Terra and Aqua: new data for epidemiology and public health

The ecology and epidemiology of malaria parasitism in wild chimpanzee reservoirs

Chimpanzees (Pan troglodytes) harbor rich assemblages of malaria parasites, including three species closely related to P. falciparum (sub-genus Laverania), the most malignant human malaria parasite. Here, we characterize the ecology and epidemiology of malaria infection in wild chimpanzee reservoirs. We used molecular assays to screen chimpanzee fecal samples, collected longitudinally and cross-sectionally from wild populations, for malaria parasite mitochondrial DNA. We found that chimpanzee malaria parasitism has an early age of onset and varies seasonally in prevalence. A subset of samples revealed Hepatocystis mitochondrial DNA, with phylogenetic analyses suggesting that Hepatocystis appears to cross species barriers more easily than Laverania. Longitudinal and cross-sectional sampling independently support the hypothesis that mean ambient temperature drives spatiotemporal variation in chimpanzee Laverania infection. Infection probability peaked at ~24.5 °C, consistent with the empirical transmission optimum of P. falciparum in humans. Forest cover was also positively correlated with spatial variation in Laverania prevalence, consistent with the observation that forest-dwelling Anophelines are the primary vectors. Extrapolating these relationships across equatorial Africa, we map spatiotemporal variation in the suitability of chimpanzee habitat for Laverania transmission, offering a hypothetical baseline indicator of human exposure risk.

An ecological niche model to predict the geographic distribution of Haemagogus janthinomys, Dyar, 1921 a yellow fever and Mayaro virus vector, in South America

Yellow fever virus (YFV) has a long history of impacting human health in South America. Mayaro virus (MAYV) is an emerging arbovirus of public health concern in the Neotropics and its full impact is yet unknown. Both YFV and MAYV are primarily maintained via a sylvatic transmission cycle but can be opportunistically transmitted to humans by the bites of infected forest dwelling Haemagogus janthinomys Dyar, 1921. To better understand the potential risk of YFV and MAYV transmission to humans, a more detailed understanding of this vector species’ distribution is critical. This study compiled a comprehensive database of 177 unique Hg. janthinomys collection sites retrieved from the published literature, digitized museum specimens and publicly accessible mosquito surveillance data. Covariate analysis was performed to optimize a selection of environmental (topographic and bioclimatic) variables associated with predicting habitat suitability, and species distributions modelled across South America using a maximum entropy (MaxEnt) approach. Our results indicate that suitable habitat for Hg. janthinomys can be found across forested regions of South America including the Atlantic forests and interior Amazon.

Mayaro virus is a neglected tropical disease and there is insufficient evidence to define its geographic range. The mosquito Haemagogus janthinomys is a primary vector of Mayaro and its distribution is largely unknown at a sub-country scale. Building compendiums of collection data and creating ecological niche models provides a more precise estimation of vector species potential habitat. Our dataset stands as one of the most expansive existing for collection data of this species combining data published in literature, publicly available data repositories and digitized museum specimen records. Comparing results of niche models with near real time environmental data can give even better predictions of areas where Mayaro virus exposure could occur. The methods and results of this study can be replicated for any disease/vector of interest so long as there is data discoverable through the scientific literature, public repositories, or other civilian and governmental agencies willing to share.

A spatiotemporal reconstruction of daily ambient temperature using satellite data in the Megalopolis of Central Mexico from 2003 to 2019

While weather stations generally capture near-surface ambient air temperature (Ta) at a high temporal resolution to calculate daily values (i.e., daily minimum, mean, and maximum Ta), their fixed locations can limit their spatial coverage and resolution even in densely populated urban areas. As a result, data from weather stations alone may be inadequate for Ta-related epidemiology particularly when the stations are not located in the areas of interest for human exposure assessment. To address this limitation in the Megalopolis of Central Mexico (MCM), we developed the first spatiotemporally resolved hybrid satellite-based land use regression Ta model for the region, home to nearly 30 million people and includes Mexico City and seven more metropolitan areas. Our model predicted daily minimum, mean, and maximum Ta for the years 2003-2019. We used data from 120 weather stations and Land Surface Temperature (LST) data from NASA's MODIS instruments on the Aqua and Terra satellites on a 1 × 1 km grid. We generated a satellite-hybrid mixed-effects model for each year, regressing Ta measurements against land use terms, day-specific random intercepts, and fixed and random LST slopes. We assessed model performance using 10-fold cross-validation at withheld stations. Across all years, the root-mean-square error ranged from 0.92 to 1.92 K and the R 2 ranged from .78 to .95. To demonstrate the utility of our model for health research, we evaluated the total number of days in the year 2010 when residents ≥65 years old were exposed to Ta extremes (above 30°C or below 5°C). Our model provides much needed high-quality Ta estimates for epidemiology studies in the MCM region.

Hybrid modeling and prediction of oyster norovirus outbreaks

This paper presents a hybrid model for predicting oyster norovirus outbreaks by combining the Artificial Neural Networks (ANNs) and Principal Component Analysis (PCA) methods and using the Moderate Resolution Imaging Spectroradiometer (MODIS) satellite remote-sensing data. Specifically, 10 years (2007-2016) of cloud-free MODIS Aqua data for water leaving reflectance and environmental data were extracted from the center of each oyster harvest area. Then, the PCA was utilized to compress the size of the MODIS Aqua data. An ANN model was trained using the first 4 years of the data from 2007 to 2010 and validated using the additional 6 years of independent datasets collected from 2011 to 2016. Results indicated that the hybrid PCA-ANN model was capable of reproducing the 10 years of historical oyster norovirus outbreaks along the Northern Gulf of Mexico coast with a sensitivity of 72.7% and specificity of 99.9%, respectively, demonstrating the efficacy of the hybrid model.

A holistic model to assess risk factors of fasciolosis in Ankole cattle

In recent decades, remote sensing (RS) technology and geographical information systems (GIS) were increasingly used as tools for epidemiological studies and the control of zoonotic diseases. Fasciolosis, a zoonotic disease caused by a trematode parasite (Fasciola spp.), is a good candidate for the application of RS and GIS in epidemiology because it is strongly influenced by the environment, i.e. the habitat of the intermediate host. In this study, we examined variables which may increase the fasciolosis risk of Ankole cattle in the degraded and overgrazed Mutara rangelands of north-eastern Rwanda. The risk variables considered included three environmental variables (normalized difference vegetation index, NDVI; normalized difference moisture index, NDMI; normalized difference water index, NDWI), two landscape metric variables (rangeland proportion, building density), two geological variables (poorly-drained soil proportion, elevation) and three animal husbandry variables (herd size, adult proportion and the body condition score). Fasciola spp. prevalence was used as the dependent variable, sampling season as a fixed factor and four principal components (PCs, condensed from the ten risk variables) as covariates in a univariate General Linear Model. Fasciola spp. prevalence was positively correlated to rangeland proportion, cattle herd size in rural areas, adult proportion and individual body condition. Moreover, high Fasciola spp. prevalence was found in densely vegetated areas with high moisture (high values of NDVI and NDMI), in combination with large proportions of poorly-drained soil at low elevations. Future investigations should focus on increased sampling across the Mutara rangelands to prepare a predictive, spatial fasciolosis risk map that would help to further improve sustainable land-use management.

Driving forces of land surface temperature anomalous changes in North America in 2002-2018

The land surface temperature (LST) changes in North America are very abnormal recently, but few studies have systematically researched these anomalies from several aspects, especially the influencing forces. After reconstructing higher quality MODIS monthly LST data (0.05° * 0.05°) in 2002-2018, we analyzed the LST changes especially anomalous changes and their driving forces in North America. Here we show that North America warmed at the rate of 0.02 °C/y. The LST changes in three regions, including frigid region in the northwestern (0.12 °C/y), the west coast from 20°N-40°N (0.07 °C/y), and the tropics south of 20°N (0.04 °C/y), were extremely abnormal. The El Nino and La Nina were the main drivers for the periodical highest and lowest LST, respectively. The North Atlantic Oscillation was closed related to the opposite change of LST in the northeastern North America and the southeastern United States, and the warming trend of the Florida peninsula in winter was closely related to enhancement of the North Atlantic Oscillation index. The Pacific Decadal Oscillation index showed a positive correlation with the LST in most Alaska. Vegetation and atmospheric water vapor also had a profound influence on the LST changes, but it had obvious difference in latitude.

Environmental Reservoirs of Vibrio cholerae: Challenges and Opportunities for Ocean-Color Remote Sensing

The World Health Organization has estimated the burden of the on-going pandemic of cholera at 1.3 to 4 million cases per year worldwide in 2016, and a doubling of case-fatality-rate to 1.8% in 2016 from 0.8% in 2015. The disease cholera is caused by the bacterium Vibrio cholerae that can be found in environmental reservoirs, living either in free planktonic form or in association with host organisms, non-living particulate matter or in the sediment, and participating in various biogeochemical cycles. An increasing number of epidemiological studies are using land- and water-based remote-sensing observations for monitoring, surveillance, or risk mapping of Vibrio pathogens and cholera outbreaks. Although the Vibrio pathogens cannot be sensed directly by satellite sensors, remotely-sensed data can be used to infer their presence. Here, we review the use of ocean-color remote-sensing data, in conjunction with information on the ecology of the pathogen, to map its distribution and forecast risk of disease occurrence. Finally, we assess how satellite-based information on cholera may help support the Sustainable Development Goals and targets on Health (Goal 3), Water Quality (Goal 6), Climate (Goal 13), and Life Below Water (Goal 14).

Gap-Filling of MODIS Time Series Land Surface Temperature (LST) Products Using Singular Spectrum Analysis (SSA)

Land surface temperature (LST) is a basic parameter in energy exchange between the land and the atmosphere, and is frequently used in many sciences such as climatology, hydrology, agriculture, ecology, etc. Time series of satellite LST data have usually deficient, missing, and unacceptable data caused by the presence of clouds in images, the presence of dust in the atmosphere, and sensor failure. In this study, the singular spectrum analysis (SSA) algorithm was used to resolve the problem of missing and outlier data caused by cloud cover. The region studied in the present research included an image frame of the Moderate Resolution Imaging Spectroradiometer (MODIS) with horizontal number 22 and vertical number 05 (h22v05). This image involved a large part of Iran, Turkmenistan, and the Caspian Sea. In this study, MODIS LST products (MOD11A1) were used during 2015 with approximately 1 km × 1 km spatial resolution and day/night LST data (daily temporal resolution). On average, the data have 36.37% gaps in each pixel profile with 730 day/night LST data. The results of the SSA algorithm in the reconstruction of LST images indicated a root mean square error (RMSE) of 2.95 Kelvin (K) between the original and reconstructed LST time series data in the study region. In general, the findings showed that the SSA algorithm using spatio-temporal interpolation can be effectively used to resolve the problem of missing data caused by cloud cover.

Dynamic malaria hotspots in an open cohort in western Kenya

Malaria hotspots, defined as areas where transmission intensity exceeds the average level, become more pronounced as transmission declines. Targeting hotspots may accelerate reductions in transmission and could be pivotal for malaria elimination. Determinants of hotspot location, particularly of their movement, are poorly understood. We used spatial statistical methods to identify foci of incidence of self-reported malaria in a large census population of 64,000 people, in 8,290 compounds over a 2.5-year study period. Regression models examine stability of hotspots and identify static and dynamic correlates with their location. Hotspot location changed over short time-periods, rarely recurring in the same area. Hotspots identified in spring versus fall season differed in their stability. Households located in a hotspot in the fall were more likely to be located in a hotspot the following fall (RR = 1.77, 95% CI: 1.66-1.89), but the opposite was true for compounds in spring hotspots (RR = 0.15, 95% CI: 0.08-0.28). Location within a hotspot was related to environmental and static household characteristics such as distance to roads or rivers. Human migration into a household was correlated with risk of hotspot membership, but the direction of the association differed based on the origin of the migration event.

Wild bonobos host geographically restricted malaria parasites including a putative new Laverania species

Malaria parasites, though widespread among wild chimpanzees and gorillas, have not been detected in bonobos. Here, we show that wild-living bonobos are endemically Plasmodium infected in the eastern-most part of their range. Testing 1556 faecal samples from 11 field sites, we identify high prevalence Laverania infections in the Tshuapa-Lomami-Lualaba (TL2) area, but not at other locations across the Congo. TL2 bonobos harbour P. gaboni, formerly only found in chimpanzees, as well as a potential new species, Plasmodium lomamiensis sp. nov. Rare co-infections with non-Laverania parasites were also observed. Phylogenetic relationships among Laverania species are consistent with co-divergence with their gorilla, chimpanzee and bonobo hosts, suggesting a timescale for their evolution. The absence of Plasmodium from most field sites could not be explained by parasite seasonality, nor by bonobo population structure, diet or gut microbiota. Thus, the geographic restriction of bonobo Plasmodium reflects still unidentified factors that likely influence parasite transmission.

Analysing trends and forecasting malaria epidemics in Madagascar using a sentinel surveillance network: a web-based application

Background

The use of a malaria early warning system (MEWS) to trigger prompt public health interventions is a key step in adding value to the epidemiological data routinely collected by sentinel surveillance systems.

Methods

This study describes a system using various epidemic thresholds and a forecasting component with the support of new technologies to improve the performance of a sentinel MEWS. Malaria-related data from 21 sentinel sites collected by Short Message Service are automatically analysed to detect malaria trends and malaria outbreak alerts with automated feedback reports.

Results

Roll Back Malaria partners can, through a user-friendly web-based tool, visualize potential outbreaks and generate a forecasting model. The system already demonstrated its ability to detect malaria outbreaks in Madagascar in 2014.

Conclusion

This approach aims to maximize the usefulness of a sentinel surveillance system to predict and detect epidemics in limited-resource environments.

Age, Spatial, and Temporal Variations in Hospital Admissions with Malaria in Kilifi County, Kenya: A 25-Year Longitudinal Observational Study

BACKGROUND

Encouraging progress has been seen with reductions in Plasmodium falciparum malaria transmission in some parts of Africa. Reduced transmission might lead to increasing susceptibility to malaria among older children due to lower acquired immunity, and this has implications for ongoing control strategies.

METHODS AND FINDINGS

We conducted a longitudinal observational study of children admitted to Kilifi County Hospital in Kenya and linked it to data on residence and insecticide-treated net (ITN) use. This included data from 69,104 children aged from 3 mo to 13 y admitted to Kilifi County Hospital between 1 January 1990 and 31 December 2014. The variation in malaria slide positivity among admissions was examined in logistic regression models using the following predictors: location of the residence, calendar time, the child's age, ITN use, and the enhanced vegetation index (a proxy for soil moisture). The proportion of malaria slide-positive admissions declined from 0.56 (95% confidence interval [CI] 0.54-0.58) in 1998 to 0.07 (95% CI 0.06-0.08) in 2009 but then increased again through to 0.24 (95% CI 0.22-0.25) in 2014. Older children accounted for most of the increase after 2009 (0.035 [95% CI 0.030-0.040] among young children compared to 0.22 [95% CI 0.21-0.23] in older children). There was a nonlinear relationship between malaria risk and prevalence of ITN use within a 2 km radius of an admitted child's residence such that the predicted malaria positive fraction varied from ~0.4 to <0.1 as the prevalence of ITN use varied from 20% to 80%. In this observational analysis, we were unable to determine the cause of the decline in malaria between 1998 and 2009, which pre-dated the dramatic scale-up in ITN distribution and use.

CONCLUSION

Following a period of reduced transmission, a cohort of older children emerged who have increased susceptibility to malaria. Further reductions in malaria transmission are needed to mitigate the increasing burden among older children, and universal ITN coverage is a promising strategy to achieve this goal.

Earth Observation, Spatial Data Quality, and Neglected Tropical Diseases

Earth observation (EO) is the use of remote sensing and in situ observations to gather data on the environment. It finds increasing application in the study of environmentally modulated neglected tropical diseases (NTDs). Obtaining and assuring the quality of the relevant spatially and temporally indexed EO data remain challenges. Our objective was to review the Earth observation products currently used in studies of NTD epidemiology and to discuss fundamental issues relating to spatial data quality (SDQ), which limit the utilization of EO and pose challenges for its more effective use. We searched Web of Science and PubMed for studies related to EO and echinococossis, leptospirosis, schistosomiasis, and soil-transmitted helminth infections. Relevant literature was also identified from the bibliographies of those papers. We found that extensive use is made of EO products in the study of NTD epidemiology; however, the quality of these products is usually given little explicit attention. We review key issues in SDQ concerning spatial and temporal scale, uncertainty, and the documentation and use of quality information. We give examples of how these issues may interact with uncertainty in NTD data to affect the output of an epidemiological analysis. We conclude that researchers should give careful attention to SDQ when designing NTD spatial-epidemiological studies. This should be used to inform uncertainty analysis in the epidemiological study. SDQ should be documented and made available to other researchers.

Mapping the zoonotic niche of Ebola virus disease in Africa

Ebola virus disease (EVD) is a complex zoonosis that is highly virulent in humans. The largest recorded outbreak of EVD is ongoing in West Africa, outside of its previously reported and predicted niche. We assembled location data on all recorded zoonotic transmission to humans and Ebola virus infection in bats and primates (1976-2014). Using species distribution models, these occurrence data were paired with environmental covariates to predict a zoonotic transmission niche covering 22 countries across Central and West Africa. Vegetation, elevation, temperature, evapotranspiration, and suspected reservoir bat distributions define this relationship. At-risk areas are inhabited by 22 million people; however, the rarity of human outbreaks emphasises the very low probability of transmission to humans. Increasing population sizes and international connectivity by air since the first detection of EVD in 1976 suggest that the dynamics of human-to-human secondary transmission in contemporary outbreaks will be very different to those of the past.

Air temperature suitability for Plasmodium falciparum malaria transmission in Africa 2000-2012: a high-resolution spatiotemporal prediction

Background

Temperature suitability for malaria transmission is a useful predictor variable for spatial models of malaria infection prevalence. Existing continental or global models, however, are synoptic in nature and so do not characterize inter-annual variability in seasonal patterns of temperature suitability, reducing their utility for predicting malaria risk.

Methods

A malaria Temperature Suitability Index (TSI) was created by first modeling minimum and maximum air temperature with an eight-day temporal resolution from gap-filled MODerate Resolution Imaging Spectroradiometer (MODIS) daytime and night-time Land Surface Temperature (LST) datasets. An improved version of an existing biological model for malaria temperature suitability was then applied to the resulting temperature information for a 13-year data series. The mechanism underlying this biological model is simulation of emergent mosquito cohorts on a two-hour time-step and tracking of each cohort throughout its life to quantify the impact air temperature has on both mosquito survival and sporozoite development.

Results

The results of this research consist of 154 monthly raster surfaces that characterize spatiotemporal patterns in TSI across Africa from April 2000 through December 2012 at a 1 km spatial resolution. Generalized TSI patterns were as expected, with consistently high values in equatorial rain forests, seasonally variable values in tropical savannas (wet and dry) and montane areas, and low values in arid, subtropical regions. Comparisons with synoptic approaches demonstrated the additional information available within the dynamic TSI dataset that is lost in equivalent synoptic products derived from long-term monthly averages.

Conclusions

The dynamic TSI dataset presented here provides a new product with far richer spatial and temporal information than any other presently available for Africa. As spatiotemporal malaria modeling endeavors evolve, dynamic predictor variables such as the malaria temperature suitability data developed here will be essential for the rational assessment of changing patterns of malaria risk.

An environmental data set for vector-borne disease modeling and epidemiology

Understanding the environmental conditions of disease transmission is important in the study of vector-borne diseases. Low- and middle-income countries bear a significant portion of the disease burden; but data about weather conditions in those countries can be sparse and difficult to reconstruct. Here, we describe methods to assemble high-resolution gridded time series data sets of air temperature, relative humidity, land temperature, and rainfall for such areas; and we test these methods on the island of Madagascar. Air temperature and relative humidity were constructed using statistical interpolation of weather station measurements; the resulting median 95^th percentile absolute errors were 2.75°C and 16.6%. Missing pixels from the MODIS11 remote sensing land temperature product were estimated using Fourier decomposition and time-series analysis; thus providing an alternative to the 8-day and 30-day aggregated products. The RFE 2.0 remote sensing rainfall estimator was characterized by comparing it with multiple interpolated rainfall products, and we observed significant differences in temporal and spatial heterogeneity relevant to vector-borne disease modeling.

Prediction of Bluetongue virus seropositivity on pastoral properties in northern Australia using remotely sensed bioclimatic variables

To monitor Bluetongue virus (BTV) activity in northern and eastern Australia the National Arbovirus Monitoring Program (NAMP) collects data from a network of sentinel herds. Groups of young cattle, previously unexposed to infection, are regularly tested to detect evidence of seroconversion. While this approach has been successful in fulfilling international surveillance requirements, it is labour and cost intensive and operationally challenging in the remote area of the northern Australian rangelands. The aim of this study was to assess the suitability of remotely sensed data as a means for predicting the distribution of BTV seroprevalence. For the period 2000-2009, bioclimatic variables were derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) and Tropical Rainfall Measuring Mission (TRMM) data products for the entire Northern Territory. A generalised linear model, based on the seasonal Normalised Difference Vegetation Index (NDVI) and minimum land surface temperature, was developed to predict BTV seropositivity. The odds of seropositivity in locations with NDVI estimates >0.45 was 3.90 (95% CI 1.11 to 13.7) times that of locations where NDVI estimates were between 0 and 0.45. Unit increases in minimum night land surface temperature in the previous winter increased the odds of seropositivity by a factor of 1.40 (95% CI 1.02 to 1.91). The area under a Receiver Operator Characteristic curve generated on the basis of the model predictions was 0.8. Uncertainty in the model's predictions was attributed to the spatio-temporal inconsistency in the precision of the available serosurveillance data. The discriminatory ability of models of this type could be improved by ensuring that exact location details and date of NAMP BTV test events are consistently recorded.

Detection and forecasting of oyster norovirus outbreaks: recent advances and future perspectives

Norovirus is a highly infectious pathogen that is commonly found in oysters growing in fecally contaminated waters. Norovirus outbreaks can cause the closure of oyster harvesting waters and acute gastroenteritis in humans associated with consumption of contaminated raw oysters. Extensive efforts and progresses have been made in detection and forecasting of oyster norovirus outbreaks over the past decades. The main objective of this paper is to provide a literature review of methods and techniques for detecting and forecasting oyster norovirus outbreaks and thereby to identify the future directions for improving the detection and forecasting of norovirus outbreaks. It is found that (1) norovirus outbreaks display strong seasonality with the outbreak peak occurring commonly in December-March in the U.S. and April-May in the Europe; (2) norovirus outbreaks are affected by multiple environmental factors, including but not limited to precipitation, temperature, solar radiation, wind, and salinity; (3) various modeling approaches may be employed to forecast norovirus outbreaks, including Bayesian models, regression models, Artificial Neural Networks, and process-based models; and (4) diverse techniques are available for near real-time detection of norovirus outbreaks, including multiplex PCR, seminested PCR, real-time PCR, quantitative PCR, and satellite remote sensing. The findings are important to the management of oyster growing waters and to future investigations into norovirus outbreaks. It is recommended that a combined approach of sensor-assisted real time monitoring and modeling-based forecasting should be utilized for an efficient and effective detection and forecasting of norovirus outbreaks caused by consumption of contaminated oysters.

Prevalence of the symbiont Cardinium in Culicoides (Diptera: Ceratopogonidae) vector species is associated with land surface temperature

Prevalence of infection by bacterial symbionts may reflect their interactions with the host and has been shown to be correlated with environmental factors. Yet, it is still unclear whether infection by symbionts is determined by environmental factors affecting the early or imago stage of the host. Here, we identified and localized the symbiont Candidatus Cardinium hertigii (Bacteroidetes) in sympatric Culicoides biting midge species, examined its abundance, and studied its association with environmental factors. The prevalence of adult infection differed, with 50.7% from C. imicola, 31.4% from C. oxystoma, and 0% from C. schultzei gp., although phylogenetic analyses showed that Cardinium in these species is almost identical. In addition, prevalence of infection differed between climate regions, with lowest prevalence in the arid region and highest prevalence in the Mediterranean region. Multivariate linear regression analysis of Cardinium prevalence together with climatic and satellite imagery data-derived environmental variables revealed that infection prevalence is significantly associated with land surface temperature and explained up to 89.7% of infection prevalence variability. These findings suggest that the observed variation of Cardinium infection of the imago stage of Culicoides may be influenced by environmental conditions during the latter's early developmental stages.

Remote sensing-based time series models for malaria early warning in the highlands of Ethiopia

BACKGROUND

Malaria is one of the leading public health problems in most of sub-Saharan Africa, particularly in Ethiopia. Almost all demographic groups are at risk of malaria because of seasonal and unstable transmission of the disease. Therefore, there is a need to develop malaria early-warning systems to enhance public health decision making for control and prevention of malaria epidemics. Data from orbiting earth-observing sensors can monitor environmental risk factors that trigger malaria epidemics. Remotely sensed environmental indicators were used to examine the influences of climatic and environmental variability on temporal patterns of malaria cases in the Amhara region of Ethiopia.

METHODS

In this study seasonal autoregressive integrated moving average (SARIMA) models were used to quantify the relationship between malaria cases and remotely sensed environmental variables, including rainfall, land-surface temperature (LST), vegetation indices (NDVI and EVI), and actual evapotranspiration (ETa) with lags ranging from one to three months. Predictions from the best model with environmental variables were compared to the actual observations from the last 12 months of the time series.

RESULTS

Malaria cases exhibited positive associations with LST at a lag of one month and positive associations with indicators of moisture (rainfall, EVI and ETa) at lags from one to three months. SARIMA models that included these environmental covariates had better fits and more accurate predictions, as evidenced by lower AIC and RMSE values, than models without environmental covariates.

CONCLUSIONS

Malaria risk indicators such as satellite-based rainfall estimates, LST, EVI, and ETa exhibited significant lagged associations with malaria cases in the Amhara region and improved model fit and prediction accuracy. These variables can be monitored frequently and extensively across large geographic areas using data from earth-observing sensors to support public health decisions.

Wind direction and proximity to larval sites determines malaria risk in Kilifi District in Kenya

Studies of the fine-scale spatial epidemiology of malaria consistently identify malaria hotspots, comprising clusters of homesteads at high transmission intensity. These hotspots sustain transmission, and may be targeted by malaria-control programmes. Here we describe the spatial relationship between the location of Anopheles larval sites and human malaria infection in a cohort study of 642 children, aged 1-10-years-old. Our data suggest that proximity to larval sites predict human malaria infection, when homesteads are upwind of larval sites, but not when homesteads are downwind of larval sites. We conclude that following oviposition, female Anophelines fly upwind in search for human hosts and, thus, malaria transmission may be disrupted by targeting vector larval sites in close proximity, and downwind to malaria hotspots.

Terra and Aqua satellites track tiger mosquito invasion: modelling the potential distribution of Aedes albopictus in north-eastern Italy

Background

The continuing spread of the Asian tiger mosquito Aedes albopictus in Europe is of increasing public health concern due to the potential risk of new outbreaks of exotic vector-borne diseases that this species can transmit as competent vector. We predicted the most favorable areas for a short term invasion of Ae. albopictus in north-eastern Italy using reconstructed daily satellite data time series (MODIS Land Surface Temperature maps, LST). We reconstructed more than 11,000 daily MODIS LST maps for the period 2001-09 (i.e. performed spatial and temporal gap-filling) in an Open Source GIS framework. We aggregated these LST maps over time and identified the potential distribution areas of Ae. albopictus by adapting published temperature threshold values using three variables as predictors (0°C for mean January temperatures, 11°C for annual mean temperatures and 1350 growing degree days filtered for areas with autumnal mean temperatures > 11°C). The resulting maps were integrated into the final potential distribution map and this was compared with the known current distribution of Ae. albopictus in north-eastern Italy.

Results

LST maps show the microclimatic characteristics peculiar to complex terrains, which would not be visible in maps commonly derived from interpolated meteorological station data. The patterns of the three indicator variables partially differ from each other, while winter temperature is the determining limiting factor for the distribution of Ae. albopictus. All three variables show a similar spatial pattern with some local differences, in particular in the northern part of the study area (upper Adige valley).

Conclusions

Reconstructed daily land surface temperature data from satellites can be used to predict areas of short term invasion of the tiger mosquito with sufficient accuracy (200 m pixel resolution size). Furthermore, they may be applied to other species of arthropod of medical interest for which temperature is a relevant limiting factor. The results indicate that, during the next few years, the tiger mosquito will probably spread toward northern latitudes and higher altitudes in north-eastern Italy, which will considerably expand the range of the current distribution of this species.

Stable and unstable malaria hotspots in longitudinal cohort studies in Kenya

BACKGROUND

Infectious diseases often demonstrate heterogeneity of transmission among host populations. This heterogeneity reduces the efficacy of control strategies, but also implies that focusing control strategies on "hotspots" of transmission could be highly effective.

METHODS AND FINDINGS

In order to identify hotspots of malaria transmission, we analysed longitudinal data on febrile malaria episodes, asymptomatic parasitaemia, and antibody titres over 12 y from 256 homesteads in three study areas in Kilifi District on the Kenyan coast. We examined heterogeneity by homestead, and identified groups of homesteads that formed hotspots using a spatial scan statistic. Two types of statistically significant hotspots were detected; stable hotspots of asymptomatic parasitaemia and unstable hotspots of febrile malaria. The stable hotspots were associated with higher average AMA-1 antibody titres than the unstable clusters (optical density [OD] = 1.24, 95% confidence interval [CI] 1.02-1.47 versus OD = 1.1, 95% CI 0.88-1.33) and lower mean ages of febrile malaria episodes (5.8 y, 95% CI 5.6-6.0 versus 5.91 y, 95% CI 5.7-6.1). A falling gradient of febrile malaria incidence was identified in the penumbrae of both hotspots. Hotspots were associated with AMA-1 titres, but not seroconversion rates. In order to target control measures, homesteads at risk of febrile malaria could be predicted by identifying the 20% of homesteads that experienced an episode of febrile malaria during one month in the dry season. That 20% subsequently experienced 65% of all febrile malaria episodes during the following year. A definition based on remote sensing data was 81% sensitive and 63% specific for the stable hotspots of asymptomatic malaria.

CONCLUSIONS

Hotspots of asymptomatic parasitaemia are stable over time, but hotspots of febrile malaria are unstable. This finding may be because immunity offsets the high rate of febrile malaria that might otherwise result in stable hotspots, whereas unstable hotspots necessarily affect a population with less prior exposure to malaria.

Human helminth co-infection: no evidence of common genetic control of hookworm and Schistosoma mansoni infection intensity in a Brazilian community

Strong statistical associations between soil-transmitted helminths and schistosomes are frequently observed in co-endemic human populations, although the underlying explanations remain poorly understood. This study investigates the contribution of host genetics and domestic environment to hookworm and Schistosoma mansoni infection intensity and evaluates the role of genetic and non-genetic factors in co-variation of infection intensity. Detailed genealogical information allowed assignment of 1303 individuals living in the Brazilian community of Americaninhas, Minas Gerais state, to 25 pedigrees (containing between two and 1159 members) residing in 303 households. The prevalence of co-infection with both hookworms and schistosomes was high (38.5%), with significant correlation between Necator americanus and S. mansoni faecal egg counts. Bivariate variance component analysis demonstrated a modest but significant species-specific heritability for intensity of N. americanus (h(2)=0.196) and S. mansoni infection (h(2)=0.230). However, after accounting for demographic, socio-economic and household risk factors, no evidence for common genetic control of intensity of hookworm and schistosome infection was observed. There was some evidence for residual clustering within households but the majority (63%) of the covariance between N. americanus and S. mansoni infection intensity remained specific to the individual and could not be explained by shared genes, shared environment or other shared demographic, socio-economic or environmental risk factors. Our results emphasize the importance of exposure to hookworm and schistosome infection in driving the association between levels of infection with these species in hosts resident in areas of high transmission and suggest that much of this common exposure occurs outside the home.

Urban structure and dengue fever in Puntarenas, Costa Rica

Dengue is currently the most important arboviral disease globally and is usually associated with built environments in tropical areas. Remotely sensed information can facilitate the study of urban mosquito-borne diseases by providing multiple temporal and spatial resolutions appropriate to investigate urban structure and ecological characteristics associated with infectious disease. In this study, coarse, medium and fine resolution satellite imagery (Moderate Resolution Imaging Spectrometer, Advanced Spaceborne Thermal Emission and Reflection Radiometer and QuickBird respectively) and ground-based data were analyzed for the Greater Puntarenas area, Costa Rica for the years 2002-04. The results showed that the mean normalized difference vegetation index (NDVI) was generally higher in the localities with lower incidence of dengue fever during 2002, although the correlation was statistically significant only in the dry season (r=-0.40; p=0.03). Dengue incidence was inversely correlated to built area and directly correlated with tree cover (r=0.75, p=0.01). Overall, the significant correlations between dengue incidence and urban structural variables (tree cover and building density) suggest that properties of urban structure may be associated with dengue incidence in tropical urban settings.

Human helminth co-infection: analysis of spatial patterns and risk factors in a Brazilian community

BACKGROUND

Individuals living in areas endemic for helminths are commonly infected with multiple species. Despite increasing emphasis given to the potential health impacts of polyparasitism, few studies have investigated the relative importance of household and environmental factors on the risk of helminth co-infection. Here, we present an investigation of exposure-related risk factors as sources of heterogeneity in the distribution of co-infection with Necator americanus and Schistosoma mansoni in a region of southeastern Brazil.

METHODOLOGY

Cross-sectional parasitological and socio-economic data from a community-based household survey were combined with remotely sensed environmental data using a geographical information system. Geo-statistical methods were used to explore patterns of mono- and co-infection with N. americanus and S. mansoni in the region. Bayesian hierarchical models were then developed to identify risk factors for mono- and co-infection in relation to community-based survey data to assess their roles in explaining observed heterogeneity in mono and co-infection with these two helminth species.

PRINCIPAL FINDINGS

The majority of individuals had N. americanus (71.1%) and/or S. mansoni (50.3%) infection; 41.0% of individuals were co-infected with both helminths. Prevalence of co-infection with these two species varied substantially across the study area, and there was strong evidence of household clustering. Hierarchical multinomial models demonstrated that relative socio-economic status, household crowding, living in the eastern watershed and high Normalized Difference Vegetation Index (NDVI) were significantly associated with N. americanus and S. mansoni co-infection. These risk factors could, however, only account for an estimated 32% of variability between households.

CONCLUSIONS

Our results demonstrate that variability in risk of N. americanus and S. mansoni co-infection between households cannot be entirely explained by exposure-related risk factors, emphasizing the possible role of other household factors in the heterogeneous distribution of helminth co-infection. Untangling the relative contribution of intrinsic host factors from household and environmental determinants therefore remains critical to our understanding of helminth epidemiology.

Spatial prediction of Plasmodium falciparum prevalence in Somalia

BACKGROUND

Maps of malaria distribution are vital for optimal allocation of resources for anti-malarial activities. There is a lack of reliable contemporary malaria maps in endemic countries in sub-Saharan Africa. This problem is particularly acute in low malaria transmission countries such as those located in the horn of Africa.

METHODS

Data from a national malaria cluster sample survey in 2005 and routine cluster surveys in 2007 were assembled for Somalia. Rapid diagnostic tests were used to examine the presence of Plasmodium falciparum parasites in finger-prick blood samples obtained from individuals across all age-groups. Bayesian geostatistical models, with environmental and survey covariates, were used to predict continuous maps of malaria prevalence across Somalia and to define the uncertainty associated with the predictions.

RESULTS

For analyses the country was divided into north and south. In the north, the month of survey, distance to water, precipitation and temperature had no significant association with P. falciparum prevalence when spatial correlation was taken into account. In contrast, all the covariates, except distance to water, were significantly associated with parasite prevalence in the south. The inclusion of covariates improved model fit for the south but not for the north. Model precision was highest in the south. The majority of the country had a predicted prevalence of < 5%; areas with > or = 5% prevalence were predominantly in the south.

CONCLUSION

The maps showed that malaria transmission in Somalia varied from hypo- to meso-endemic. However, even after including the selected covariates in the model, there still remained a considerable amount of unexplained spatial variation in parasite prevalence, indicating effects of other factors not captured in the study. Nonetheless the maps presented here provide the best contemporary information on malaria prevalence in Somalia.

High resolution population maps for low income nations: combining land cover and census in East Africa

Background

Between 2005 and 2050, the human population is forecast to grow by 2.7 billion, with the vast majority of this growth occurring in low income countries. This growth is likely to have significant social, economic and environmental impacts, and make the achievement of international development goals more difficult. The measurement, monitoring and potential mitigation of these impacts require high resolution, contemporary data on human population distributions. In low income countries, however, where the changes will be concentrated, the least information on the distribution of population exists. In this paper we investigate whether satellite imagery in combination with land cover information and census data can be used to create inexpensive, high resolution and easily-updatable settlement and population distribution maps over large areas.

Methodology/Principal Findings

We examine various approaches for the production of maps of the East African region (Kenya, Uganda, Burundi, Rwanda and Tanzania) and where fine resolution census data exists, test the accuracies of map production approaches and existing population distribution products. The results show that combining high resolution census, settlement and land cover information is important in producing accurate population distribution maps.

Conclusions

We find that this semi-automated population distribution mapping at unprecedented spatial resolution produces more accurate results than existing products and can be undertaken for as little as $0.01 per km². The resulting population maps are a product of the Malaria Atlas Project (MAP: http://www.map.ox.ac.uk) and are freely available.

The malaria Atlas Project: developing global maps of malaria risk

The primary goal of the recently launched Malaria Atlas Project is to develop the science of malaria cartography.

Mapping the environmental coverage of the INDEPTH demographic surveillance system network in rural Africa

OBJECTIVES

The INDEPTH DSS network was founded in 1998 to provide an international network of field sites for continuous demographic evaluation of populations and their health. Results from the network have been used to derive estimates of mortality, morbidity and health equity. Spatial extrapolation and logical summaries of these findings are dependent on the network covering a representative sample of the environments in a region and their interrelationships being known. Here, we investigate how comprehensive is the coverage of the network of rural DSS sites in Africa in terms of the range of ecological zones found across the continent.

METHODS

We used satellite imagery to define an environmental signature for each INDEPTH DSS site, and then calculate Euclidean distances from these signatures to the environmental signatures of every image pixel across Africa. These distances were then mapped and a gridded population surface used to mask uninhabited areas to illustrate the extent of the environmental coverage of the INDEPTH network. Environmental similarities between DSS sites were also calculated, hierarchically clustered and visualized as a dendrogram to examine between site relationships. Finally, an ecozonation of Africa was used to analyse the per-ecozone environmental similarity of the INDEPTH DSS network.

RESULTS AND CONCLUSIONS

The current INDEPTH DSS network in Africa spans all the major environmental zones, but within these zones the environmental coverage of the network varies. These variations were mapped by ecozone. These maps provide valuable information in determining the confidence with which relationships derived from rural INDEPTH DSS sites can be extended to other areas. The results also indicate suites of sites that form environmentally cohesive groups and from which data can be logically summarized. Finally, the results highlight areas where the location of new INDEPTH DSS sites would increase significantly the environmental coverage of the network.

Global environmental data for mapping infectious disease distribution

This contribution documents the satellite data archives, data processing methods and temporal Fourier analysis (TFA) techniques used to create the remotely sensed datasets on the DVD distributed with this volume. The aim is to provide a detailed reference guide to the genesis of the data, rather than a standard review. These remotely sensed data cover the entire globe at either 1 x 1 or 8 x 8 km spatial resolution. We briefly evaluate the relationships between the 1 x 1 and 8 x 8 km global TFA products to explore their inter-compatibility. The 8 x 8 km TFA surfaces are used in the mapping procedures detailed in the subsequent disease mapping reviews, since the 1 x 1 km products have been validated less widely. Details are also provided on additional, current and planned sensors that should be able to provide continuity with these environmental variable surfaces, as well as other sources of global data that may be used for mapping infectious disease.