimation of the relationship between malaria and helminths making it difficult to establish 106 conclusively the burden of malaria-STH dual infections. [11,12] Consistent with findings of these 107 reviews, a recently conducted systemic review and meta-analysis of 55 studies which enrolled 37,559 108 children across low and middle-income countries (LMIC) found a wide variation in the prevalence of 109 malaria-helminth co-infections, ranging from 7-76% across LMICs.
[13] These findings may be due to 110 the low sensitivity of the diagnostic methods employed for the detection of malaria-helminth co-

111
infections in the primary studies included in the systematic review.

112
The is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted March 15, 2022. ; https://doi.org/10.1101/2022.03.14.22272330 doi: medRxiv preprint 156 data points with spatial reference, 6,229 data points had prevalence rates above zero for T. Trichiura, 157 12,945 data points had prevalence rates above zero for hookworm and 9,728 for A.lumbricoides.

158
Incomplete data (those without spatial reference and explicit prevalence rates) were excluded from the 159 analyses.

160
Statistical analysis 161 162 Spatial auto-correlation 163 We used Moran's I statistics as the indicator of spatial auto-correlation. We plotted the prevalence of 164 P. falciparum on the x-axis against the spatially lagged variables of hookworm, T. trichiura and A. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint province. There were a total of 5,935 administrative level 2 in the study areas reported in this paper.

196
To identify the spatial clusters of an implementation unit, we used local Moran's I algorithm in Geoda coldspots, and areas endemic to malaria-STH co-infection across the study areas. We considered areas 202 with high malaria-STH co-infection and alpha level ≤0.001 as statistically significant hotspots.

205
The spatial auto-correlation plot was a linear fit whose slope corresponds to Moran's I and its values is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted March 15, 2022. ; https://doi.org/10.1101/2022.03.14.22272330 doi: medRxiv preprint prevalence of P.falciparum and HW co-infection, there were specific areas (second administrative 214 unit) with statistically significant co-infection (p < 0.001) which can be considered as hotspots for the 215 co-infection.

216
The second administrative units with statistically significant P.falciparum and HW co-infection 221 districts in the Lake Victoria region and eastern Uganda ( Figure 2B). In Nigeria, hotspots for co-

infection with P. falciparum and HW co-infection included coastal communities in the southern and
223 south-eastern parts of the country ( Figure 2B, Table S1). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted March 15, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted March 15, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022 High P.falciparum and high T. trichiura co-infection were more prevalent in the Central African sub-

258
DRC, Equatorial Guinea, Gabon, and coastal communities of southern Nigeria had a high prevalence 259 of P. falciparum and high T. trichiura co-infection ( Figure 3A). Hotspots for P. falciparum and T.  Table S1). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted March 15, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted March 15, 2022. ; https://doi.org/10.1101/2022.03.14.22272330 doi: medRxiv preprint

280
High P.falciparum and high A. lumbricoides (AS) co-infection was more prevalent in Central Africa 281 and parts of West Africa ( Figure 4A). The countries with high prevalence of P.falciparum and high  Table S1). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted March 15, 2022. ; https://doi.org/10.1101/2022.03.14.22272330 doi: medRxiv preprint

306
Discussion 307 308 Our analysis shows spatial variations in the estimates of prevalence of P.falciparum-STH co-

309
infections and identified hotspots across many countries in SSA where the co-infections were high.

310
Consistent with previous studies [3,12,[22][23][24], our findings also highlight inter-and intra-country is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

355
In conclusion, we have demonstrated wide spatial heterogeneity within the SSA and within countries, is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted March 15, 2022. ; https://doi.org/10.1101/2022.03.14.22272330 doi: medRxiv preprint control programmes. Nevertheless, it is important that our findings are confirmed by further empirical 362 studies as this will provide the platform for research agenda that will lead to the establishment of the 363 much-needed platform for the implementation of the integrated programmes that are cost-effective,

364
and make optimum use of limited human resources frequently found in SSA. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted March 15, 2022. ; https://doi.org/10.1101/2022.03.14.22272330 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted March 15, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022