Using image segmentation models to analyse high-resolution earth observation data: new tools to monitor disease risks in changing environments.

BACKGROUND: In the near future, the incidence of mosquito-borne diseases may expand to new sites due to changes in temperature and rainfall patterns caused by climate change. Therefore, there is a need to use recent technological advances to improve vector surveillance methodologies. Unoccupied Aerial Vehicles (UAVs), often called drones, have been used to collect high-resolution imagery to map detailed information on mosquito habitats and direct control measures to specific areas. Supervised classification approaches have been largely used to automatically detect vector habitats. However, manual data labelling for model training limits their use for rapid responses. Open-source foundation models such as the Meta AI Segment Anything Model (SAM) can facilitate the manual digitalization of high-resolution images. This pre-trained model can assist in extracting features of interest in a diverse range of images. Here, we evaluated the performance of SAM through the Samgeo package, a Python-based wrapper for geospatial data, as it has not been applied to analyse remote sensing images for epidemiological studies. RESULTS: We tested the identification of two land cover classes of interest: water bodies and human settlements, using different UAV acquired imagery across five malaria-endemic areas in Africa, South America, and Southeast Asia. We employed manually placed point prompts and text prompts associated with specific classes of interest to guide the image segmentation and assessed the performance in the different geographic contexts. An average Dice coefficient value of 0.67 was obtained for buildings segmentation and 0.73 for water bodies using point prompts. Regarding the use of text prompts, the highest Dice coefficient value reached 0.72 for buildings and 0.70 for water bodies. Nevertheless, the performance was closely dependent on each object, landscape characteristics and selected words, resulting in varying performance. CONCLUSIONS: Recent models such as SAM can potentially assist manual digitalization of imagery by vector control programs, quickly identifying key features when surveying an area of interest. However, accurate segmentation still requires user-provided manual prompts and corrections to obtain precise segmentation. Further evaluations are necessary, especially for applications in rural areas.

Genetic markers associated with the widespread insecticide resistance in malaria vector Anopheles funestus populations across Tanzania.

BACKGROUND: Anopheles funestus is a leading vector of malaria in most parts of East and Southern Africa, yet its ecology and responses to vector control remain poorly understood compared with other vectors such as Anopheles gambiae and Anopheles arabiensis. This study presents the first large-scale survey of the genetic and phenotypic expression of insecticide resistance in An. funestus populations in Tanzania. METHODS: We performed insecticide susceptibility bioassays on An. funestus mosquitoes in nine regions with moderate-to-high malaria prevalence in Tanzania, followed by genotyping for resistance-associated mutations (CYP6P9a, CYP6P9b, L119F-GSTe2) and structural variants (SV4.3 kb, SV6.5 kb). Generalized linear models were used to assess relationships between genetic markers and phenotypic resistance. An interactive R Shiny tool was created to visualize the data and support evidence-based interventions. RESULTS: Pyrethroid resistance was universal but reversible by piperonyl-butoxide (PBO). However, carbamate resistance was observed in only five of the nine districts, and dichloro-diphenyl-trichloroethane (DDT) resistance was found only in the Kilombero valley, south-eastern Tanzania. Conversely, there was universal susceptibility to the organophosphate pirimiphos-methyl in all sites. Genetic markers of resistance had distinct geographical patterns, with CYP6P9a-R and CYP6P9b-R alleles, and the SV6.5 kb structural variant absent or undetectable in the north-west but prevalent in all other sites, while SV4.3 kb was prevalent in the north-western and western regions but absent elsewhere. Emergent L119F-GSTe2, associated with deltamethrin resistance, was detected in heterozygous form in districts bordering Mozambique, Malawi and the Democratic Republic of Congo. The resistance landscape was most complex in western Tanzania, in Tanganyika district, where all five genetic markers were detected. There was a notable south-to-north spread of resistance genes, especially CYP6P9a-R, though this appears to be interrupted, possibly by the Rift Valley. CONCLUSIONS: This study underscores the need to expand resistance monitoring to include An. funestus alongside other vector species, and to screen for both the genetic and phenotypic signatures of resistance. The findings can be visualized online via an interactive user interface and could inform data-driven decision-making for resistance management and vector control. Since this was the first large-scale survey of resistance in Tanzania's An. funestus, we recommend regular updates with greater geographical and temporal coverage.

Geospatial modelling of dry season habitats of the malaria vector, Anopheles funestus, in south-eastern Tanzania.

BACKGROUND: Anopheles funestus is a major malaria vector in Eastern and Southern Africa and is currently the dominant malaria-transmitting vector in many parts of Tanzania. Previous research has identified its preference for specific aquatic habitats, especially those that persist in dry months. This observation suggests the potential for targeted control through precise habitat mapping and characterization. In this study, we investigated the influence of habitat characteristics, land cover and human population densities on An. funestus distribution during dry seasons. Based on the results, we developed a habitat suitability model for this vector species in south-eastern Tanzania. METHODS: Eighteen villages in south-eastern Tanzania were surveyed during the dry season from September-December 2021. Water bodies were systematically inspected for mosquito larvae and characterized by their physico-chemical characteristics and surrounding environmental features. A generalized linear model was used to assess the presence of An. funestus larvae as a function of the physico-chemical characteristics, land use and human population densities. The results obtained from this model were used to generate spatially explicit predictions of habitat suitability in the study districts. RESULTS: Of the 1466 aquatic habitats surveyed, 440 were positive for An. funestus, with river streams having the highest positivity (74%; n = 322) followed by ground pools (15%; n = 67). The final model had an 83% accuracy in predicting positive An. funestus habitats, with the most important characteristics being permanent waters, clear waters with or without vegetation or movement and shading over the habitats. There was also a positive association of An. funestus presence with forested areas and a negative association with built-up areas. Human population densities had no influence on An. funestus distribution. CONCLUSIONS: The results of this study underscore the crucial role of both the specific habitat characteristics and key environmental factors, notably land cover, in the distribution of An. funestus. In this study area, An. funestus predominantly inhabits river streams and ground pools, with a preference for clear, perennial waters with shading. The strong positive association with more pristine environments with tree covers and the negative association with built-up areas underscore the importance of ecological transitions in vector distribution and malaria transmission risk. Such spatially explicit predictions could enable more precise interventions, particularly larval source management, to accelerate malaria control.

Perspectives of African stakeholders on gene drives for malaria control and elimination: a multi-country survey.

BACKGROUND: Gene drive modified mosquitoes (GDMMs) have the potential to address Africa's persistent malaria problem, but are still in early stages of development and testing. Continuous engagement of African stakeholders is crucial for successful evaluation and implementation of these technologies. The aim of this multi-country study was, therefore, to explore the insights and recommendations of key stakeholders across Africa on the potential of GDMMs for malaria control and elimination in the continent. METHODS: A concurrent mixed-methods study design was used, involving a structured survey administered to 180 stakeholders in 25 countries in sub-Saharan Africa, followed by 18 in-depth discussions with selected groups and individuals. Stakeholders were drawn from academia, research and regulatory institutions, government ministries of health and environment, media and advocacy groups. Thematic content analysis was used to identify key topics from the in-depth discussions, and descriptive analysis was done to summarize information from the survey data. RESULTS: Despite high levels of awareness of GDMMs among the stakeholders (76.7%), there was a relatively low-level of understanding of their key attributes and potential for malaria control (28.3%). When more information about GDMMs was provided to the stakeholders, they readily discussed their insights and concerns, and offered several recommendations to ensure successful research and implementation of the technology. These included: (i) increasing relevant technical expertise within Africa, (ii) generating local evidence on safety, applicability, and effectiveness of GDMMs, and (iii) developing country-specific regulations for safe and effective governance of GDMMs. A majority of the respondents (92.9%) stated that they would support field trials or implementation of GDMMs in their respective countries. This study also identified significant misconceptions regarding the phase of GDMM testing in Africa, as several participants incorrectly asserted that GDMMs were already present in Africa, either within laboratories or released into the field. CONCLUSION: Incorporating views and recommendations of African stakeholders in the ongoing research and development of GDMMs is crucial for instilling stakeholder confidence on their potential application. These findings will enable improved planning for GDMMs in Africa as well as improved target product profiles for the technologies to maximize their potential for solving Africa's enduring malaria challenge.

Effects of vegetation densities on the performance of attractive targeted sugar baits (ATSBs) for malaria vector control: a semi-field study.

BACKGROUND: Attractive targeted sugar baits (ATSBs) control sugar-feeding mosquitoes with oral toxicants, and may effectively complement core malaria interventions, such as insecticide-treated nets even where pyrethroid-resistance is widespread. The technology is particularly efficacious in arid and semi-arid areas. However, their performance remains poorly-understood in tropical areas with year-round malaria transmission, and where the abundant vegetation constitutes competitive sugar sources for mosquitoes. This study compared the efficacies of ATSBs (active ingredient: 2% boric acid) in controlled settings with different vegetation densities. METHODS: Potted mosquito-friendly plants were introduced inside semi-field chambers (9.6 m by 9.6 m) to simulate densely-vegetated, sparsely-vegetated, and bare sites without any vegetation (two chambers/category). All chambers had volunteer-occupied huts. Laboratory-reared Anopheles arabiensis were released nightly (200/chamber) and host-seeking females recaptured using human landing catches outdoors (8.00 p.m.-9.00 p.m.) and CDC-light traps indoors (9.00 p.m.-6.00 a.m.). Additionally, resting mosquitoes were collected indoors and outdoors each morning using Prokopack aspirators. The experiments included a "before-and-after" set-up (with pre-ATSBs, ATSBs and post-ATSBs phases per chamber), and a "treatment vs. control" set-up (where similar chambers had ATSBs or no ATSBs). The experiments lasted 84 trap-nights. RESULTS: In the initial tests when all chambers had no vegetation, the ATSBs reduced outdoor-biting by 69.7%, indoor-biting by 79.8% and resting mosquitoes by 92.8%. In tests evaluating impact of vegetation, the efficacy of ATSBs against host-seeking mosquitoes was high in bare chambers (outdoors: 64.1% reduction; indoors: 46.8%) but modest or low in sparsely-vegetated (outdoors: 34.5%; indoors: 26.2%) and densely-vegetated chambers (outdoors: 25.4%; indoors: 16.1%). Against resting mosquitoes, the ATSBs performed modestly across settings (non-vegetated chambers: 37.5% outdoors and 38.7% indoors; sparsely-vegetated: 42.9% outdoors and 37.5% indoors; densely-vegetated: 45.5% outdoors and 37.5% indoors). Vegetation significantly reduced the ATSBs efficacies against outdoor-biting and indoor-biting mosquitoes but not resting mosquitoes. CONCLUSION: While vegetation can influence the performance of ATSBs, the devices remain modestly efficacious in both sparsely-vegetated and densely-vegetated settings. Higher efficacies may occur in places with minimal or completely no vegetation, but such environments are naturally unlikely to sustain Anopheles populations or malaria transmission in the first place. Field studies therefore remain necessary to validate the efficacies of ATSBs in the tropics.

Common predators and factors influencing their abundance in Anopheles funestus aquatic habitats in rural south-eastern Tanzania.

BACKGROUND: The role of larval predators in regulating the Anopheles funestus population in various malaria-endemic countries remains relatively unknown. This study aimed to investigate the common predators that co-exist with Anopheles funestus group larvae and evaluate factors that influence their abundance in rural south-eastern Tanzania. METHODS: Mosquito larvae and predators were sampled concurrently using standard dipper (350 ml) or 10 L bucket in previously identified aquatic habitats in selected villages in southern Tanzania. Predators and mosquito larvae were identified using standard identification keys. All positive habitats were geo-located and their physical features characterized. Water physicochemical parameters such as dissolved oxygen (DO), pH, electrical conductivity (EC), total dissolved solids (TDS) and temperature were also recorded. RESULTS: A total of 85 previously identified An. funestus aquatic habitats in nine villages were sampled for larvae and potential predators. A total of 8,295 predators were sampled. Of these Coenagrionidae 57.7% (n = 4785), Corixidae 12.8% (n = 1,060), Notonectidae 9.9% (n = 822), Aeshnidae 4.9% (n = 405), Amphibian 4.5% (n = 370), Dytiscidae 3.8% (n = 313) were common. A total of 5,260 mosquito larvae were sampled, whereby Anopheles funestus group were 60.3% (n = 3,170), Culex spp. 24.3% (n = 1,279), An. gambie s.l. 8.3% (n = 438) and other anophelines 7.1% (n = 373). Permanent and aquatic habitats larger than 100m2 were positively associated with An. funestus group larvae (P<0.05) and predator abundance (P<0.05). Habitats with submerged vegetation were negatively associated with An. funestus group larvae (P<0.05). Only dissolved oxygen (DO) was positively and significantly affect the abundance of An. funestus group larvae (P<0.05). While predators' abundance was not impacted by all physicochemical parameters. CONCLUSION: Six potential predator families were common in aquatic habitats of An. funestus group larvae. Additional studies are needed to demonstrate the efficacy of different predators on larval density and adult fitness traits. Interventions leveraging the interaction between mosquitoes and predators can be established to disrupt the transmission potential and survival of the An. funestus mosquitoes.

The needs and opportunities for housing improvement for malaria control in southern Tanzania.

BACKGROUND: Malaria disproportionately affects low-income households in rural communities where poor housing is common. Despite evidence that well-constructed and mosquito-proofed houses can reduce malaria risk, housing improvement is rarely included in malaria control toolboxes. This study assessed the need, magnitude, and opportunities for housing improvement to control malaria in rural Tanzania. METHODS: A mixed-methods study was conducted in 19 villages across four district councils in southern Tanzania. A structured survey was administered to 1292 community members to assess need, perceptions, and opportunities for housing improvement for malaria control. Direct observations of 802 houses and surrounding environments were done to identify the actual needs and opportunities, and to validate the survey findings. A market survey was done to assess availability and cost of resources and services necessary for mosquito-proofing homes. Focus group discussions were conducted with key stakeholders to explore insights on the potential and challenges of housing improvement as a malaria intervention. RESULTS: Compared to other methods for malaria control, housing improvement was among the best understood and most preferred by community members. Of the 735 survey respondents who needed housing improvements, a majority needed window screening (91.1%), repairs of holes in walls (79.4%), door covers (41.6%), closing of eave spaces (31.2%) and better roofs (19.0%). Community members invested significant efforts to improve their own homes against malaria and other dangers, but these efforts were often slow and delayed due to high costs and limited household incomes. Study participants suggested several mechanisms of support to improve their homes, including government loans and subsidies. CONCLUSION: Addressing the need for housing improvement is a critical component of malaria control efforts in southern Tanzania. In this study, a majority of the community members surveyed needed modest modifications and had plans to work on those modifications. Without additional support, their efforts were however generally slow; households would take years to sufficiently mosquito-proof their houses. It is, therefore, crucial to bring together the key players across sectors to reduce barriers in malaria-proofing housing in endemic settings. These may include government subsidies or partnerships with businesses to make housing improvement more accessible and affordable to residents.

Mass Trapping and Larval Source Management for Mosquito Elimination on Small Maldivian Islands.

Globally, environmental impacts and insecticide resistance are forcing pest control organizations to adopt eco-friendly and insecticide-free alternatives to reduce the risk of mosquito-borne diseases, which affect millions of people, such as dengue, chikungunya or Zika virus. We used, for the first time, a combination of human odor-baited mosquito traps (at 6.0 traps/ha), oviposition traps (7.2 traps/ha) and larval source management (LSM) to practically eliminate populations of the Asian tiger mosquito Aedes albopictus (peak suppression 93.0% (95% CI 91.7-94.4)) and the Southern house mosquito Culex quinquefasciatus (peak suppression 98.3% (95% CI 97.0-99.5)) from a Maldivian island (size: 41.4 ha) within a year and thereafter observed a similar collapse of populations on a second island (size 49.0 ha; trap densities 4.1/ha and 8.2/ha for both trap types, respectively). On a third island (1.6 ha in size), we increased the human odor-baited trap density to 6.3/ha and then to 18.8/ha (combined with LSM but without oviposition traps), after which the Aedes mosquito population was eliminated within 2 months. Such suppression levels eliminate the risk of arboviral disease transmission for local communities and safeguard tourism, a vital economic resource for small island developing states. Terminating intense insecticide use (through fogging) benefits human and environmental health and restores insect biodiversity, coral reefs and marine life in these small and fragile island ecosystems. Moreover, trapping poses a convincing alternative to chemical control and reaches impact levels comparable to contemporary genetic control strategies. This can benefit numerous communities and provide livelihood options in small tropical islands around the world where mosquitoes pose both a nuisance and disease threat.

Using ecological observations to improve malaria control in areas where Anopheles funestus is the dominant vector.

The most important malaria vectors in sub-Saharan Africa are Anopheles gambiae, Anopheles arabiensis, Anopheles funestus, and Anopheles coluzzii. Of these, An. funestus presently dominates in many settings in east and southern Africa. While research on this vector species has been impeded by difficulties in creating laboratory colonies, available evidence suggests it has certain ecological vulnerabilities that could be strategically exploited to greatly reduce malaria transmission in areas where it dominates. This paper examines the major life-history traits of An. funestus, its aquatic and adult ecologies, and its responsiveness to key interventions. It then outlines a plausible strategy for reducing malaria transmission by the vector and sustaining the gains over the medium to long term. To illustrate the propositions, the article uses data from south-eastern Tanzania where An. funestus mediates over 85% of malaria transmission events and is highly resistant to key public health insecticides, notably pyrethroids. Both male and female An. funestus rest indoors and the females frequently feed on humans indoors, although moderate to high degrees of zoophagy can occur in areas with large livestock populations. There are also a few reports of outdoor-biting by the species, highlighting a broader range of behavioural phenotypes that can be considered when designing new interventions to improve vector control. In comparison to other African malaria vectors, An. funestus distinctively prefers permanent and semi-permanent aquatic habitats, including river streams, ponds, swamps, and spring-fed pools. The species is therefore well-adapted to sustain its populations even during dry months and can support year-round malaria transmission. These ecological features suggest that highly effective control of An. funestus could be achieved primarily through strategic combinations of species-targeted larval source management and high quality insecticide-based methods targeting adult mosquitoes in shelters. If done consistently, such an integrated strategy has the potential to drastically reduce local populations of An. funestus and significantly reduce malaria transmission in areas where this vector species dominates. To sustain the gains, the programmes should be complemented with gradual environmental improvements such as house modification to maintain biting exposure at a bare minimum, as well as continuous engagements of the resident communities and other stakeholders.

Effects of agricultural pesticides on the susceptibility and fitness of malaria vectors in rural south-eastern Tanzania.

BACKGROUND: Agricultural pesticides may exert strong selection pressures on malaria vectors during the aquatic life stages and may contribute to resistance in adult mosquitoes. This could reduce the performance of key vector control interventions such as indoor-residual spraying and insecticide-treated nets. The aim of this study was to investigate effects of agrochemicals on susceptibility and fitness of the malaria vectors across farming areas in Tanzania. METHODS: An exploratory mixed-methods study was conducted to assess pesticide use in four villages (V1-V4) in south-eastern Tanzania. Anopheles gambiae (s.l.) larvae were collected from agricultural fields in the same villages and their emergent adults examined for insecticide susceptibility, egg-laying and wing lengths (as proxy for body size). These tests were repeated using two groups of laboratory-reared An. arabiensis, one of which was pre-exposed for 48 h to sub-lethal aquatic doses of agricultural pesticides found in the villages. RESULTS: Farmers lacked awareness about the linkages between the public health and agriculture sectors but were interested in being more informed. Agrochemical usage was reported as extensive in V1, V2 and V3 but minimal in V4. Similarly, mosquitoes from V1 to V3 but not V4 were resistant to pyrethroids and either pirimiphos-methyl or bendiocarb, or both. Adding the synergist piperonyl butoxide restored potency of the pyrethroids. Pre-exposure of laboratory-reared mosquitoes to pesticides during aquatic stages did not affect insecticide susceptibility in emergent adults of the same filial generation. There was also no effect on fecundity, except after pre-exposure to organophosphates, which were associated with fewer eggs and smaller mosquitoes. Wild mosquitoes were smaller than laboratory-reared ones, but fecundity was similar. CONCLUSIONS: Safeguarding the potential of insecticide-based interventions requires improved understanding of how agricultural pesticides influence important life cycle processes and transmission potential of mosquito vectors. In this study, susceptibility of mosquitoes to public health insecticides was lower in villages reporting frequent use of pesticides compared to villages with little or no pesticide use. Variations in the fitness parameters, fecundity and wing length marginally reflected the differences in exposure to agrochemicals and should be investigated further. Pesticide use may exert additional life cycle constraints on mosquito vectors, but this likely occurs after multi-generational exposures.

Wild populations of malaria vectors can mate both inside and outside human dwellings.

BACKGROUND: Wild populations of Anopheles mosquitoes are generally thought to mate outdoors in swarms, although once colonized, they also mate readily inside laboratory cages. This study investigated whether the malaria vectors Anopheles funestus and Anopheles arabiensis can also naturally mate inside human dwellings. METHOD: Mosquitoes were sampled from three volunteer-occupied experimental huts in a rural Tanzanian village at 6:00 p.m. each evening, after which the huts were completely sealed and sampling was repeated at 11:00 p.m and 6 a.m. the next morning to compare the proportions of inseminated females. Similarly timed collections were done inside local unsealed village houses. Lastly, wild-caught larvae and pupae were introduced inside or outside experimental huts constructed inside two semi-field screened chambers. The huts were then sealed and fitted with exit traps, allowing mosquito egress but not entry. Mating was assessed in subsequent days by sampling and dissecting emergent adults caught indoors, outdoors and in exit traps. RESULTS: Proportions of inseminated females inside the experimental huts in the village increased from approximately 60% at 6 p.m. to approximately 90% the following morning despite no new mosquitoes entering the huts after 6 p.m. Insemination in the local homes increased from approximately 78% to approximately 93% over the same time points. In the semi-field observations of wild-caught captive mosquitoes, the proportions of inseminated An. funestus were 20.9% (95% confidence interval [CI]: ± 2.8) outdoors, 25.2% (95% CI: ± 3.4) indoors and 16.8% (± 8.3) in exit traps, while the proportions of inseminated An. arabiensis were 42.3% (95% CI: ± 5.5) outdoors, 47.4% (95% CI: ± 4.7) indoors and 37.1% (CI: ± 6.8) in exit traps. CONCLUSION: Wild populations of An. funestus and An. arabiensis in these study villages can mate both inside and outside human dwellings. Most of the mating clearly happens before the mosquitoes enter houses, but additional mating happens indoors. The ecological significance of such indoor mating remains to be determined. The observed insemination inside the experimental huts fitted with exit traps and in the unsealed village houses suggests that the indoor mating happens voluntarily even under unrestricted egress. These findings may inspire improved vector control, such as by targeting males indoors, and potentially inform alternative methods for colonizing strongly eurygamic Anopheles species (e.g. An. funestus) inside laboratories or semi-field chambers.

Habitat characteristics and insecticide susceptibility of Aedes aegypti in the Ifakara area, south-eastern Tanzania.

BACKGROUND: Aedes-borne diseases such as dengue and chikungunya constitute constant threats globally. In Tanzania, these diseases are transmitted by Aedes aegypti, which is widely distributed in urban areas, but whose ecology remains poorly understood in small towns and rural settings. METHODS: A survey of Ae. aegypti aquatic habitats was conducted in and around Ifakara, a fast-growing town in south-eastern Tanzania. The study area was divided into 200 × 200 m search grids, and habitats containing immature Aedes were characterized. Field-collected Ae. aegypti were tested for susceptibility to common public health insecticides (deltamethrin, permethrin, bendiocarb and pirimiphos-methyl) in the dry and rainy seasons. RESULTS: Of 1515 and 1933 aquatic habitats examined in the dry and rainy seasons, 286 and 283 contained Aedes immatures, respectively (container index, CI: 18.9-14.6%). In the 2315 and 2832 houses visited in the dry and rainy seasons, 114 and 186 houses had at least one Aedes-positive habitat, respectively (house index, HI: 4.9-6.6%). The main habitat types included: (i) used vehicle tires and discarded containers; (ii) flowerpots and clay pots; and (iii) holes made by residents on trunks of coconut trees when harvesting the coconuts. Used tires had highest overall abundance of Ae. aegypti immatures, while coconut tree-holes had highest densities per habitat. Aedes aegypti adults were susceptible to all tested insecticides in both seasons, except bendiocarb, against which resistance was observed in the rainy season. CONCLUSIONS: To our knowledge, this is the first study on ecology and insecticide susceptibility of Ae. aegypti in Ifakara area, and will provide a basis for future studies on its pathogen transmission activities and control. The high infestation levels observed indicate significant risk of Aedes-borne diseases, requiring immediate action to prevent potential outbreaks in the area. While used tires, discarded containers and flowerpots are key habitats for Ae. aegypti, this study also identified coconut harvesting as an important risk factor, and the associated tree-holes as potential targets for Aedes control. Since Ae. aegypti mosquitoes in the area are still susceptible to most insecticides, effective control could be achieved by combining environmental management, preferably involving communities, habitat removal and insecticide spraying.

Preferred resting surfaces of dominant malaria vectors inside different house types in rural south-eastern Tanzania.

BACKGROUND: Malaria control in Africa relies extensively on indoor residual spraying (IRS) and insecticide-treated nets (ITNs). IRS typically targets mosquitoes resting on walls, and in few cases, roofs and ceilings, using contact insecticides. Unfortunately, little attention is paid to where malaria vectors actually rest indoors, and how such knowledge could be used to improve IRS. This study investigated preferred resting surfaces of two major malaria vectors, Anopheles funestus and Anopheles arabiensis, inside four common house types in rural south-eastern Tanzania. METHODS: The assessment was done inside 80 houses including: 20 with thatched roofs and mud walls, 20 with thatched roofs and un-plastered brick walls, 20 with metal roofs and un-plastered brick walls, and 20 with metal roofs and plastered brick walls, across four villages. In each house, resting mosquitoes were sampled in mornings (6 a.m.-8 a.m.), evenings (6 p.m.-8 p.m.) and at night (11 p.m.-12.00 a.m.) using Prokopack aspirators from multiple surfaces (walls, undersides of roofs, floors, furniture, utensils, clothing, curtains and bed nets). RESULTS: Overall, only 26% of An. funestus and 18% of An. arabiensis were found on walls. In grass-thatched houses, 33-55% of An. funestus and 43-50% of An. arabiensis rested under roofs, while in metal-roofed houses, only 16-20% of An. funestus and 8-30% of An. arabiensis rested under roofs. Considering all data together, approximately 40% of mosquitoes rested on surfaces not typically targeted by IRS, i.e. floors, furniture, utensils, clothing and bed nets. These proportions were particularly high in metal-roofed houses (47-53% of An. funestus; 60-66% of An. arabiensis). CONCLUSION: While IRS typically uses contact insecticides to target adult mosquitoes on walls, and occasionally roofs and ceilings, significant proportions of vectors rest on surfaces not usually sprayed. This gap exceeds one-third of malaria mosquitoes in grass-thatched houses, and can reach two-thirds in metal-roofed houses. Where field operations exclude roofs during IRS, the gaps can be much greater. In conclusion, there is need for locally-obtained data on mosquito resting behaviours and how these influence the overall impact and costs of IRS. This study also emphasizes the need for alternative approaches, e.g. house screening, which broadly tackle mosquitoes beyond areas reachable by IRS and ITNs.