Evaluation of an open forecasting challenge to assess skill of West Nile virus neuroinvasive disease prediction

BACKGROUND

West Nile virus (WNV) is the leading cause of mosquito-borne illness in the continental USA. WNV occurrence has high spatiotemporal variation, and current approaches to targeted control of the virus are limited, making forecasting a public health priority. However, little research has been done to compare strengths and weaknesses of WNV disease forecasting approaches on the national scale. We used forecasts submitted to the 2020 WNV Forecasting Challenge, an open challenge organized by the Centers for Disease Control and Prevention, to assess the status of WNV neuroinvasive disease (WNND) prediction and identify avenues for improvement.

METHODS

We performed a multi-model comparative assessment of probabilistic forecasts submitted by 15 teams for annual WNND cases in US counties for 2020 and assessed forecast accuracy, calibration, and discriminatory power. In the evaluation, we included forecasts produced by comparison models of varying complexity as benchmarks of forecast performance. We also used regression analysis to identify modeling approaches and contextual factors that were associated with forecast skill.

RESULTS

Simple models based on historical WNND cases generally scored better than more complex models and combined higher discriminatory power with better calibration of uncertainty. Forecast skill improved across updated forecast submissions submitted during the 2020 season. Among models using additional data, inclusion of climate or human demographic data was associated with higher skill, while inclusion of mosquito or land use data was associated with lower skill. We also identified population size, extreme minimum winter temperature, and interannual variation in WNND cases as county-level characteristics associated with variation in forecast skill.

CONCLUSIONS

Historical WNND cases were strong predictors of future cases with minimal increase in skill achieved by models that included other factors. Although opportunities might exist to specifically improve predictions for areas with large populations and low or high winter temperatures, areas with high case-count variability are intrinsically more difficult to predict. Also, the prediction of outbreaks, which are outliers relative to typical case numbers, remains difficult. Further improvements to prediction could be obtained with improved calibration of forecast uncertainty and access to real-time data streams (e.g. current weather and preliminary human cases).

Linking rattiness, geography and environmental degradation to spillover Leptospira infections in marginalised urban settings: An eco-epidemiological community-based cohort study in Brazil

Background

Zoonotic spillover from animal reservoirs is responsible for a significant global public health burden, but the processes that promote spillover events are poorly understood in complex urban settings. Endemic transmission of Leptospira, the agent of leptospirosis, in marginalised urban communities occurs through human exposure to an environment contaminated by bacteria shed in the urine of the rat reservoir. However, it is unclear to what extent transmission is driven by variation in the distribution of rats or by the dispersal of bacteria in rainwater runoff and overflow from open sewer systems.

Methods

We conducted an eco-epidemiological study in a high-risk community in Salvador, Brazil, by prospectively following a cohort of 1401 residents to ascertain serological evidence for leptospiral infections. A concurrent rat ecology study was used to collect information on the fine-scale spatial distribution of 'rattiness', our proxy for rat abundance and exposure of interest. We developed and applied a novel geostatistical framework for joint spatial modelling of multiple indices of disease reservoir abundance and human infection risk.

Results

The estimated infection rate was 51.4 (95%CI 40.4, 64.2) infections per 1000 follow-up events. Infection risk increased with age until 30 years of age and was associated with male gender. Rattiness was positively associated with infection risk for residents across the entire study area, but this effect was stronger in higher elevation areas (OR 3.27 95% CI 1.68, 19.07) than in lower elevation areas (OR 1.14 95% CI 1.05, 1.53).

Conclusions

These findings suggest that, while frequent flooding events may disperse bacteria in regions of low elevation, environmental risk in higher elevation areas is more localised and directly driven by the distribution of local rat populations. The modelling framework developed may have broad applications in delineating complex animal-environment-human interactions during zoonotic spillover and identifying opportunities for public health intervention.

Funding

This work was supported by the Oswaldo Cruz Foundation and Secretariat of Health Surveillance, Brazilian Ministry of Health, the National Institutes of Health of the United States (grant numbers F31 AI114245, R01 AI052473, U01 AI088752, R01 TW009504 and R25 TW009338); the Wellcome Trust (102330/Z/13/Z), and by the Fundação de Amparo à Pesquisa do Estado da Bahia (FAPESB/JCB0020/2016). MTE was supported by a Medical Research UK doctorate studentship. FBS participated in this study under a FAPESB doctorate scholarship.

Detection and persistence of environmental DNA (eDNA) of the different developmental stages of a vector mosquito, Culex pipiens pallens

Preventing mosquito-borne infectious diseases requires that vector mosquitoes are monitored and controlled. Targeting immature mosquitoes (eggs, larvae, and pupae), which have less mobility than adults, is an effective management approach. However, conducting these surveys is often difficult due to the limitations of morphological classification and survey costs. The application of environmental DNA (eDNA) analysis can solve these issues because it allows easy estimation of species distribution and morphology-independent species identification. Although a few previous studies have reported mosquito eDNA detection, there is a gap in knowledge regarding the dynamics related to the persistence of immature mosquito eDNA. We used Culex pipiens pallens, a vector of West Nile fever, as a model species. First, we developed a species-specific detection assay and confirmed its specificity using in silico and in vitro tests. Next, we conducted laboratory experiments using breeding tanks. Water samples were collected at each developmental stage. In addition, water samples were collected daily until the seventh day after emergence from the pupae. We quantified eDNA using real-time PCR with the developed assay to investigate the dynamics of mosquito eDNA. The specificity of the developed assay was confirmed by in silico and in vitro tests. Mosquito eDNA was detected at all developmental stages and detected up to seven days after emergence of pupae. In particular, high concentrations of eDNA were detected immediately after hatching from eggs and after emergence from pupae. Highly frequent positive eDNA signals were continuously detected between egg hatching and pupa hatching. Mosquito eDNA was detected immediately after the eggs were introduced, and eDNA-positive detections continued until pupae emergence, suggesting that eDNA analysis is useful for monitoring mosquito larvae. In the future, monitoring immature mosquitoes using eDNA analysis will contribute to prevent mosquito-borne infectious diseases.

Integrated Forecasts Based on Public Health Surveillance and Meteorological Data Predict West Nile Virus in a High-Risk Region of North America

BACKGROUND

West Nile virus (WNV), a global arbovirus, is the most prevalent mosquito-transmitted infection in the United States. Forecasts of WNV risk during the upcoming transmission season could provide the basis for targeted mosquito control and disease prevention efforts. We developed the Arbovirus Mapping and Prediction (ArboMAP) WNV forecasting system and used it in South Dakota from 2016 to 2019. This study reports a post hoc forecast validation and model comparison.

OBJECTIVES

Our objective was to validate historical predictions of WNV cases with independent data that were not used for model calibration. We tested the hypothesis that predictive models based on mosquito surveillance data combined with meteorological variables were more accurate than models based on mosquito or meteorological data alone.

METHODS

The ArboMAP system incorporated models that predicted the weekly probability of observing one or more human WNV cases in each county. We compared alternative models with different predictors including a) a baseline model based only on historical WNV cases, b) mosquito models based on seasonal patterns of infection rates, c) environmental models based on lagged meteorological variables, including temperature and vapor pressure deficit, d) combined models with mosquito infection rates and lagged meteorological variables, and e) ensembles of two or more combined models. During the WNV season, models were calibrated using data from previous years and weekly predictions were made using data from the current year. Forecasts were compared with observed cases to calculate the area under the receiver operating characteristic curve (AUC) and other metrics of spatial and temporal prediction error.

RESULTS

Mosquito and environmental models outperformed the baseline model that included county-level averages and seasonal trends of WNV cases. Combined models were more accurate than models based only on meteorological or mosquito infection variables. The most accurate model was a simple ensemble mean of the two best combined models. Forecast accuracy increased rapidly from early June through early July and was stable thereafter, with a maximum AUC of 0.85. The model predictions captured the seasonal pattern of WNV as well as year-to-year variation in case numbers and the geographic pattern of cases.

DISCUSSION

The predictions reached maximum accuracy early enough in the WNV season to allow public health responses before the peak of human cases in August. This early warning is necessary because other indicators of WNV risk, including early reports of human cases and mosquito abundance, are poor predictors of case numbers later in the season. https://doi.org/10.1289/EHP10287.

First detection of WNV RNA presence in field-collected mosquitoes in Cyprus

West Nile virus (WNV) infections have increased over recent years to the extent that WNV has become one of the most widespread arboviruses in the world, with potential consequences for both human and animal health. While much is known about WNV and the vectors that transmit it from their primary hosts across continental Europe, little is known about the epidemiology of the disease on the island of Cyprus. In this study, the aim was to investigate the prevalence of WNV infection in potential mosquito vectors for the first time in the Republic of Cyprus, using WNV surveillance of mosquitoes. Mosquitoes were collected in 2019, during which an outbreak in humans had occurred, and sampled mosquitoes were then examined for WNV infection by testing them for the presence of WNV RNA. Of 126 mosquito pools tested, one pool, containing Culex pipiens mosquitoes sampled from the Nicosia district, was found to be positive for the presence of WNV RNA. The positive pool found in this study represents the first demonstration of WNV in mosquitoes in Cyprus and confirms that human cases in Cyprus are likely the result of transmission via local Culex mosquitoes.

A multivariate geostatistical framework for combining multiple indices of abundance for disease vectors and reservoirs: a case study of rattiness in a low-income urban Brazilian community

A key requirement in studies of endemic vector-borne or zoonotic disease is an estimate of the spatial variation in vector or reservoir host abundance. For many vector species, multiple indices of abundance are available, but current approaches to choosing between or combining these indices do not fully exploit the potential inferential benefits that might accrue from modelling their joint spatial distribution. Here, we develop a class of multivariate generalized linear geostatistical models for multiple indices of abundance. We illustrate this novel methodology with a case study on Norway rats in a low-income urban Brazilian community, where rat abundance is a likely risk factor for human leptospirosis. We combine three indices of rat abundance to draw predictive inferences on a spatially continuous latent process, rattiness, that acts as a proxy for abundance. We show how to explore the association between rattiness and spatially varying environmental factors, evaluate the relative importance of each of the three contributing indices and assess the presence of residual, unexplained spatial variation, and identify rattiness hotspots. The proposed methodology is applicable more generally as a tool for understanding the role of vector or reservoir host abundance in predicting spatial variation in the risk of human disease.

West Nile Virus Mosquito Vectors in North America

In North America, the geographic distribution, ecology, and vectorial capacity of a diverse assemblage of mosquito species belonging to the genus Culex determine patterns of West Nile virus transmission and disease risk. East of the Mississippi River, mostly ornithophagic Culex pipiens L. complex mosquitoes drive intense enzootic transmission with relatively small numbers of human cases. Westward, the presence of highly competent Culex tarsalis (Coquillett) under arid climate and hot summers defines the regions with the highest human risk. West Nile virus human risk distribution is not uniform geographically or temporally within all regions. Notable geographic 'hotspots' persist with occasional severe outbreaks. Despite two decades of comprehensive research, several questions remain unresolved, such as the role of non-Culex bridge vectors, which are not involved in the enzootic cycle, but may be involved in virus transmission to humans. The absence of bridge vectors also may help to explain the frequent lack of West Nile virus 'spillover' into human populations despite very intense enzootic amplification in the eastern United States. This article examines vectorial capacity and the eco-epidemiology of West Nile virus mosquito vectors in four geographic regions of North America and presents some of the unresolved questions.

Models and Surveillance Systems to Detect and Predict West Nile Virus Outbreaks

Over the past 20 yr, many models have been developed to predict risk for West Nile virus (WNV; Flaviviridae: Flavivirus) disease in the human population. These models have aided our understanding of the meteorological and land-use variables that drive spatial and temporal patterns of human disease risk. During the same period, electronic data systems have been adopted by surveillance programs across much of the United States, including a growing interest in integrated data services that preserve the autonomy and attribution of credit to originating agencies but facilitate data sharing, analysis, and visualization at local, state, and national scales. At present, nearly all predictive models have been limited to the scientific literature, with few having been implemented for use by public-health and vector-control decision makers. The current article considers the development of models for spatial patterns, early warning, and early detection of WNV over the last 20 yr and considers some possible paths toward increasing the utility of these models for guiding interventions.

Linking Bird and Mosquito Data to Assess Spatiotemporal West Nile Virus Risk in Humans

West Nile virus (WNV; family Flaviviridae) causes a disease in humans that may develop into a deadly neuroinvasive disease. In North America, several peridomestic bird species can develop sufficient viremia to infect blood-feeding mosquito vectors without succumbing to the virus. Mosquito species from the genus Culex, Aedes and Ochlerotatus display variable host preferences, ranging between birds and mammals, including humans, and may bridge transmission among avian hosts and contribute to spill-over transmission to humans. In this study, we aimed to test the effect of density of three mosquito species and two avian species on WNV mosquito infection rates and investigated the link between spatiotemporal clusters of high mosquito infection rates and clusters of human WNV cases. We based our study around the city of Ottawa, Canada, between the year 2007 and 2014. We found a large effect size of density of two mosquito species on mosquito infection rates. We also found spatiotemporal overlap between a cluster of high mosquito infection rates and a cluster of human WNV cases. Our study is innovative because it suggests a role of avian and mosquito densities on mosquito infection rates and, in turn, on hotspots of human WNV cases.

Risk mapping of West Nile virus circulation in Spain, 2015

West Nile fever is an emergent disease in Europe. The objective of this study was to conduct a predictive risk mapping of West Nile Virus (WNV) circulation in Spain based on historical data of WNV circulation. Areas of Spain with evidence of WNV circulation were mapped based on data from notifications to the surveillance systems and a literature review. A logistic regression-based spatial model was used to assess the probability of WNV circulation. Data were analyzed at municipality level. Mean temperatures of the period from June to October, presence of wetlands and presence of Special Protection Areas for birds were considered as potential predictors. Two predictors of WNV circulation were identified: higher temperature [adjusted odds ratio (AOR) 2.07, 95% CI 1.82-2.35, p<0.01] and presence of wetlands (3.37, 95% CI 1.89-5.99, p<0.01). Model validations indicated good predictions: area under the ROC curve was 0.895 (95% CI 0.870-0.919) for internal validation and 0.895 (95% CI 0.840-0.951) for external validation. This model could support improvements of WNV risk- based surveillance in Spain. The importance of a comprehensive surveillance for WNF, including human, animal and potential vectors is highlighted, which could additionally result in model refinements.

Review of West Nile virus circulation and outbreak risk in Madagascar: Entomological and ornithological perspectives

West Nile fever (WNF) is a zoonotic disease, occurring nearly globally. In Madagascar, West Nile virus (WNV) was first detected in 1978 from wild birds and the virus is currently distributed across the island, but no epidemic or epizootic period has been recorded. One fatal human case of WNV infection was reported in 2011, suggesting a "tip of the iceberg" phenomenon of a possible WNF epidemic/epizootic on the island. The main objective of this literature-based survey is to review patterns of WNV circulation in Madagascar from the entomological and ornithological points of view. Among the 235 mosquito species described from Madagascar, 29 species are widely associated with WNV infection; 16 of them are found naturally infected with WNV on the island and categorized into major, candidate, and potential vectors of WNV according to their vector capacity. This study upholds the hypothesis that WNV enzooticity is independent of annual movements of migratory birds passing through Madagascar. Moreover, the lack of regular migratory bird flux between Africa and Madagascar would reduce the probability of transmission and the subsequent reintroduction of the virus into locally occurring mosquito species. Given that Palearctic migratory birds are strongly implicated in the transmission of WNV, we highlight notable differences in the movements and species diversity of these birds in Madagascar as compared to eastern and northern Africa. Risk factors from this two-pronged approach are presented for the emergence of WNF outbreak.

Temporal and Spatial Variability of Entomological Risk Indices for West Nile Virus Infection in Northern Colorado: 2006-2013

West Nile virus (WNV) is enzootic in northern Colorado. Annual surveillance activities in Fort Collins, CO, include collecting female Culex mosquitoes and testing them for the presence of WNV RNA in order to calculate 1) Culex female abundance, 2) WNV infection rate, and 3) the vector index (VI). These entomological risk indices inform public policy regarding the need for emergency adulticiding. Currently, these are calculated on a city-wide basis. In this study, we present descriptive data from historical surveillance records spanning 2006-2013 to discern seasonal and yearly patterns of entomological risk for WNV infection. Also, we retrospectively test the hypothesis that entomological risk is correlated with human transmission risk and is heterogeneous within the City of Fort Collins. Four logistically relevant zones within the city were established and used to test this hypothesis. Zones in the eastern portion of the city consistently had significantly higher Culex abundance and VI compared with zones in the west, leading to higher entomological risk indicators for human WNV infection in the east. Moreover, the relative risk of a reported human case of WNV infection was significantly higher in the eastern zones of the city. Our results suggest that a more spatially targeted WNV management program may better mitigate human risk for WNV infection in Fort Collins, and possibly other cities where transmission is enzootic, while at the same time reducing pesticide use.

Ecological approaches in veterinary epidemiology: mapping the risk of bat-borne rabies using vegetation indices and night-time light satellite imagery

Rabies remains a disease of significant public health concern. In the Americas, bats are an important source of rabies for pets, livestock, and humans. For effective rabies control and prevention, identifying potential areas for disease occurrence is critical to guide future research, inform public health policies, and design interventions. To anticipate zoonotic infectious diseases distribution at coarse scale, veterinary epidemiology needs to advance via exploring current geographic ecology tools and data using a biological approach. We analyzed bat-borne rabies reports in Chile from 2002 to 2012 to establish associations between rabies occurrence and environmental factors to generate an ecological niche model (ENM). The main rabies reservoir in Chile is the bat species Tadarida brasiliensis; we mapped 726 occurrences of rabies virus variant AgV4 in this bat species and integrated them with contemporary Normalized Difference Vegetation Index (NDVI) data from the Moderate Resolution Imaging Spectroradiometer (MODIS). The correct prediction of areas with rabies in bats and the reliable anticipation of human rabies in our study illustrate the usefulness of ENM for mapping rabies and other zoonotic pathogens. Additionally, we highlight critical issues with selection of environmental variables, methods for model validation, and consideration of sampling bias. Indeed, models with weak or incorrect validation approaches should be interpreted with caution. In conclusion, ecological niche modeling applications for mapping disease risk at coarse geographic scales have a promising future, especially with refinement and enrichment of models with additional information, such as night-time light data, which increased substantially the model’s ability to anticipate human rabies.

Modeling Culex tarsalis abundance on the northern Colorado front range using a landscape-level approach

Remote sensing and Geographic Information System (GIS) data can be used to identify larval mosquito habitats and predict species distribution and abundance across a landscape. An understanding of the landscape features that impact abundance and dispersal can then be applied operationally in mosquito control efforts to reduce the transmission of mosquito-borne pathogens. In an effort to better understand the effects of landscape heterogeneity on the abundance of the West Nile virus (WNV) vector Culex tarsalis, we determined associations between GIS-based environmental data at multiple spatial extents and monthly abundance of adult Cx. tarsalis in Larimer County and Weld County, CO. Mosquito data were collected from Centers for Disease Control and Prevention miniature light traps operated as part of local WNV surveillance efforts. Multiple regression models were developed for prediction of monthly Cx. tarsalis abundance for June, July, and August using 4 years of data collected over 2007-10. The models explained monthly adult mosquito abundance with accuracies ranging from 51-61% in Fort Collins and 57-88% in Loveland-Johnstown. Models derived using landscape-level predictors indicated that adult Cx. tarsalis abundance is negatively correlated with elevation. In this case, low-elevation areas likely more abundantly include habitats for Cx. tarsalis. Model output indicated that the perimeter of larval sites is a significant predictor of Cx. tarsalis abundance at a spatial extent of 500 m in Loveland-Johnstown in all months examined. The contribution of irrigated crops at a spatial extent of 500 m improved model fit in August in both Fort Collins and Loveland-Johnstown. These results emphasize the significance of irrigation and the manual control of water across the landscape to provide viable larval habitats for Cx. tarsalis in the study area. Results from multiple regression models can be applied operationally to identify areas of larval Cx. tarsalis production (irrigated crops lands and standing water) and assign priority in larval treatments to areas with a high density of larval sites at relevant spatial extents around urban locations.

West Nile virus state of the art report of MALWEST Project

During the last three years Greece is experiencing the emergence of West Nile virus (WNV) epidemics. Within this framework, an integrated surveillance and control programme (MALWEST project) with thirteen associate partners was launched aiming to investigate the disease and suggest appropriate interventions. One out of seven work packages of the project is dedicated to the State of the Art report for WNV. Three expert working groups on humans, animals and mosquitoes were established. Medical databases (PubMed, Scopus) were searched together with websites: e.g., WHO, CDC, ECDC. In total, 1,092 relevant articles were initially identified and 258 of them were finally included as references regarding the current knowledge about WNV, along with 36 additional sources (conference papers, reports, book chapters). The review is divided in three sections according to the fields of interest: (1) WNV in humans (epidemiology, molecular characteristics, transmission, diagnosis, treatment, prevention, surveillance); (2) WNV in animals (epidemiological and transmission characteristics concerning birds, horses, reptiles and other animal species) and (3) WNV in mosquitoes (control, surveillance). Finally, some examples of integrated surveillance programmes are presented. The introduction and establishment of the disease in Greece and other European countries further emphasizes the need for thorough research and broadening of our knowledge on this viral pathogen.

Spatio-temporal epidemiology of human West Nile virus disease in South Dakota

Despite a cold temperate climate and low human population density, the Northern Great Plains has become a persistent hot spot for human West Nile virus (WNV) disease in North America. Understanding the spatial and temporal patterns of WNV can provide insights into the epidemiological and ecological factors that influence disease emergence and persistence. We analyzed the 1,962 cases of human WNV disease that occurred in South Dakota from 2002-2012 to identify the geographic distribution, seasonal cycles, and interannual variability of disease risk. The geographic and seasonal patterns of WNV have changed since the invasion and initial epidemic in 2002-2003, with cases shifting toward the eastern portion of South Dakota and occurring earlier in the transmission season in more recent years. WNV cases were temporally autocorrelated at lags of up to six weeks and early season cumulative case numbers were correlated with seasonal totals, indicating the possibility of using these data for short-term early detection of outbreaks. Epidemiological data are likely to be most effective for early warning of WNV virus outbreaks if they are integrated with entomological surveillance and environmental monitoring to leverage the strengths and minimize the weaknesses of each information source.

Predicting human West Nile virus infections with mosquito surveillance data

West Nile virus (WNV) has become established across the Americas with recent heightened activity causing significant human illness. Surveillance methods to predict the risk of human infection are urgently needed to initiate timely preventative measures and justify the expense of implementing costly or unpopular control measures, such as aerial spraying or curfews. We quantified the links between mosquito surveillance data and the spatiotemporal patterns of 3,827 human WNV cases reported over 5 years in Colorado from 2003 to 2007. Mosquito data were strongly predictive of variation in the number of human WNV infections several weeks in advance in both a spatiotemporal statewide analysis and temporal variation within counties with substantial numbers of human cases. We outline several ways to further improve the predictive power of these data and we quantify the loss of information if no funds are available for testing mosquitoes for WNV. These results demonstrate that mosquito surveillance provides a valuable public health tool for assessing the risk of human arboviral infections, allocating limited public health resources, and justifying emergency control actions.

The contribution of Culex pipiens complex mosquitoes to transmission and persistence of West Nile virus in North America

Mosquitoes within the Culex pipiens complex have been implicated as major vectors of West Nile virus (WNV) in North America due to their seasonal abundance, vector competence and high field infection rates. However, the role of Cx. p. pipiens complex mosquitoes in enzootic amplification of WNV among avian hosts and epidemic transmission to humans varies throughout its geographical distribution. In the northeastern United States, Cx. p. pipiens is recognized as the primary enzootic vector responsible for amplification of virus among wild bird populations. However, because this mosquito is strongly ornithophilic, its role in transmission to humans appears to be more limited in this region. In the north central and Mid-Atlantic States by contrast, Cx. p. pipiens shows an increased affinity for human hosts and has been incriminated as a key bridge vector. In southern regions of the United States, Culex p. quinquefasciatus are more opportunistic feeders, and are thought to be principal enzootic and epidemic vectors. In western regions of the United States where Culex tarsalis predominates, especially in rural areas, Cx. p. pipiens and Cx. p. quinquefasciatus play roles that are more limited and are recognized as secondary vectors. In the southwestern United States Cx. p. quinquefasciatus also appears to be the predominant vector in urban habitats, but only a secondary vector in more rural environs. The direct involvement of Cx. p. pipiens form molestus in WNV transmission is largely unknown, but human-biting Cx. p. pipiens are more likely to have a probability of genetic ancestry with Cx. p. pipiens form molestus. The detection of WNV from overwintering populations of diapausing Cx. p. pipiens and non-diapausing Cx. p. quinquefaciatus and their role in local overwintering of WNV are addressed.

Zoonotic infections among employees from Great Smoky Mountains and Rocky Mountain National Parks, 2008-2009

U.S. National Park Service employees may have prolonged exposure to wildlife and arthropods, placing them at increased risk of infection with endemic zoonoses. To evaluate possible zoonotic risks present at both Great Smoky Mountains (GRSM) and Rocky Mountain (ROMO) National Parks, we assessed park employees for baseline seroprevalence to specific zoonotic pathogens, followed by evaluation of incident infections over a 1-year study period. Park personnel showed evidence of prior infection with a variety of zoonotic agents, including California serogroup bunyaviruses (31.9%), Bartonella henselae (26.7%), spotted fever group rickettsiae (22.2%), Toxoplasma gondii (11.1%), Anaplasma phagocytophilum (8.1%), Brucella spp. (8.9%), flaviviruses (2.2%), and Bacillus anthracis (1.5%). Over a 1-year study period, we detected incident infections with leptospirosis (5.7%), B. henselae (5.7%), spotted fever group rickettsiae (1.5%), T. gondii (1.5%), B. anthracis (1.5%), and La Crosse virus (1.5%) in staff members at GRSM, and with spotted fever group rickettsiae (8.5%) and B. henselae (4.3%) in staff at ROMO. The risk of any incident infection was greater for employees who worked as resource managers (OR 7.4; 95% CI 1.4,37.5; p=0.02), and as law enforcement rangers/rescue crew (OR 6.5; 95% CI 1.1,36.5; p=0.03), relative to those who worked primarily in administration or management. The results of this study increase our understanding of the pathogens circulating within both parks, and can be used to inform the development of effective guidelines and interventions to increase visitor and staff awareness and help prevent exposure to zoonotic agents.

National and regional associations between human West Nile virus incidence and demographic, landscape, and land use conditions in the coterminous United States

The incidence of human West Nile virus (WNV) varies spatially and temporally and is influenced by a wide range of biotic and abiotic factors. There are numerous important vector species, with variable geographic ranges and ecologies, considered crucial to the transmission of WNV in the coterminous United States. To date there has been a lack of a systematic investigation in the United States, at a regional scale, of the wide variety of landscape, land use, and demographic influences on WNV incidence. In this study, we use published vector species distribution maps, as well as prominent landscape features, to define six distinct regions of the coterminous United States. We relate data on demographic, landscape, and land use conditions to the incidence of human WNV by region recorded at county level in the coterminous United States from 2002-2009. The observed relationships varied by region with the Great Plains, Northwest, and Southwest regions showing high WNV incidence associated with rural irrigated landscapes, indicating the importance of Culex tarsalis as the primary vector. In the Southeast, the percent of the population in poverty was positively associated with high WNV incidence, potentially indicating the quality of housing in relation to the vector Culex quinquefasciatus, a mosquito that often feeds indoors. The Northeast region human WNV incidence was positively associated with agricultural landscapes, potentially implying the importance of Culex restuans in a region generally thought of as being dominated by Culex pipiens transmission. There was strong spatial autocorrelation in most of the regions, but with a spatial autologistic term accounted for in binary logistic regression models, there were significant landscape, land use, and demographic covariates for each region.

The effect of spatial and temporal subsetting on Culex tarsalis abundance models--a design for sensible reduction of vector surveillance

Early identification of increasing mosquito activity is critical to effective mosquito control, particularly when increasing host-seeking behavior may be associated with increased risk of mosquito-borne disease. In this paper, we analyzed the temporal abundance pattern of the West Nile Virus vector, Culex tarsalis, in Fort Collins, CO, using an autoregressive integrated moving average model. We determined that an autoregressive model order 5 with lagged minimum temperatures was best at describing the seasonal abundance of Cx. tarsalis. We then tested the effect of using both temporal and spatial subsets of the data to determine the effect of reduced sampling effort on abundance predictions. We found that, if reduced trapping is necessary due to limited resources, removal of the least productive 1/3 or 1/4 of the traps produced the least erroneous predictions of seasonality represented in the observed data. We show that this productivity-based subset scheme performs better than other sampling effort reductions in generating the best estimate of Cx. tarsalis abundance per trap-night.

Using geographic information systems and decision support systems for the prediction, prevention, and control of vector-borne diseases

Emerging and resurging vector-borne diseases cause significant morbidity and mortality, especially in the developing world. We focus on how advances in mapping, Geographic Information System, and Decision Support System technologies, and progress in spatial and space-time modeling, can be harnessed to prevent and control these diseases. Major themes, which are addressed using examples from tick-borne Lyme borreliosis; flea-borne plague; and mosquito-borne dengue, malaria, and West Nile virus disease, include (a) selection of spatial and space-time modeling techniques, (b) importance of using high-quality and biologically or epidemiologically relevant data, (c) incorporation of new technologies into operational vector and disease control programs, (d) transfer of map-based information to stakeholders, and (e) adaptation of technology solutions for use in resource-poor environments. We see great potential for the use of new technologies and approaches to more effectively target limited surveillance, prevention, and control resources and to reduce vector-borne and other infectious diseases.

Population genetic data suggest a role for mosquito-mediated dispersal of West Nile virus across the western United States

After introduction, West Nile virus (WNV) spread rapidly across the western United States between the years 2001 and 2004. This westward movement is thought to have been mediated by random dispersive movements of resident birds. Little attention has been placed on the role of mosquito vectors in virus dispersal across North America. The mosquito vector largely responsible for WNV amplification and transmission of WNV in the western USA is Culex tarsalis. Here we present population genetic data that suggest a potential role for C. tarsalis in the dispersal of WNV across the western USA. Population genetic structure across the species range of C. tarsalis in the USA was characterized in 16 states using 12 microsatellite loci. structure and geneland analyses indicated the presence of three broad population clusters. Barriers to gene flow were resolved near the Sonoran desert in southern Arizona and between the eastern Rocky Mountains and High Plains plateau. Small genetic distances among populations within clusters indicated that gene flow was not obstructed over large portions of the West Coast and within the Great Plains region. Overall, gene flow in C. tarsalis appears to be extensive, potentially mediated by movement of mosquitoes among neighbouring populations and hindered in geographically limited parts of its range. The pattern of genetic clustering in C. tarsalis is congruent with the pattern of invasion of WNV across the western United States, raising the possibility that movement of this important vector may be involved in viral dispersal.

Landscape epidemiology of vector-borne diseases

Landscape epidemiology describes how the temporal dynamics of host, vector, and pathogen populations interact spatially within a permissive environment to enable transmission. The spatially defined focus, or nidus, of transmission may be characterized by vegetation as well as by climate, latitude, elevation, and geology. The ecological complexity, dimensions, and temporal stability of the nidus are determined largely by pathogen natural history and vector bionomics. Host populations, transmission efficiency, and therefore pathogen amplification vary spatially, thereby creating a heterogeneous surface that may be defined by remote sensing and statistical tools. The current review describes the evolution of landscape epidemiology as a science and exemplifies selected aspects by contrasting the ecology of two different recent disease outbreaks in North America caused by West Nile virus, an explosive, highly virulent mosquito-borne virus producing ephemeral nidi, and Borrelia burgdorferi, a slowly amplifying chronic pathogen producing semipermanent nidi.

Mapping environmental dimensions of dengue fever transmission risk in the Aburrá Valley, Colombia

Dengue fever (DF) is endemic in Medellín, the second largest Colombian city, and surrounding municipalities. We used DF case and satellite environmental data to investigate conditions associated with suitable areas for DF occurrence in 2008 in three municipalities (Bello, Medellín and Itagüí). We develop spatially stratified tests of ecological niche models, and found generally good predictive ability, with all model tests yielding results significantly better than random expectations. We concluded that Bello and Medellín present ecological conditions somewhat different from, and more suitable for DF than, those of Itagüí. We suggest that areas predicted by our models as suitable for DF could be considered as at-risk, and could be used to guide campaigns for DF prevention in these municipalities.

Mosquitoes and West Nile virus along a river corridor from prairie to montane habitats in eastern Colorado

We conducted studies on mosquitoes and West Nile virus (WNV) along a riparian corridor following the South Platte River and Big Thompson River in northeastern Colorado and extending from an elevation of 1,215 m in the prairie landscape of the eastern Colorado plains to 1,840 m in low montane areas at the eastern edge of the Rocky Mountains in the central part of the state. Mosquito collection during June-September 2007 in 20 sites along this riparian corridor yielded a total of 199,833 identifiable mosquitoes of 17 species. The most commonly collected mosquitoes were, in descending order: Aedes vexans, Culex tarsalis, Ae. dorsalis, Ae. trivittatus, Ae. melanimon, Cx. pipiens, and Culiseta inornata. Species richness was higher in the plains than in foothills-montane areas, and abundances of several individual species, including the WNV vectors Cx. tarsalis and Cx. pipiens and the nuisance-biter and potential secondary WNV vector Ae. vexans, decreased dramatically from the plains (1,215-1,487 m) to foothills-montane areas (1,524-1,840 m). Ae. vexans and Cx. tarsalis had a striking pattern of uniformly high abundances between 1,200-1,450 m followed by a gradual decrease in abundance above 1,450 m to reach very low numbers above 1,550 m. Culex species were commonly infected with WNV in the plains portion of the riparian corridor in 2007, with 14 of 16 sites yielding WNV-infected Cx. tarsalis and infection rates for Cx. tarsalis females exceeding 2.0 per 1,000 individuals in ten of the sites. The Vector Index for abundance of WNV-infected Cx. tarsalis females during June-September exceeded 0.5 in six plains sites along the South Platte River but was uniformly low (0-0.1) in plains, foothills and montane sites above 1,500 m along the Big Thompson River. A population genetic analysis of Cx. tarsalis revealed that all collections from the ≈190 km riparian transect in northeastern Colorado were genetically uniform but that these collections were genetically distinct from collections from Delta County on the western slope of the Continental Divide. This suggests that major waterways in the Great Plains serve as important dispersal corridors for Cx. tarsalis but that the Continental Divide is a formidable barrier to this WNV vector.

Relationship between distance from major larval habitats and abundance of adult mosquitoes in semiarid plains landscapes in Colorado

We examined the relationship between distance from major larval habitats and abundance of adult mosquitoes in the semiarid plains landscape characteristic of eastern Colorado. Mosquito collection was conducted from late June to early August 2007 and included trap locations at distances ranging from <10 m up to 20-150 m and 160-373 m from three major larval habitats: one area along a river corridor and two small reservoirs. The study yielded 65,140 mosquitoes of 14 species, and five species were sufficiently abundant to be included in statistical analyses: Aedes vexans (Meigen), Culex tarsalis Coquillett, Ochlerotatus dorsalis (Meigen) (=Ae. dorsalis), Ochlerotatus melanimon (Dyar) (=Ae. melanimon), and Culex pipiens L. Distance to nearest major larval habitat was not strongly related to Culex abundance within the approximately = 400-m range from larval habitats examined in this study. Abundance of Ae. vexans declined significantly with distance from the larval habitat, whereas abundance was significantly higher in the 20-150- and 160-373-m classes compared with areas within 10 m of the larval habitat for both Ochlerotatus species. Except for Ae. vexans, however, we did not find monotonic increasing or decreasing abundance trends associated with distance from larval habitats for the 400-m range examined. This, combined with a finding that fine-scale habitat heterogeneity influenced abundance for most of the mosquitoes examined, underscores the importance of considering not only distance from larval habitat but also fine-scale habitat heterogeneity to understand how important nuisance-biters and West Nile virus (family Flaviviridae, genus Flavivirus, WNV) vectors use the landscape. We also discuss how these results relate to previous studies from western North America and explore their relevance to operational implementation of adulticides to suppress mosquito vectors during WNV disease outbreaks in the Great Plains.

Canine pulmonary angiostrongylosis: the influence of climate on parasite distribution

The geographic range of Angiostrongylus vasorum is expanding, leading to increased disease. Although observed cases of canine pulmonary angiostrongylosis have been dutifully reported in the literature, the state of biological knowledge remains too poor to predict future patterns of spread with any confidence. Nevertheless, there is an urgent need to identify areas that are likely to be suitable for parasite establishment. Preliminary attempts to do this using a climatic envelope approach suggest that several new areas are open to colonisation, even without invoking climate change. The risk of parasite importation into these areas should be mitigated, e.g. by restricting movement of dogs unless tested or treated for A. vasorum, and monitored by focused surveillance of definitive and intermediate hosts. These efforts will benefit from newly developed diagnostic tests.

Predictable ecology and geography of West Nile virus transmission in the central United States

West Nile virus (WNV) arrived in North America and spread rapidly through the western hemisphere. We present a series of tests to determine whether ecological factors are consistently associated with WNV transmission to humans. We analyzed human WNV cases in the states of Illinois, Indiana, and Ohio in 2002 and 2003, building ecological niche models to associate WNV case occurrences with ecological and environmental parameters. In essentially all tests, both within states, among states, between years, and across the region, we found high predictivity of WNV case distributions, suggesting that one or more elements in the WNV transmission cycle has a strong ecological determination. Areas in the geographic region included in this study predicted as suitable for WNV transmission tended to have lower values of the vegetation indices in the summer months, pointing to consistent ecological differences between suitable and unsuitable areas.

West Nile virus in the Americas

Since the first detection of West Nile virus in the Western Hemisphere in 1999, the virus has spread rapidly across the North American continent and as far south as Argentina. An unprecedented pattern of large annual epidemics of human neuroinvasive disease continues in North America, resulting in considerable public health impact. The high infection incidence in humans has resulted in non-mosquito transmission modes, such as through transfused blood and transplanted organs. West Nile virus incursion into Latin America and the Caribbean Islands has resulted in surprisingly low human, avian, and equine morbidity and mortality despite evidence that West Nile virus strains circulating in those regions are similar to those in North America.