House-to-house human movement drives dengue virus transmission

Modeling tools for dengue risk mapping - a systematic review

Introduction

The global spread and the increased frequency and magnitude of epidemic dengue in the last 50 years underscore the urgent need for effective tools for surveillance, prevention, and control. This review aims at providing a systematic overview of what predictors are critical and which spatial and spatio-temporal modeling approaches are useful in generating risk maps for dengue.

Methods

A systematic search was undertaken, using the PubMed, Web of Science, WHOLIS, Centers for Disease Control and Prevention (CDC) and OvidSP databases for published citations, without language or time restrictions. A manual search of the titles and abstracts was carried out using predefined criteria, notably the inclusion of dengue cases. Data were extracted for pre-identified variables, including the type of predictors and the type of modeling approach used for risk mapping.

Results

A wide variety of both predictors and modeling approaches was used to create dengue risk maps. No specific patterns could be identified in the combination of predictors or models across studies. The most important and commonly used predictors for the category of demographic and socio-economic variables were age, gender, education, housing conditions and level of income. Among environmental variables, precipitation and air temperature were often significant predictors. Remote sensing provided a source of varied land cover data that could act as a proxy for other predictor categories. Descriptive maps showing dengue case hotspots were useful for identifying high-risk areas. Predictive maps based on more complex methodology facilitated advanced data analysis and visualization, but their applicability in public health contexts remains to be established.

Conclusions

The majority of available dengue risk maps was descriptive and based on retrospective data. Availability of resources, feasibility of acquisition, quality of data, alongside available technical expertise, determines the accuracy of dengue risk maps and their applicability to the field of public health. A large number of unknowns, including effective entomological predictors, genetic diversity of circulating viruses, population serological profile, and human mobility, continue to pose challenges and to limit the ability to produce accurate and effective risk maps, and fail to support the development of early warning systems.

Electronic supplementary material

The online version of this article (doi:10.1186/1476-072X-13-50) contains supplementary material, which is available to authorized users.

Chikungunya cases identified through passive surveillance and household investigations--Puerto Rico, May 5-August 12, 2014

Chikungunya and dengue are mosquito-borne, viral, acute febrile illnesses that can be difficult to distinguish clinically. Whereas dengue is endemic in many countries in the Caribbean and the Americas, the first locally acquired chikungunya case in the Western Hemisphere was reported from the Caribbean island of St. Martin in December 2013 and was soon followed by cases in many parts of the region. In January 2014, the Puerto Rico Department of Health (PRDH) and CDC initiated chikungunya surveillance by building on an existing passive dengue surveillance system. To assess the extent of chikungunya in Puerto Rico, the severity of illnesses, and the health care-seeking behaviors of residents, PRDH and CDC analyzed data from passive surveillance and investigations conducted around the households of laboratory-positive chikungunya patients. Passive surveillance indicated that the first locally acquired, laboratory-positive chikungunya case in Puerto Rico was in a patient with illness onset on May 5, 2014. By August 12, a total of 10,201 suspected chikungunya cases (282 per 100,000 residents) had been reported. Specimens from 2,910 suspected cases were tested, and 1,975 (68%) were positive for chikungunya virus (CHIKV) infection. Four deaths were reported. The household investigations found that, of 250 participants, 70 (28%) tested positive for current or recent CHIKV infection, including 59 (84%) who reported illness within the preceding 3 months. Of 25 laboratory-positive participants that sought medical care, five (20%) were diagnosed with chikungunya and two (8%) were reported to PRDH. These investigative efforts indicated that chikungunya cases were underrecognized and underreported, prompting PRDH to conduct information campaigns to increase knowledge of the disease among health care professionals and the public. PRDH and CDC recommended that health care providers manage suspected chikungunya cases as they do dengue because of the similarities in symptoms and increased risk for complications in dengue patients that are not appropriately managed. Residents of and travelers to the tropics can minimize their risk for both chikungunya and dengue by taking standard measures to avoid mosquito bites.

Advances in the understanding, management, and prevention of dengue

Dengue causes more human morbidity globally than any other vector-borne viral disease. Recent research has led to improved epidemiological methods that predict disease burden and factors involved in transmission, a better understanding of immune responses in infection, and enhanced animal models. In addition, a number of control measures, including preventative vaccines, are in clinical trials. However, significant gaps remain, including the need for better surveillance in large parts of the world, methods to predict which individuals will develop severe disease, and immunologic correlates of protection against dengue illness. During the next decade, dengue will likely expand its geographic reach and become an increasing burden on health resources in affected areas. Licensed vaccines and antiviral agents are needed in order to effectively control dengue and limit disease.

Seroprevalence of neutralizing antibodies against dengue virus in two localities in the state of Morelos, Mexico

Humoral immune response against dengue virus (DENV) is an important component in dengue-endemic transmission. We conducted a cross-sectional nested cohort study to determine the seroprevalence and frequency of neutralizing antibodies against DENV serotypes in two endemic localities in the state of Morelos, Mexico. The cohort participants (N = 1,196) were screened to determine previous exposure to DENV. Overall seroprevalence was 76.6% (95% confidence interval [95% CI] = 73.6-79.2), and prevalence of neutralizing antibodies in the 5- to 9-year-old group was 82.5% (95% CI = 67.2-92.7), 45% (95% CI = 29.3-61.5), and 65% (95% CI = 48.3-79.4) for DENV-1, DENV-2, and DENV-3, respectively. For participants older than 10 years, the observed seroprevalence was above 60% for each serotype, except DENV-4 in the 10- to 25-year-old group (42.9%); 81% of humoral responses were multitypic. The outcomes of our study contribute to understanding the immune component of dengue transmission and provide focal information for the evaluation of vaccine candidates under development.

The spatial dynamics of dengue virus in Kamphaeng Phet, Thailand

BACKGROUND

Dengue is endemic to the rural province of Kamphaeng Phet, Northern Thailand. A decade of prospective cohort studies has provided important insights into the dengue viruses and their generated disease. However, as elsewhere, spatial dynamics of the pathogen remain poorly understood. In particular, the spatial scale of transmission and the scale of clustering are poorly characterized. This information is critical for effective deployment of spatially targeted interventions and for understanding the mechanisms that drive the dispersal of the virus.

METHODOLOGY/PRINCIPAL FINDINGS

We geocoded the home locations of 4,768 confirmed dengue cases admitted to the main hospital in Kamphaeng Phet province between 1994 and 2008. We used the phi clustering statistic to characterize short-term spatial dependence between cases. Further, to see if clustering of cases led to similar temporal patterns of disease across villages, we calculated the correlation in the long-term epidemic curves between communities. We found that cases were 2.9 times (95% confidence interval 2.7-3.2) more likely to live in the same village and be infected within the same month than expected given the underlying spatial and temporal distribution of cases. This fell to 1.4 times (1.2-1.7) for individuals living in villages 1 km apart. Significant clustering was observed up to 5 km. We found a steadily decreasing trend in the correlation in epidemics curves by distance: communities separated by up to 5 km had a mean correlation of 0.28 falling to 0.16 for communities separated between 20 km and 25 km. A potential explanation for these patterns is a role for human movement in spreading the pathogen between communities. Gravity style models, which attempt to capture population movement, outperformed competing models in describing the observed correlations.

CONCLUSIONS

There exists significant short-term clustering of cases within individual villages. Effective spatially and temporally targeted interventions deployed within villages may target ongoing transmission and reduce infection risk.

A spatial simulation model for dengue virus infection in urban areas

Background

The World Health Organization estimates that the global number of dengue infections range between 80–100 million per year, with some studies estimating approximately three times higher numbers. Furthermore, the geographic range of dengue virus transmission is extending with the disease now occurring more frequently in areas such as southern Europe. Ae. aegypti, one of the most prominent dengue vectors, is endemic to the far north-east of Australia and the city of Cairns frequently experiences dengue outbreaks which sometimes lead to large epidemics.

Method

A spatially-explicit, individual-based mathematical model that accounts for the spread of dengue infection as a result of human movement and mosquito dispersion is presented. The model closely couples the four key sub-models necessary for representing the overall dynamics of the physical system, namely those describing mosquito population dynamics, human movement, virus transmission and vector control. Important features are the use of high quality outbreak data and mosquito trapping data for calibration and validation and a strategy to derive local mosquito abundance based on vegetation coverage and census data.

Results

The model has been calibrated using detailed 2003 dengue outbreak data from Cairns, together with census and mosquito trapping data, and is shown to realistically reproduce a further dengue outbreak. The simulation results replicating the 2008/2009 Cairns epidemic support several hypotheses (formulated previously) aimed at explaining the large-scale epidemic which occurred in 2008/2009; specifically, while warmer weather and increased human movement had only a small effect on the spread of the virus, a shorter virus strain-specific extrinsic incubation time can explain the observed explosive outbreak of 2008/2009.

Conclusion

The proof-of-concept simulation model described in this study has potential as a tool for understanding factors contributing to dengue spread as well as planning and optimizing dengue control, including reducing the Ae. aegypti vector population and for estimating the effectiveness and cost-effectiveness of future vaccination programmes. This model could also be applied to other vector borne viral diseases such as chikungunya, also spread by Ae. aegypti and, by re-parameterisation of the vector sub-model, to dengue and chikungunya viruses spread by Aedes albopictus.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2334-14-447) contains supplementary material, which is available to authorized users.

Shifting patterns of Aedes aegypti fine scale spatial clustering in Iquitos, Peru

Background

Empiric evidence shows that Aedes aegypti abundance is spatially heterogeneous and that some areas and larval habitats produce more mosquitoes than others. There is a knowledge gap, however, with regards to the temporal persistence of such Ae. aegypti abundance hotspots. In this study, we used a longitudinal entomologic dataset from the city of Iquitos, Peru, to (1) quantify the spatial clustering patterns of adult Ae. aegypti and pupae counts per house, (2) determine overlap between clusters, (3) quantify the temporal stability of clusters over nine entomologic surveys spaced four months apart, and (4) quantify the extent of clustering at the household and neighborhood levels.

Methodologies/Principal Findings

Data from 13,662 household entomological visits performed in two Iquitos neighborhoods differing in Ae. aegypti abundance and dengue virus transmission was analyzed using global and local spatial statistics. The location and extent of Ae. aegypti pupae and adult hotspots (i.e., small groups of houses with significantly [p<0.05] high mosquito abundance) were calculated for each of the 9 entomologic surveys. The extent of clustering was used to quantify the probability of finding spatially correlated populations. Our analyses indicate that Ae. aegypti distribution was highly focal (most clusters do not extend beyond 30 meters) and that hotspots of high vector abundance were common on every survey date, but they were temporally unstable over the period of study.

Conclusions/Significance

Our findings have implications for understanding Ae. aegypti distribution and for the design of surveillance and control activities relying on household-level data. In settings like Iquitos, where there is a relatively low percentage of Ae. aegypti in permanent water-holding containers, identifying and targeting key premises will be significantly challenged by shifting hotspots of Ae. aegypti infestation. Focusing efforts in large geographic areas with historically high levels of transmission may be more effective than targeting Ae. aegypti hotspots.

We carried out a comprehensive study of the long-term trends in household-level Aedes aegypti spatial distribution within a well-defined urban area endemic for dengue virus. By using a dataset consisting of 13,662 household entomological visits performed in two neighborhoods in Iquitos, Peru, we quantified the ∼3 year spatial clustering patterns of Ae. aegypti among houses and the temporal persistence of vector abundance hotspots. Our results provide strong support for the conclusion that Ae. aegypti distribution is highly focal and that hotspots of high vector abundance at the level of small groups of houses are common, but temporally unstable. Results from our study have implications for understanding the spatio-temporal patterns of Ae. aegypti abundance and for the design of surveillance and control activities that are based on household-level entomological data.

Demographic and clinical features of suspected dengue and dengue haemorrhagic fever in the Northern Province of Sri Lanka, a region afflicted by an internal conflict for more than 30 years-a retrospective analysis

OBJECTIVES

The aim of this study was to determine the demographic, clinical, and notification data of suspected dengue fever (DF) and dengue hemorrhagic fever (DHF) cases admitted to Jaffna Teaching Hospital, Sri Lanka.

METHODS

The data were collected from bed head tickets of all patients presenting with clinically suspected DF/DHF from October 2009 to September 2010.

RESULTS

A total of 1085 clinically suspected DF/DHF cases were identified, with high numbers occurring during December 2009 to March 2010. The majority of the reported patients were females (n = 550, 50.7%) and approximately three-quarters of the patients (n = 797, 73.5%) were adults. All had fever, but fever spikes were noted in only 129 cases (11.9%; 95% confidence interval (CI) 10.1-13.9%). Over 50% of cases had vomiting (95% CI 47.5-53.5%). Haemorrhages were noted in 266 (24.5%), with gum bleeding in 99 patients (37.2%). Low white blood cell and platelet counts were noted in 27.1% and 85.6% of cases, respectively. Of the 1085 cases, only 24 (2.2%) were screened for dengue IgM/IgG and only 458 cases (42.2%) were notified to the Epidemiology Unit, Ministry of Health, Sri Lanka.

CONCLUSIONS

The absence of laboratory diagnosis and poor notification to the Epidemiology Unit were the major drawbacks noted.

Heterogeneous feeding patterns of the dengue vector, Aedes aegypti, on individual human hosts in rural Thailand

BACKGROUND

Mosquito biting frequency and how bites are distributed among different people can have significant epidemiologic effects. An improved understanding of mosquito vector-human interactions would refine knowledge of the entomological processes supporting pathogen transmission and could reveal targets for minimizing risk and breaking pathogen transmission cycles.

METHODOLOGY AND PRINCIPAL FINDINGS

We used human DNA blood meal profiling of the dengue virus (DENV) vector, Aedes aegypti, to quantify its contact with human hosts and to infer epidemiologic implications of its blood feeding behavior. We determined the number of different people bitten, biting frequency by host age, size, mosquito age, and the number of times each person was bitten. Of 3,677 engorged mosquitoes collected and 1,186 complete DNA profiles, only 420 meals matched people from the study area, indicating that Ae. aegypti feed on people moving transiently through communities to conduct daily business. 10-13% of engorged mosquitoes fed on more than one person. No biting rate differences were detected between high- and low-dengue transmission seasons. We estimate that 43-46% of engorged mosquitoes bit more than one person within each gonotrophic cycle. Most multiple meals were from residents of the mosquito collection house or neighbors. People ≤ 25 years old were bitten less often than older people. Some hosts were fed on frequently, with three hosts bitten nine times. Interaction networks for mosquitoes and humans revealed biologically significant blood feeding hotspots, including community marketplaces.

CONCLUSION AND SIGNIFICANCE

High multiple-feeding rates and feeding on community visitors are likely important features in the efficient transmission and rapid spread of DENV. These results help explain why reducing vector populations alone is difficult for dengue prevention and support the argument for additional studies of mosquito feeding behavior, which when integrated with a greater understanding of human behavior will refine estimates of risk and strategies for dengue control.

Exploratory space-time analysis of dengue incidence in Trinidad: a retrospective study using travel hubs as dispersal points, 1998-2004

BACKGROUND

Dengue is an acute arboviral disease responsible for most of the illness and death in tropical and subtropical regions. Over the last 25 years there has been increase epidemic activity of the disease in the Caribbean, with the co-circulation of multiple serotypes. An understanding of the space and time dynamics of dengue could provide health agencies with important clues for reducing its impact.

METHODS

Dengue Haemorrhagic Fever (DHF) cases observed for the period 1998-2004 were georeferenced using Geographic Information System software. Spatial clustering was calculated for individual years and for the entire study period using the Nearest Neighbor Index. Space and time interaction between DHF cases was determined using the Knox Test while the Nearest Neighbor Hierarchical method was used to extract DHF hot spots. All space and time distances calculated were validated using the Pearson r significance test.

RESULTS

Results shows that (1) a decrease in mean distance between DHF cases correlates with activity leading up to an outbreak, (2) a decrease in temporal distance between DHF cases leads to increased geographic spread of the disease, with an outbreak occurrence about every 2 years, and (3) a general pattern in the movement of dengue incidents from more rural to urban settings leading up to an outbreak with hotspot areas associated with transportation hubs in Trinidad.

CONCLUSION

Considering only the spatial dimension of the disease, results suggest that DHF cases become more concentrated leading up to an outbreak. However, with the additional consideration of time, results suggest that when an outbreak occurs incidents occur more rapidly in time leading to a parallel increase in the rate of distribution of the disease across space. The results of this study can be used by public health officers to help visualize and understand the spatial and temporal patterns of dengue, and to prepare warnings for the public. Dengue space-time patterns and hotspot detection will provide useful information to support public health officers in their efforts to control and predict dengue spread over critical hotspots allowing better allocation of resources.

Assessing dengue infection risk in the southern region of Taiwan: implications for control

Dengue, one of the most important mosquito-borne diseases, is a major international public health concern. This study aimed to assess potential dengue infection risk from Aedes aegypti in Kaohsiung and the implications for vector control. Here we investigated the impact of dengue transmission on human infection risk using a well-established dengue-mosquito-human transmission dynamics model. A basic reproduction number (R 0)-based probabilistic risk model was also developed to estimate dengue infection risk. Our findings confirm that the effect of biting rate plays a crucial role in shaping R 0 estimates. We demonstrated that there was 50% risk probability for increased dengue incidence rates exceeding 0.5-0.8 wk-1 for temperatures ranging from 26°C to 32°C. We further demonstrated that the weekly increased dengue incidence rate can be decreased to zero if vector control efficiencies reach 30-80% at temperatures of 19-32°C. We conclude that our analysis on dengue infection risk and control implications in Kaohsiung provide crucial information for policy-making on disease control.

Lineage II of Southeast Asian/American DENV-2 is associated with a severe dengue outbreak in the Peruvian Amazon

During 2010 and 2011, the Loreto region of Peru experienced a dengue outbreak of unprecedented magnitude and severity for the region. This outbreak coincided with the reappearance of dengue virus-2 (DENV-2) in Loreto after almost 8 years. Whole-genome sequence indicated that DENV-2 from the outbreak belonged to lineage II of the southeast Asian/American genotype and was most closely related to viruses circulating in Brazil during 2007 and 2008, whereas DENV-2 previously circulating in Loreto grouped with lineage I (DENV-2 strains circulating in South America since 1990). One amino acid substitution (NS5 A811V) in the 2010 and 2011 isolates resulted from positive selection. However, the 2010 and 2011 DENV-2 did not replicate to higher titers in monocyte-derived dendritic cells and did not infect or disseminate in a higher proportion of Aedes aegypti than DENV-2 isolates previously circulating in Loreto. These results suggest that factors other than enhanced viral replication played a role in the severity of this outbreak.

Towards a Hybrid Agent-based Model for Mosquito Borne Disease

Agent-based models (ABM) are used to simulate the spread of infectious disease through a population. Detailed human movement, demography, realistic business location networks, and in-host disease progression are available in existing ABMs, such as the Epidemic Simulation System (EpiSimS). These capabilities make possible the exploration of pharmaceutical and non-pharmaceutical mitigation strategies used to inform the public health community. There is a similar need for the spread of mosquito borne pathogens due to the re-emergence of diseases such as chikungunya and dengue fever. A network-patch model for mosquito dynamics has been coupled with EpiSimS. Mosquitoes are represented as a "patch" or "cloud" associated with a location. Each patch has an ordinary differential equation (ODE) mosquito dynamics model and mosquito related parameters relevant to the location characteristics. Activities at each location can have different levels of potential exposure to mosquitoes based on whether they are inside, outside, or somewhere in-between. As a proof of concept, the hybrid network-patch model is used to simulate the spread of chikungunya through Washington, DC. Results are shown for a base case, followed by varying the probability of transmission, mosquito count, and activity exposure. We use visualization to understand the pattern of disease spread.

Characterizing the likelihood of dengue emergence and detection in naïve populations

Background

Vector-borne disease transmission is dependent on the many nuances of the contact event between infectious and susceptible hosts. Virus acquisition from a viremic human to a susceptible mosquito is often assumed to be nearly perfect and almost always uniform across the infectious period. Dengue transmission models that have previously addressed variability in human to vector transmission dynamics do not account for the variation in infectiousness of a single individual, and subsequent infection of naïve mosquitoes. Understanding the contribution of this variability in human infectiousness is especially important in the context of introduction events where an infected individual carries the virus into a population of competent vectors. Furthermore, it could affect the ability to detect an epidemic (and the timing of detection) following introduction.

Methods

We constructed a stochastic, compartmental model to describe the heterogeneity of human viremia and calculate the probability of a successful introduction, taking into account the viremia level (and thus acquisition potential) of the index case on, and after, the day of introduction into a susceptible population and varying contact rates between the human and mosquito populations. We then compared the results of this model with those generated by a simpler model that has the same average infectiousness but only a single infectious class.

Results

We found that the infectivity of the index case as well as the contact rate affected the probability of emergence, but that contact rate had the most significant effect. We also found that the interaction between contact rate and the infectiousness of the index case affected the time to detection relative to the peak of the epidemic curve. Additionally, when compared to our model that accounts for variable infectiousness, a model with a single infectious class underestimates the probability of emergence and transmission intensity.

Conclusion

Understanding the interplay between individual human heterogeneity of infectiousness and the rate of contact with the vector population will be important when predicting the likelihood, detection, and magnitude of an outbreak.

Strengths and weaknesses of Global Positioning System (GPS) data-loggers and semi-structured interviews for capturing fine-scale human mobility: findings from Iquitos, Peru

Quantifying human mobility has significant consequences for studying physical activity, exposure to pathogens, and generating more realistic infectious disease models. Location-aware technologies such as Global Positioning System (GPS)-enabled devices are used increasingly as a gold standard for mobility research. The main goal of this observational study was to compare and contrast the information obtained through GPS and semi-structured interviews (SSI) to assess issues affecting data quality and, ultimately, our ability to measure fine-scale human mobility. A total of 160 individuals, ages 7 to 74, from Iquitos, Peru, were tracked using GPS data-loggers for 14 days and later interviewed using the SSI about places they visited while tracked. A total of 2,047 and 886 places were reported in the SSI and identified by GPS, respectively. Differences in the concordance between methods occurred by location type, distance threshold (within a given radius to be considered a match) selected, GPS data collection frequency (i.e., 30, 90 or 150 seconds) and number of GPS points near the SSI place considered to define a match. Both methods had perfect concordance identifying each participant's house, followed by 80-100% concordance for identifying schools and lodgings, and 50-80% concordance for residences and commercial and religious locations. As the distance threshold selected increased, the concordance between SSI and raw GPS data increased (beyond 20 meters most locations reached their maximum concordance). Processing raw GPS data using a signal-clustering algorithm decreased overall concordance to 14.3%. The most common causes of discordance as described by a sub-sample (n=101) with whom we followed-up were GPS units being accidentally off (30%), forgetting or purposely not taking the units when leaving home (24.8%), possible barriers to the signal (4.7%) and leaving units home to recharge (4.6%). We provide a quantitative assessment of the strengths and weaknesses of both methods for capturing fine-scale human mobility.

Time-varying, serotype-specific force of infection of dengue virus

Infectious disease models play a key role in public health planning. These models rely on accurate estimates of key transmission parameters such as the force of infection (FoI), which is the per-capita risk of a susceptible person being infected. The FoI captures the fundamental dynamics of transmission and is crucial for gauging control efforts, such as identifying vaccination targets. Dengue virus (DENV) is a mosquito-borne, multiserotype pathogen that currently infects ∼390 million people a year. Existing estimates of the DENV FoI are inaccurate because they rely on the unrealistic assumption that risk is constant over time. Dengue models are thus unreliable for designing vaccine deployment strategies. Here, we present to our knowledge the first time-varying (daily), serotype-specific estimates of DENV FoIs using a spline-based fitting procedure designed to examine a 12-y, longitudinal DENV serological dataset from Iquitos, Peru (11,703 individuals, 38,416 samples, and 22,301 serotype-specific DENV infections from 1999 to 2010). The yearly DENV FoI varied markedly across time and serotypes (0-0.33), as did daily basic reproductive numbers (0.49-4.72). During specific time periods, the FoI fluctuations correlated across serotypes, indicating that different DENV serotypes shared common transmission drivers. The marked variation in transmission intensity that we detected indicates that intervention targets based on one-time estimates of the FoI could underestimate the level of effort needed to prevent disease. Our description of dengue virus transmission dynamics is unprecedented in detail, providing a basis for understanding the persistence of this rapidly emerging pathogen and improving disease prevention programs.

Assessing the relationship between vector indices and dengue transmission: a systematic review of the evidence

BACKGROUND

Despite doubts about methods used and the association between vector density and dengue transmission, routine sampling of mosquito vector populations is common in dengue-endemic countries worldwide. This study examined the evidence from published studies for the existence of any quantitative relationship between vector indices and dengue cases.

METHODOLOGY/PRINCIPAL FINDINGS

From a total of 1205 papers identified in database searches following Cochrane and PRISMA Group guidelines, 18 were included for review. Eligibility criteria included 3-month study duration and dengue case confirmation by WHO case definition and/or serology. A range of designs were seen, particularly in spatial sampling and analyses, and all but 3 were classed as weak study designs. Eleven of eighteen studies generated Stegomyia indices from combined larval and pupal data. Adult vector data were reported in only three studies. Of thirteen studies that investigated associations between vector indices and dengue cases, 4 reported positive correlations, 4 found no correlation and 5 reported ambiguous or inconclusive associations. Six out of 7 studies that measured Breteau Indices reported dengue transmission at levels below the currently accepted threshold of 5.

CONCLUSIONS/SIGNIFICANCE

There was little evidence of quantifiable associations between vector indices and dengue transmission that could reliably be used for outbreak prediction. This review highlighted the need for standardized sampling protocols that adequately consider dengue spatial heterogeneity. Recommendations for more appropriately designed studies include: standardized study design to elucidate the relationship between vector abundance and dengue transmission; adult mosquito sampling should be routine; single values of Breteau or other indices are not reliable universal dengue transmission thresholds; better knowledge of vector ecology is required.

Recasting the theory of mosquito-borne pathogen transmission dynamics and control

Mosquito-borne diseases pose some of the greatest challenges in public health, especially in tropical and sub-tropical regions of the world. Efforts to control these diseases have been underpinned by a theoretical framework developed for malaria by Ross and Macdonald, including models, metrics for measuring transmission, and theory of control that identifies key vulnerabilities in the transmission cycle. That framework, especially Macdonald's formula for R₀ and its entomological derivative, vectorial capacity, are now used to study dynamics and design interventions for many mosquito-borne diseases. A systematic review of 388 models published between 1970 and 2010 found that the vast majority adopted the Ross–Macdonald assumption of homogeneous transmission in a well-mixed population. Studies comparing models and data question these assumptions and point to the capacity to model heterogeneous, focal transmission as the most important but relatively unexplored component in current theory. Fine-scale heterogeneity causes transmission dynamics to be nonlinear, and poses problems for modeling, epidemiology and measurement. Novel mathematical approaches show how heterogeneity arises from the biology and the landscape on which the processes of mosquito biting and pathogen transmission unfold. Emerging theory focuses attention on the ecological and social context for mosquito blood feeding, the movement of both hosts and mosquitoes, and the relevant spatial scales for measuring transmission and for modeling dynamics and control.

Determinants of heterogeneous blood feeding patterns by Aedes aegypti in Iquitos, Peru

BACKGROUND

Heterogeneous mosquito biting results in different individuals in a population receiving an uneven number of bites. This is a feature of many vector-borne disease systems that, if understood, could guide preventative control efforts toward individuals who are expected to contribute most to pathogen transmission. We aimed to characterize factors determining biting patterns of Aedes aegypti, the principal mosquito vector of dengue virus.

METHODOLOGY/PRINCIPAL FINDINGS

Engorged female Ae. aegypti and human cheek swabs were collected from 19 houses in Iquitos, Peru. We recorded the body size, age, and sex of 275 consenting residents. Movement in and out of the house over a week (time in house) and mosquito abundance were recorded on eight separate occasions in each household over twelve months. We identified the individuals bitten by 96 engorged mosquitoes over this period by amplifying specific human microsatellite markers in mosquito blood meals and human cheek swabs. Using a multinomial model assuming a saturating relationship (power), we found that, relative to other residents of a home, an individual's likelihood of being bitten in the home was directly proportional to time spent in their home and body surface area (p<0.05). A linear function fit the relationship equally well (ΔAIC<1).

CONCLUSIONS/SIGNIFICANCE

Our results indicate that larger people and those who spend more time at home are more likely to receive Ae. aegypti bites in their homes than other household residents. These findings are consistent with the idea that measurable characteristics of individuals can inform predictions of the extent to which different people will be bitten. This has implications for an improved understanding of heterogeneity in different people's contributions to pathogen transmission, and enhanced interventions that include the people and places that contribute most to pathogen amplification and spread.

Mapping population and pathogen movements

For most of human history, populations have been relatively isolated from each other, and only recently has there been extensive contact between peoples, flora and fauna from both old and new worlds. The reach, volume and speed of modern travel are unprecedented, with human mobility increasing in high income countries by over 1000-fold since 1800. This growth is putting people at risk from the emergence of new strains of familiar diseases, and from completely new diseases, while ever more cases of the movement of both disease vectors and the diseases they carry are being seen. Pathogens and their vectors can now move further, faster and in greater numbers than ever before. Equally however, we now have access to the most detailed and comprehensive datasets on human mobility and pathogen distributions ever assembled, in order to combat these threats. This short review paper provides an overview of these datasets, with a particular focus on low income regions, and covers briefly approaches used to combine them to help us understand and control some of the negative effects of population and pathogen movements.

Socially structured human movement shapes dengue transmission despite the diffusive effect of mosquito dispersal

For sexually and directly transmitted infectious diseases, social connections influence transmission because they determine contact between individuals. For pathogens that are indirectly transmitted by arthropod vectors, the movement of the vectors is thought to diminish the role of social connections. Results from a recent study of mosquito-borne dengue virus (DENV), however, indicate that human movement alone can explain significant spatial variation in urban transmission rates. Because movement patterns are structured by social ties, this result suggests that social proximity may be a good predictor of infection risk for DENV and other pathogens transmitted by the mosquito Aedes aegypti. Here we investigated the effect of socially structured movement on DENV transmission using a spatially explicit, agent-based transmission model. When individual movements overlap to a high degree within social groups we were able to recreate infection patterns similar to those detected in dengue-endemic, northeastern Peru. Our results are consistent with the hypothesis that social proximity drives fine-scale heterogeneity in DENV transmission rates, a result that was robust to the influence of mosquito dispersal. This heterogeneity in transmission caused by socially structured movements appeared to be hidden by the diffusive effect of mosquito dispersal in aggregated infection dynamics, which implies this heterogeneity could be present and active in real dengue systems without being easily noticed. Accounting for socially determined, overlapping human movements could substantially improve the efficiency and efficacy of dengue surveillance and disease prevention programs as well as result in more accurate estimates of important epidemiological quantities, such as R0.

Heterogeneity, mixing, and the spatial scales of mosquito-borne pathogen transmission

The Ross-Macdonald model has dominated theory for mosquito-borne pathogen transmission dynamics and control for over a century. The model, like many other basic population models, makes the mathematically convenient assumption that populations are well mixed; i.e., that each mosquito is equally likely to bite any vertebrate host. This assumption raises questions about the validity and utility of current theory because it is in conflict with preponderant empirical evidence that transmission is heterogeneous. Here, we propose a new dynamic framework that is realistic enough to describe biological causes of heterogeneous transmission of mosquito-borne pathogens of humans, yet tractable enough to provide a basis for developing and improving general theory. The framework is based on the ecological context of mosquito blood meals and the fine-scale movements of individual mosquitoes and human hosts that give rise to heterogeneous transmission. Using this framework, we describe pathogen dispersion in terms of individual-level analogues of two classical quantities: vectorial capacity and the basic reproductive number, . Importantly, this framework explicitly accounts for three key components of overall heterogeneity in transmission: heterogeneous exposure, poor mixing, and finite host numbers. Using these tools, we propose two ways of characterizing the spatial scales of transmission—pathogen dispersion kernels and the evenness of mixing across scales of aggregation—and demonstrate the consequences of a model's choice of spatial scale for epidemic dynamics and for estimation of , both by a priori model formulas and by inference of the force of infection from time-series data.

Pathogens transmitted by mosquitoes, such as malaria and dengue, are notorious for the biological complexity associated with how they are transmitted within local communities. Yet mathematical models for these pathogens, which are critical tools for making recommendations for control policy, are based around concepts originally designed to describe how molecules interact in chemical systems. To provide those interested in mosquito-borne diseases a more appropriate tool for modeling their transmission, we introduce a mathematical framework that is based on the spatial locations where mosquitoes lay eggs and feed on blood and how mosquitoes and hosts move about those locations. Analysis of this framework shows that the transmission contributions of different hosts and locations can be calculated, and that overall potential for transmission in a community depends on three concepts: heterogeneous exposure (some people bitten by mosquitoes more than others), poor mixing (non-random contacts between hosts and mosquitoes), and finite population sizes (each host can contribute at most one new infection towards the population total). Together, these factors determine critical levels of vaccination coverage to eliminate a pathogen and the spatial areas over which transmission should be modeled and studied in the field.

Direct feeding on dengue patients yields new insights into human-to-mosquito dengue virus transmission

Evaluation of: Nguyen MN, Duong TH, Trung VT et al. Host and viral features of human dengue cases shape the population of infected and infectious Aedes aegypti mosquitoes. Proc. Natl Acad. Sci. USA 110(22), 9072–9077 (2013). Dengue virus (DENV) is a mosquito-borne virus of immense and growing global health importance. Despite decades of research, many fundamental components of DENV biology remain poorly understood. The Nguyen et al. study shines new light on one such component: the relationship between DENV blood viremia and infectiousness to mosquitoes. By directly feeding mosquitoes on infected people, the authors identified the viremia levels required for mosquitoes to become infected with each of the four DENV serotypes. Based on these results, the authors then indicated that ambulatory dengue cases who did not visit a hospital had viremia levels sufficient to infect mosquitoes. In avoiding laboratory surrogates, this study has significantly improved our understanding of DENV with implications for modeling and vaccine development.

The effects of vector movement and distribution in a mathematical model of dengue transmission

BACKGROUND

Mathematical models have been used to study the dynamics of infectious disease outbreaks and predict the effectiveness of potential mass vaccination campaigns. However, models depend on simplifying assumptions to be tractable, and the consequences of making such assumptions need to be studied. Two assumptions usually incorporated by mathematical models of vector-borne disease transmission is homogeneous mixing among the hosts and vectors and homogeneous distribution of the vectors.

METHODOLOGY/PRINCIPAL FINDINGS

We explored the effects of mosquito movement and distribution in an individual-based model of dengue transmission in which humans and mosquitoes are explicitly represented in a spatial environment. We found that the limited flight range of the vector in the model greatly reduced its ability to transmit dengue among humans. A model that does not assume a limited flight range could yield similar attack rates when transmissibility of dengue was reduced by 39%. A model in which mosquitoes are distributed uniformly across locations behaves similarly to one in which the number of mosquitoes per location is drawn from an exponential distribution with a slightly higher mean number of mosquitoes per location. When the models with different assumptions were calibrated to have similar human infection attack rates, mass vaccination had nearly identical effects.

CONCLUSIONS/SIGNIFICANCE

Small changes in assumptions in a mathematical model of dengue transmission can greatly change its behavior, but estimates of the effectiveness of mass dengue vaccination are robust to some simplifying assumptions typically made in mathematical models of vector-borne disease.

Spatial clustering of Aedes aegypti related to breeding container characteristics in Coastal Ecuador: implications for dengue control

Mosquito management within households remains central to the control of dengue virus transmission. An important factor in these management decisions is the spatial clustering of Aedes aegypti. We measured spatial clustering of Ae. aegypti in the town of Borbón, Ecuador and assessed what characteristics of breeding containers influenced the clustering. We used logistic regression to assess the spatial extent of that clustering. We found strong evidence for juvenile mosquito clustering within 20 m and for adult mosquito clustering within 10 m, and stronger clustering associations for containers ≥ 40 L than those < 40 L. Aedes aegypti clusters persisted after adjusting for various container characteristics, suggesting that patterns are likely attributable to short dispersal distances rather than shared characteristics of containers in cluster areas. These findings have implications for targeting Ae. aegypti control efforts.

The complex relationship between weather and dengue virus transmission in Thailand

Using a novel analytical approach, weather dynamics and seasonal dengue virus transmission cycles were profiled for each Thailand province, 1983-2001, using monthly assessments of cases, temperature, humidity, and rainfall. We observed systematic differences in the structure of seasonal transmission cycles of different magnitude, the role of weather in regulating seasonal cycles, necessary versus optimal transmission "weather-space," basis of large epidemics, and predictive indicators that estimate risk. Larger epidemics begin earlier, develop faster, and are predicted at Onset change-point when case counts are low. Temperature defines a viable range for transmission; humidity amplifies the potential within that range. This duality is central to transmission. Eighty percent of 1.2 million severe dengue cases occurred when mean temperature was 27-29.5°C and mean humidity was > 75%. Interventions are most effective when applied early. Most cases occur near Peak, yet small reductions at Onset can substantially reduce epidemic magnitude. Monitoring the Quiet-Phase is fundamental in effectively targeting interventions pre-emptively.

An epidemic of dengue-1 in a remote village in rural Laos

In the Lao PDR (Laos), urban dengue is an increasingly recognised public health problem. We describe a dengue-1 virus outbreak in a rural northwestern Lao forest village during the cool season of 2008. The isolated strain was genotypically “endemic” and not “sylvatic,” belonging to the genotype 1, Asia 3 clade. Phylogenetic analyses of 37 other dengue-1 sequences from diverse areas of Laos between 2007 and 2010 showed that the geographic distribution of some strains remained focal overtime while others were dispersed throughout the country. Evidence that dengue viruses have broad circulation in the region, crossing country borders, was also obtained. Whether the outbreak arose from dengue importation from an urban centre into a dengue-naïve community or crossed into the village from a forest cycle is unknown. More epidemiological and entomological investigations are required to understand dengue epidemiology and the importance of rural and forest dengue dynamics in Laos.

Dengue disease is caused by a virus transmitted by mosquitoes. In Southeast Asia, where it is endemic, it represents a very important public health problem. Major outbreaks, including severe cases and death, occur every year. Two distinct transmission cycles have been described. Most common is the human-mosquito-human cycle observed throughout most tropical regions of the world, often associated with urban locations and always human habitations, often producing explosive outbreaks, whereas “sylvatic” dengue, genetically different, circulates in forest wild animals and has been reported to be able to infect humans. In the Lao PDR, a developing country where dengue is endemic, data on this disease are sparse. This study reports an unusual outbreak of dengue that occurred during the cold season in a village in a forested area. It also is the first extensive analysis of dengue virus nucleotide sequences, from 39 patients across the country, from Laos. Results suggest three patterns of dengue circulation in Laos: local transmission, transmission over the whole country, and transmission implicating bordering countries. The dengue virus isolated from patients in the forest village outbreak proved to be genetically similar to those found in urbanized areas throughout the country. More investigations are needed to understand the relationships between dengue in forested and urban areas.

Estimates of dengue force of infection in children in Colombo, Sri Lanka

Dengue is the most important vector-borne viral disease worldwide and a major cause of childhood fever burden in Sri Lanka, which has experienced a number of large epidemics in the past decade. Despite this, data on the burden and transmission of dengue virus in the Indian Subcontinent are lacking. As part of a longitudinal fever surveillance study, we conducted a dengue seroprevalence survey among children aged <12 years in Colombo, Sri Lanka. We used a catalytic model to estimate the risk of primary infection among seronegative children. Over 50% of children had IgG antibodies to dengue virus and seroprevalence increased with age. The risk of primary infection was 14.1% per year (95% CI: 12.7%–15.6%), indicating that among initially seronegative children, approximately 1 in 7 experience their first infection within 12 months. There was weak evidence to suggest that the force of primary infection could be lower for children aged 6 years and above. We estimate that there are approximately 30 primary dengue infections among children <12 years in the community for every case notified to national surveillance, although this ratio is closer to 100∶1 among infants. Dengue represents a considerable infection burden among children in urban Sri Lanka, with levels of transmission comparable to those in the more established epidemics of Southeast Asia.

Dengue is an increasing problem in the Asian subcontinent, but little research exists on dengue burden and transmission in this region. Dengue ranges from mild fever to pronounced circulatory shock and potentially death. However, clinical disease gives an incomplete picture of how much dengue is circulating, because many infections are asymptomatic. Presence of antibodies to dengue virus provides evidence of past infection. By studying how antibody prevalence changes with age, the force of infection can be estimated, a key measure of population transmission that quantifies the risk of a first infection among dengue-naive (seronegative) individuals. We estimated the force of dengue primary infection by applying a catalytic model to data from a serological study of children in Colombo, Sri Lanka. Over 70% of children experienced at least one infection by the age of 12 years, and the median age at infection was 4.7 years. Among dengue-naive children 14% can be expected to experience a dengue infection within 12 months. The high force of infection at young ages indicates a very high level of dengue virus transmission in this urban setting that is comparable with levels seen in other regions with well-established epidemics, including Southeast Asia and Latin America.

Performance of the tourniquet test for diagnosing dengue in Peru

The tourniquet test (TT) is a physical examination maneuver often performed on patients suspected of having dengue. It has been incorporated into dengue diagnostic guidelines and is used in clinical studies. However, little is known about TT performance characteristics in different patient types or epidemiologic conditions. In the dengue-endemic city of Iquitos, Peru, we performed TTs and dengue laboratory assays on 13,548 persons with febrile disease, recruited through either active (n = 1,095) or passive (n = 12,453) surveillance. The sensitivity was 52% and 56%, the specificity was 58% and 68%, the positive predictive value was 45% and 55%, and the negative predictive value was 64% and 69% for persons enrolled in active and passive surveillance, respectively. We demonstrated that the TT was more sensitive identifying dengue disease in women and those of younger age and that sensitivity increased the later a person came to a medical clinic for care.

Host and viral features of human dengue cases shape the population of infected and infectious Aedes aegypti mosquitoes

Dengue is the most prevalent arboviral disease of humans. The host and virus variables associated with dengue virus (DENV) transmission from symptomatic dengue cases (n = 208) to Aedes aegypti mosquitoes during 407 independent exposure events was defined. The 50% mosquito infectious dose for each of DENV-1-4 ranged from 6.29 to 7.52 log10 RNA copies/mL of plasma. Increasing day of illness, declining viremia, and rising antibody titers were independently associated with reduced risk of DENV transmission. High early DENV plasma viremia levels in patients were a marker of the duration of human infectiousness, and blood meals containing high concentrations of DENV were positively associated with the prevalence of infectious mosquitoes 14 d after blood feeding. Ambulatory dengue cases had lower viremia levels compared with hospitalized dengue cases but nonetheless at levels predicted to be infectious to mosquitoes. These data define serotype-specific viremia levels that vaccines or drugs must inhibit to prevent DENV transmission.

Using GPS technology to quantify human mobility, dynamic contacts and infectious disease dynamics in a resource-poor urban environment

Empiric quantification of human mobility patterns is paramount for better urban planning, understanding social network structure and responding to infectious disease threats, especially in light of rapid growth in urbanization and globalization. This need is of particular relevance for developing countries, since they host the majority of the global urban population and are disproportionally affected by the burden of disease. We used Global Positioning System (GPS) data-loggers to track the fine-scale (within city) mobility patterns of 582 residents from two neighborhoods from the city of Iquitos, Peru. We used ∼2.3 million GPS data-points to quantify age-specific mobility parameters and dynamic co-location networks among all tracked individuals. Geographic space significantly affected human mobility, giving rise to highly local mobility kernels. Most (∼80%) movements occurred within 1 km of an individual's home. Potential hourly contacts among individuals were highly irregular and temporally unstructured. Only up to 38% of the tracked participants showed a regular and predictable mobility routine, a sharp contrast to the situation in the developed world. As a case study, we quantified the impact of spatially and temporally unstructured routines on the dynamics of transmission of an influenza-like pathogen within an Iquitos neighborhood. Temporally unstructured daily routines (e.g., not dominated by a single location, such as a workplace, where an individual repeatedly spent significant amount of time) increased an epidemic's final size and effective reproduction number by 20% in comparison to scenarios modeling temporally structured contacts. Our findings provide a mechanistic description of the basic rules that shape human mobility within a resource-poor urban center, and contribute to the understanding of the role of fine-scale patterns of individual movement and co-location in infectious disease dynamics. More generally, this study emphasizes the need for careful consideration of human social interactions when designing infectious disease mitigation strategies, particularly within resource-poor urban environments.