BackgroundThe suppression of indoor malaria transmission requires additional interventions that complement the use of insecticide treated nets (ITNs) and indoor residual spraying (IRS). Previous studies have examined the impact of house structure on malaria transmission in areas of low transmission. This study was conducted in a high transmission setting and presents further evidence about the association between specific house characteristics and the abundance of endophilic malaria vectors.MethodsMosquitoes were sampled using CDC light traps from 72 randomly selected houses in two villages on a monthly basis from 2008 to 2011 in rural Southern Tanzania. Generalized linear models using Poisson distributions were used to analyze the association of house characteristics (eave gaps, wall types, roof types, number of windows, rooms and doors, window screens, house size), number of occupants and ITN usage with mean catches of malaria vectors (An.gambiae s.l. and An. funestus).ResultsA total of 36490 female An. gambiae s.l. were collected in Namwawala village and 21266 in Idete village. As for An. funestus females, 2268 were collected in Namwawala and 3398 in Idete. Individually, each house factor had a statistically significant impact (p < 0.05) on the mean catches for An. gambiae s.l. but not An. funestus. A multivariate analysis indicated that the combined absence or presence of eaves, treated or untreated bed-nets, the number of house occupants, house size, netting over windows, and roof type were significantly related (p < 0.05) to An.gambiae s.l. and An. funestus house entry in both villages.ConclusionsDespite significant reductions in vector density and malaria transmission caused by high coverage of ITNs, high numbers of host-seeking malaria vectors are still found indoors due to house designs that favour mosquito entry. In addition to ITNs and IRS, significant efforts should focus on improving house design to prevent mosquito entry and eliminate indoor malaria transmission.

BackgroundMalaria vector control in Tanzania is based on use of long-lasting insecticide treated nets (LLINs) and indoor residual spraying (IRS), which both rely on the use of chemical insecticides. The effectiveness of these control tools is endangered by the development of insecticide resistance in the major malaria vectors. This study was carried out to monitor the susceptibility status of major malaria vectors to insecticides used for IRS and LLINs in mainland Tanzania.MethodsMosquito larvae were collected in 20 sites of Tanzania mainland in 2015. Phenotypic resistance was determined using standard WHO susceptibility tests. Molecular assay were used to determine distribution of Anopheles gambiae sub-species. A microplate assay approach was used for identifying enzyme levels on single mosquitoes from each sites compared with a susceptible reference strain, An. gambiae sensu stricto (s.s.) Kisumu strain.Results Anopheles arabiensis was the dominant malaria specie in the country, accounting for 52% of the sibling species identified, while An. gambiae s.s. represented 48%. In Arumeru site, the dominant species was An. arabiensis, which was resistant to both pyrethroids (permethrin and deltamethrin), and pirimiphos-methyl, and had significant elevated levels of GSTs, non-specific esterases, and oxidase enzymes. An. arabiensis was also a dominant species in Kilombero and Kondoa sites, both were resistant to permethrin and deltamethrin with significant activity levels of oxidase enzymes. Resistance to bendiocarb was recorded in Ngara site where specie composition is evenly distributed between An. gambiae s.s. and An.arabiensis. Also bendiocarb resistance was recorded in Mbozi site, where An. gambiae s.s. is the dominant species.ConclusionsOverall, this study confirmed resistance to all four insecticide classes in An. gambiae sensu lato in selected locations in Tanzania. Results are discussed in relation to resistance mechanisms and the optimization of resistance management strategies.

BackgroundLong-Lasting Insecticidal Nets (LLINs) are one of the major malaria vector control tools, with most countries adopting free or subsidised universal coverage campaigns of populations at-risk from malaria. It is essential to understand LLIN durability so that public health policy makers can select the most cost effective nets that last for the longest time, and estimate the optimal timing of repeated distribution campaigns. However, there is limited knowledge from few countries of the durability of LLINs under user conditions.Methods/DesignThis study investigates LLIN durability in eight districts of Tanzania, selected for their demographic, geographic and ecological representativeness of the country as a whole. We use a two-stage approach: First, LLINs from recent national net campaigns will be evaluated retrospectively in 3,420 households. Those households will receive one of three leading LLIN products at random (Olyset®, PermaNet®2.0 or Netprotect®) and will be followed up for three years in a prospective study to compare their performance under user conditions. LLIN durability will be evaluated by measuring Attrition (the rate at which nets are discarded by households), Bioefficacy (the insecticidal efficacy of the nets measured by knock-down and mortality of mosquitoes), Chemical content (g/kg of insecticide available in net fibres) and physical Degradation (size and location of holes). In addition, we will extend the current national mosquito insecticide Resistance monitoring program to additional districts and use these data sets to provide GIS maps for use in health surveillance and decision making by the National Malaria Control Program (NMCP).DiscussionThe data will be of importance to policy makers and vector control specialists both in Tanzania and the SSA region to inform best practice for the maintenance of high and cost-effective coverage and to maximise current health gains in malaria control.Electronic supplementary materialThe online version of this article (doi:10.1186/1471-2458-14-1266) contains supplementary material, which is available to authorized users.

BackgroundLong-lasting insecticidal nets (LLINs) are the first line choice for malaria vector control in sub-Saharan Africa, with most countries adopting universal coverage campaigns. However, there is only limited information on LLIN durability under user conditions. Therefore, this study aimed to assess the durability of Olyset® LLINs distributed during campaigns between 2009 and 2011 in Tanzania.MethodsA retrospective field survey was conducted in eight districts in Tanzania mainland to assess the durability of Olyset campaign nets. Household questionnaires were used to assess attrition, i.e. net loss. All nets remaining in households were collected. A sub-sample of 198 Olyset campaign nets was examined for bio-efficacy against Anopheles gambiae s.s. mosquitoes, permethrin content and physical integrity following standard World Health Organization (WHO) methods.ResultsOf 6067 campaign nets reported to have been received between 2009 and 2011, 35 % (2145 nets) were no longer present. Most of those nets had been discarded (84 %) mainly because they were too torn (94 %). Of the 198 sub-sampled Olyset LLINs, 61 % were still in serviceable physical condition sufficient to provide personal protection while 39 % were in unserviceable physical condition according to WHO proportionate Hole Index (pHI). More than 96 % (116/120) of nets in serviceable condition passed WHO bioefficacy criteria while all nets in unserviceable condition passed WHO bioefficacy criteria. Overall mean permethrin content was 16.5 g/kg (95 % CI 16.2–16.9) with 78 % of the sub-sampled nets retaining recommended permethrin content regardless of their age or physical condition. Nets aged 4 years and above had a mean permethrin content of 14 g/kg (95 % CI 12.0–16.0). The only statistically significant predictor of reduced physical net integrity was rats in the house.ConclusionsTwo-to-four years after a mass campaign, only 39 % of distributed nets remain both present and in serviceable physical condition, a functional survival considerably below WHO assumptions of 50 % survival of a ‘three-year’ net. However, the majority of nets still retained substantial levels of permethrin and could still be bio-chemically useful against mosquitoes if their holes were repaired, adding evidence to the value of net care and repair campaigns.

BackgroundLarval source management strategies can play an important role in malaria elimination programmes, especially for tackling outdoor biting species and for eliminating parasite and vector populations when they are most vulnerable during the dry season. Effective larval source management requires tools for identifying geographic foci of vector proliferation and malaria transmission where these efforts may be concentrated. Previous studies have relied on surface topographic wetness to indicate hydrological potential for vector breeding sites, but this is unsuitable for karst (limestone) landscapes such as Zanzibar where water flow, especially in the dry season, is subterranean and not controlled by surface topography.MethodsWe examine the relationship between dry and wet season spatial patterns of diagnostic positivity rates of malaria infection amongst patients reporting to health facilities on Unguja, Zanzibar, with the physical geography of the island, including land cover, elevation, slope angle, hydrology, geology and geomorphology in order to identify transmission hot spots using Boosted Regression Trees (BRT) analysis.ResultsThe distribution of both wet and dry season malaria infection rates can be predicted using freely available static data, such as elevation and geology. Specifically, high infection rates in the central and southeast regions of the island coincide with outcrops of hard dense limestone which cause locally elevated water tables and the location of dolines (shallow depressions plugged with fine-grained material promoting the persistence of shallow water bodies).ConclusionsThis analysis provides a tractable tool for the identification of malaria hotspots which incorporates subterranean hydrology, which can be used to target larval source management strategies.

BackgroundMalaria transmission, primarily mediated by Anopheles gambiae, persists in Dar es Salaam (DSM) despite high coverage with bed nets, mosquito-proofed housing and larviciding. New or improved vector control strategies are required to eliminate malaria from DSM, but these will only succeed if they are delivered to the minority of locations where residual transmission actually persists. Hotspots of spatially clustered locations with elevated malaria infection prevalence or vector densities were, therefore, mapped across the city in an attempt to provide a basis for targeting supplementary interventions.MethodsTwo phases of a city-wide population-weighted random sample of cross-sectional household surveys of malaria infections were complemented by two matching phases of geographically overlapping, high-resolution, longitudinal vector density surveys; spanning 2010–2013. Spatial autocorrelations were explored using Moran’s I and hotspots were detected using flexible spatial scan statistics.ResultsSeven hotspots of spatially clustered elevated vector density and eight of malaria infection prevalence were detected over both phases. Only a third of vectors were collected in hotspots in phase 1 (30 %) and phase 2 (33 %). Malaria prevalence hotspots accounted for only half of malaria infections detected in phase 1 (55 %) and phase 2 (47 %). Three quarters (76 % in phase 1 and 74 % in phase 2) of survey locations with detectable vector populations were outside of hotspots. Similarly, more than half of locations with higher infection prevalence (>10 %) occurred outside of hotspots (51 % in phase 1 and 54 % in phase 2). Vector proliferation hazard (exposure to An. gambiae) and malaria infection risk were only very loosely associated with each other (Odds ratio (OR) [95 % Confidence Interval (CI)] = 1.56 [0.89, 1.78], P = 0.52)).ConclusionMany small, scattered loci of local malaria transmission were haphazardly scattered across the city, so interventions targeting only currently identifiable spatially aggregated hotspots will have limited impact. Routine, spatially comprehensive, longitudinal entomological and parasitological surveillance systems, with sufficient sensitivity and spatial resolution to detect these scattered loci, are required to eliminate transmission from this typical African city. Intervention packages targeted to both loci and hotspots of transmission will need to suppress local vector proliferation, treat infected residents and provide vulnerable residents with supplementary protective measures against exposure.

This study investigated whether passively collected routine health facility data can be used for mapping spatial heterogeneities in malaria transmission at the level of local government housing cluster administrative units in Dar es Salaam, Tanzania. From June 2012 to January 2013, residential locations of patients tested for malaria at a public health facility were traced based on their local leaders' names and geo-referencing the point locations of these leaders' houses. Geographic information systems (GIS) were used to visualise the spatial distribution of malaria infection rates. Spatial scan statistics was deployed to detect spatial clustering of high infection rates. Among 2407 patients tested for malaria, 46.6% (1121) could be traced to their 411 different residential housing clusters. One small spatially aggregated cluster of neighbourhoods with high prevalence was identified. While the home residence housing cluster leader was unambiguously identified for 73.8% (240/325) of malaria-positive patients, only 42.3% (881/2082) of those with negative test results were successfully traced. It was concluded that recording simple points of reference during routine health facility visits can be used for mapping malaria infection burden on very fine geographic scales, potentially offering a feasible approach to rational geographic targeting of malaria control interventions. However, in order to tap the full potential of this approach, it would be necessary to optimise patient tracing success and eliminate biases by blinding personnel to test results.