Leveraging risk maps of malaria vector abundance to guide control efforts reduces malaria incidence in Eastern Province, Zambia

Comparing field-collected versus remotely-sensed variables to model malaria risk in the highlands of western Uganda

BACKGROUND

Malaria risk is not uniform across relatively small geographic areas, such as within a village. This heterogeneity in risk is associated with factors including demographic characteristics, individual behaviours, home construction, and environmental conditions, the importance of which varies by setting, making prediction difficult. This study attempted to compare the ability of statistical models to predict malaria risk at the household level using either (i) free easily-obtained remotely-sensed data or (ii) results from a resource-intensive household survey.

METHODS

The results of a household malaria survey conducted in 3 villages in western Uganda were combined with remotely-sensed environmental data to develop predictive models of two outcomes of interest (1) a positive ultrasensitive rapid diagnostic test (uRDT) and (2) inpatient admission for malaria within the last year. Generalized additive models were fit to each result using factors from the remotely-sensed data, the household survey, or a combination of both. Using a cross-validation approach, each model's ability to predict malaria risk for out-of-sample households (OOS) and villages (OOV) was evaluated.

RESULTS

Models fit using only environmental variables provided a better fit and higher OOS predictive power for uRDT result (AIC = 362, AUC = 0.736) and inpatient admission (AIC = 623, AUC = 0.672) compared to models using household variables (uRDT AIC = 376, Admission AIC = 644, uRDT AUC = 0.667, Admission AUC = 0.653). Combining the datasets did not result in a better fit or higher OOS predictive power for uRDT results (AIC = 367, AUC = 0.671), but did for inpatient admission (AIC = 615, AUC = 0.683). Household factors performed best when predicting OOV uRDT results (AUC = 0.596) and inpatient admission (AUC = 0.553), but not much better than a random classifier.

CONCLUSIONS

These results suggest that residual malaria risk is driven more by the external environment than home construction within the study area, possibly due to transmission regularly occurring outside of the home. Additionally, they suggest that when predicting malaria risk the benefit may not outweigh the high costs of attaining detailed information on household predictors. Instead, using remotely-sensed data provides an equally effective, cost-efficient alternative.

Anopheles rufipes implicated in malaria transmission both indoors and outdoors alongside Anopheles funestus and Anopheles arabiensis in rural south-east Zambia

BACKGROUND

The primary malaria vector-control interventions, indoor residual spraying and long-lasting insecticidal nets, are effective against indoor biting and resting mosquito species. Consequently, outdoor biting and resting malaria vectors might elude the primary interventions and sustain malaria transmission. Varied vector biting and resting behaviour calls for robust entomological surveillance. This study investigated the bionomics of malaria vectors in rural south-east Zambia, focusing on species composition, their resting and host-seeking behaviour and sporozoite infection rates.

METHODS

The study was conducted in Nyimba District, Zambia. Randomly selected households served as sentinel houses for monthly collection of mosquitoes indoors using CDC-light traps (CDC-LTs) and pyrethrum spray catches (PSC), and outdoors using only CDC-LTs for 12 months. Mosquitoes were identified using morphological taxonomic keys. Specimens belonging to the Anopheles gambiae complex and Anopheles funestus group were further identified using molecular techniques. Plasmodium falciparum sporozoite infection was determined using sandwich enzyme-linked immunosorbent assays.

RESULTS

From 304 indoor and 257 outdoor light trap-nights and 420 resting collection, 1409 female Anopheles species mosquitoes were collected and identified morphologically; An. funestus (n = 613; 43.5%), An. gambiae sensu lato (s.l.)(n = 293; 20.8%), Anopheles pretoriensis (n = 282; 20.0%), Anopheles maculipalpis (n = 130; 9.2%), Anopheles rufipes (n = 55; 3.9%), Anopheles coustani s.l. (n = 33; 2.3%), and Anopheles squamosus (n = 3, 0.2%). Anopheles funestus sensu stricto (s.s.) (n = 144; 91.1%) and Anopheles arabiensis (n = 77; 77.0%) were the dominant species within the An. funestus group and An. gambiae complex, respectively. Overall, outdoor CDC-LTs captured more Anopheles mosquitoes (mean = 2.25, 95% CI 1.22-3,28) than indoor CDC-LTs (mean = 2.13, 95% CI 1.54-2.73). Fewer resting mosquitoes were collected with PSC (mean = 0.44, 95% CI 0.24-0.63). Sporozoite infectivity rates for An. funestus, An. arabiensis and An. rufipes were 2.5%, 0.57% and 9.1%, respectively. Indoor entomological inoculation rates (EIRs) for An. funestus s.s, An. arabiensis and An. rufipes were estimated at 4.44, 1.15 and 1.20 infectious bites/person/year respectively. Outdoor EIRs for An. funestus s.s. and An. rufipes at 7.19 and 4.31 infectious bites/person/year, respectively.

CONCLUSION

The findings of this study suggest that An. rufipes may play an important role in malaria transmission alongside An. funestus s.s. and An. arabiensis in the study location.

Impact of aerial humidity on seasonal malaria: an ecological study in Zambia

Background

Seasonal patterns of malaria cases in many parts of Africa are generally associated with rainfall, yet in the dry seasons, malaria transmission declines but does not always cease. It is important to understand what conditions support these periodic cases. Aerial moisture is thought to be important for mosquito survival and ability to forage, but its role during the dry seasons has not been well studied. During the dry season aerial moisture is minimal, but intermittent periods may arise from the transpiration of peri-domestic trees or from some other sources in the environment. These periods may provide conditions to sustain pockets of mosquitoes that become active and forage, thereby transmitting malaria. In this work, humidity along with other ecological variables that may impact malaria transmission have been examined.

Methods

Negative binomial regression models were used to explore the association between peri-domestic tree humidity and local malaria incidence. This was done using sensitive temperature and humidity loggers in the rural Southern Province of Zambia over three consecutive years. Additional variables including rainfall, temperature and elevation were also explored.

Results

A negative binomial model with no lag was found to best fit the malaria cases for the full year in the evaluated sites of the Southern Province of Zambia. Local tree and granary night-time humidity and temperature were found to be associated with local health centre-reported incidence of malaria, while rainfall and elevation did not significantly contribute to this model. A no lag and one week lag model for the dry season alone also showed a significant effect of humidity, but not temperature, elevation, or rainfall.

Conclusion

The study has shown that throughout the dry season, periodic conditions of sustained humidity occur that may permit foraging by resting mosquitoes, and these periods are associated with increased incidence of malaria cases. These results shed a light on conditions that impact the survival of the common malaria vector species, Anopheles arabiensis, in arid seasons and suggests how they emerge to forage when conditions permit.

Developing High-Resolution Population and Settlement Data for Impactful Malaria Interventions in Zambia

Foundational high-resolution geospatial data products for population, settlements, infrastructure, and boundaries may greatly enhance the efficient planning of resource allocation during health sector interventions. To ensure the relevance and sustainability of such products, government partners must be involved from the beginning in their creation, improvement, and/or management, so they can be successfully applied to public health campaigns, such as malaria control and prevention. As an example, Zambia had an ambitious strategy of reaching the entire population with malaria vector control campaigns by late 2020 or early 2021, but they lacked the requisite accurate and up-to-date data on infrastructure and population distribution. To address this gap, the Geo-Referenced Infrastructure and Demographic Data for Development (GRID3) program, Akros, and other partners developed maps and planning templates to aid Zambia's National Malaria Elimination Program (NMEP) in operationalizing its strategy.

Multiple Novel Clades of Anopheline Mosquitoes Caught Outdoors in Northern Zambia

Residual vector populations that do not come in contact with the most frequently utilized indoor-directed interventions present major challenges to global malaria eradication. Many of these residual populations are mosquito species about which little is known. As part of a study to assess the threat of outdoor exposure to malaria mosquitoes within the Southern and Central Africa International Centers of Excellence for Malaria Research, foraging female anophelines were collected outside households in Nchelenge District, northern Zambia. These anophelines proved to be more diverse than had previously been reported in the area. In order to further characterize the anopheline species, sequencing and phylogenetic approaches were utilized. Anopheline mosquitoes were collected from outdoor light traps, morphologically identified, and sent to Johns Hopkins Bloomberg School of Public Health for sequencing. Sanger sequencing from 115 field-derived samples yielded mitochondrial COI sequences, which were aligned with a homologous 488 bp gene segment from known anophelines (n = 140) retrieved from NCBI. Nuclear ITS2 sequences (n = 57) for at least one individual from each unique COI clade were generated and compared against NCBI’s nucleotide BLAST database to provide additional evidence for taxonomical identity and structure. Molecular and morphological data were combined for assignment of species or higher taxonomy. Twelve phylogenetic groups were characterized from the COI and ITS2 sequence data, including the primary vector species Anopheles funestus s.s. and An. gambiae s.s. An unexpectedly large proportion of the field collections were identified as An. coustani and An. sp. 6. Six phylogenetic groups remain unidentified to species-level. Outdoor collections of anopheline mosquitoes in areas frequented by people in Nchelenge, northern Zambia, proved to be extremely diverse. Morphological misidentification and underrepresentation of some anopheline species in sequence databases confound efforts to confirm identity of potential malaria vector species. The large number of unidentified anophelines could compromise the malaria vector surveillance and malaria control efforts not only in northern Zambia but other places where surveillance and control are focused on indoor-foraging and resting anophelines. Therefore, it is critical to continue development of methodologies that allow better identification of these populations and revisiting and cleaning current genomic databases.

Using geospatial models to map zero-dose children: factors associated with zero-dose vaccination status before and after a mass measles and rubella vaccination campaign in Southern province, Zambia

INTRODUCTION

Despite gains in global coverage of childhood vaccines, many children remain undervaccinated. Although mass vaccination campaigns are commonly conducted to reach these children their effectiveness is unclear. We evaluated the effectiveness of a mass vaccination campaign in reaching zero-dose children.

METHODS

We conducted a prospective study in 10 health centre catchment areas in Southern province, Zambia in November 2020. About 2 months before a national mass measles and rubella vaccination campaign conducted by the Ministry of Health, we used aerial satellite maps to identify built structures. These structures were visited and diphtheria-tetanus-pertussis (DTP) and measles zero-dose children were identified (children who had not received any DTP or measles-containing vaccines, respectively). After the campaign, households where measles zero-dose children were previously identified were targeted for mop-up vaccination and to assess if these children were vaccinated during the campaign. A Bayesian geospatial model was used to identify factors associated with zero-dose status and measles zero-dose children being reached during the campaign. We also produced fine-scale zero-dose prevalence maps and identified optimal locations for additional vaccination sites.

RESULTS

Before the vaccination campaign, 17.3% of children under 9 months were DTP zero-dose and 4.3% of children 9-60 months were measles zero-dose. Of the 461 measles zero-dose children identified before the vaccination campaign, 338 (73.3%) were vaccinated during the campaign and 118 (25.6%) were reached by a targeted mop-up activity. The presence of other children in the household, younger age, greater travel time to health facilities and living between health facility catchment areas were associated with zero-dose status. Mapping zero-dose prevalence revealed substantial heterogeneity within and between catchment areas. Several potential locations were identified for additional vaccination sites.

CONCLUSION

Fine-scale variation in zero-dose prevalence and the impact of accessibility to healthcare facilities on vaccination coverage were identified. Geospatial modelling can aid targeted vaccination activities.

Satellite Observations and Malaria: New Opportunities for Research and Applications

Satellite remote sensing provides a wealth of information about environmental factors that influence malaria transmission cycles and human populations at risk. Long-term observations facilitate analysis of climate–malaria relationships, and high-resolution data can be used to assess the effects of agriculture, urbanization, deforestation, and water management on malaria. New sources of very-high-resolution satellite imagery and synthetic aperture radar data will increase the precision and frequency of observations. Cloud computing platforms for remote sensing data combined with analysis-ready datasets and high-level data products have made satellite remote sensing more accessible to nonspecialists. Further collaboration between the malaria and remote sensing communities is needed to develop and implement useful geospatial data products that will support global efforts toward malaria control, elimination, and eradication.

Comparing prioritization strategies for delivering indoor residual spray (IRS) implementation, using a network approach

Background

Indoor residual spraying (IRS) is an effective method to control malaria-transmitting Anopheles mosquitoes and often complements insecticide-treated mosquito nets, the predominant malaria vector control intervention. With insufficient funds to cover every household, malaria control programs must balance the malaria risk to a particular human community against the financial cost of spraying that community. This study creates a framework for modelling the distance to households for targeting IRS implementation, and applies it to potential risk prioritization strategies in four provinces (Luapula, Muchinga, Eastern, and Northern) in Zambia.

Methods

Optimal network models were used to assess the travel distance of routes between operations bases and human communities identified through remote sensing. Network travel distances were compared to Euclidean distances, to demonstrate the importance of accounting for road routes. The distance to reaching communities for different risk prioritization strategies were then compared assuming sufficient funds to spray 50% of households, using four underlying malarial risk maps: (a) predicted Plasmodium falciparum parasite rate in 2–10 years olds (PfPR), or (b) predicted probability of the presence of each of three main malaria transmitting anopheline vectors (Anopheles arabiensis, Anopheles funestus, Anopheles gambiae).

Results

The estimated one-way network route distance to reach communities to deliver IRS ranged from 0.05 to 115.69 km. Euclidean distance over and under-estimated these routes by − 101.21 to 41.79 km per trip, as compared to the network route method. There was little overlap between risk map prioritization strategies, both at a district-by-district scale, and across all four provinces. At both scales, agreement for inclusion or exclusion from IRS across all four prioritization strategies occurred in less than 10% of houses. The distances to reaching prioritized communities were either lower, or not statistically different from non-prioritized communities, at both scales of strategy.

Conclusion

Variation in distance to targeted communities differed depending on risk prioritization strategy used, and higher risk prioritization did not necessarily translate into greater distances in reaching a human community. These findings from Zambia suggest that areas with higher malaria burden may not necessarily be more remote than areas with lower malaria burden.