Gregory E. Glass

Spatiotemporal Transmission Dynamics of Hemorrhagic Fever with Renal Syndrome in China, 2005–2012

Background Hemorrhagic fever with renal syndrome (HFRS) is a rodent-borne disease caused by many serotypes of hantaviruses. In China, HFRS has been recognized as a severe public health problem with 90% of the total reported cases in the world. This study describes the spatiotemporal dynamics of HFRS cases in China and identifies the regions, time, and populations at highest risk, which could help the planning and implementation of key preventative measures. Methods Data on all reported HFRS cases at the county level from January 2005 to December 2012 were collected from Chinese Center for Disease Control and Prevention. Geographic Information System-based spatiotemporal analyses including Local Indicators of Spatial Association and Kulldorffs space-time scan statistic were performed to detect local high-risk space-time clusters of HFRS in China. In addition, cases from high-risk and low-risk counties were compared to identify significant demographic differences. Results A total of 100,868 cases were reported during 2005–2012 in mainland China. There were significant variations in the spatiotemporal dynamics of HFRS. HFRS cases occurred most frequently in June, November, and December. There was a significant positive spatial autocorrelation of HFRS incidence during the study periods, with Morans I values ranging from 0.46 to 0.56 (P<0.05). Several distinct HFRS cluster areas were identified, mainly concentrated in northeastern, central, and eastern of China. Compared with cases from low-risk areas, a higher proportion of cases were younger, non-farmer, and floating residents in high-risk counties. Conclusions This study identified significant space-time clusters of HFRS in China during 2005–2012 indicating that preventative strategies for HFRS should be particularly focused on the northeastern, central, and eastern of China to achieve the most cost-effective outcomes.

Disparities in influenza mortality and transmission related to sociodemographic factors within Chicago in the pandemic of 1918

Significance The pervasiveness of influenza among humans and its rapid spread during pandemics create a false sense that all humans are affected equally. In this work, we show that neighborhood-level social determinants were associated with greater burdens of pandemic influenza in 1918 and several other diseases in a major US city. We show that literacy, homeownership, and unemployment were associated with cumulative influenza mortality as well as measures of the speed of transmission using a unique dataset describing the home location and week of death of individuals who died during the influenza pandemic in 1918. Our results suggest that, similar to other infectious diseases, social disparities should be a focus of research and public health response in future pandemics. Social factors have been shown to create differential burden of influenza across different geographic areas. We explored the relationship between potential aggregate-level social determinants and mortality during the 1918 influenza pandemic in Chicago using a historical dataset of 7,971 influenza and pneumonia deaths. Census tract-level social factors, including rates of illiteracy, homeownership, population, and unemployment, were assessed as predictors of pandemic mortality in Chicago. Poisson models fit with generalized estimating equations (GEEs) were used to estimate the association between social factors and the risk of influenza and pneumonia mortality. The Poisson model showed that influenza and pneumonia mortality increased, on average, by 32.2% for every 10% increase in illiteracy rate adjusted for population density, homeownership, unemployment, and age. We also found a significant association between transmissibility and population density, illiteracy, and unemployment but not homeownership. Lastly, analysis of the point locations of reported influenza and pneumonia deaths revealed fine-scale spatiotemporal clustering. This study shows that living in census tracts with higher illiteracy rates increased the risk of influenza and pneumonia mortality during the 1918 influenza pandemic in Chicago. Our observation that disparities in structural determinants of neighborhood-level health lead to disparities in influenza incidence in this pandemic suggests that disparities and their determinants should remain targets of research and control in future pandemics.

Assessment of the severity of Ebola virus disease in Sierra Leone in 2014-2015.

The current Ebola virus disease (EVD) epidemic in West Africa is unprecedented in scale, and Sierra Leone is the most severely affected country. The case fatality risk (CFR) and hospitalization fatality risk (HFR) were used to characterize the severity of infections in confirmed and probable EVD cases in Sierra Leone. Proportional hazards regression models were used to investigate factors associated with the risk of death in EVD cases. In total, there were 17 318 EVD cases reported in Sierra Leone from 23 May 2014 to 31 January 2015. Of the probable and confirmed EVD cases with a reported final outcome, a total of 2536 deaths and 886 recoveries were reported. CFR and HFR estimates were 74·2% [95% credibility interval (CrI) 72·6-75·5] and 68·9% (95% CrI 66·2-71·6), respectively. Risks of death were higher in the youngest (0-4 years) and oldest (⩾60 years) age groups, and in the calendar month of October 2014. Sex and occupational status did not significantly affect the mortality of EVD. The CFR and HFR estimates of EVD were very high in Sierra Leone.

Malaria elimination in Botswana, 2012–2014: achievements and challenges

BackgroundBotswana significantly reduced its malaria burden between 2000 and 2012. Incidence dropped from 0.99 to 0.01xa0% and deaths attributed to malaria declined from 12 to 3. The country initiated elimination strategies in October 2012. We examine the progress and challenges during implementation and identify future needs for a successful program in Botswana.MethodsA national, rapid notification and response strategy was developed. Cases detected through the routine passive surveillance system at health facilities were intended to initiate screening of contacts around a positive case during follow up. Positive cases were reported to district health management teams to activate district rapid response teams (DRRT). The health facility and the DRRT were to investigate the cases, and screen household members within 100xa0m of case households within 48xa0h of notification using rapid diagnostic tests (RDT) and microscopy. Positive malaria cases detected in health facilities were used for spatial analysis.ResultsThere were 1808 malaria cases recorded in Botswana during 26xa0months from October, 2012 to December, 2014. Males were more frequently infected (59xa0%) than females. Most cases (60xa0%) were reported from Okavango district which experienced an outbreak in 2013 and 2014. Among the factors creating challenges for malaria eradication, only 1148 cases (63.5xa0%) were captured by the required standardized notification forms. In total, 1080 notified cases were diagnosed by RDT. Of the positive malaria cases, only 227 (12.6xa0%) were monitored at the household level. One hundred (8.7xa0%) cases were associated with national or transnational movement of patients. Local movements of infected individuals within Botswana accounted for 31 cases while 69 (6.01xa0%) cases were imported from other countries. Screening individuals in and around index households identified 37 additional, asymptomatic infections. Oscillating, sporadic and new malaria hot-spots were detected in Botswana during the study period.ConclusionBotswana’s experience shows some of the practical challenges of elimination efforts. Among them are the substantial movements of human infections within and among countries, and the persistence of asymptomatic reservoir infections. Programmatically, challenges include improving the speed of communicating and improving the thoroughness when responding to newly identified cases. The country needs further sustainable interventions to target infections if it is to successfully achieve its elimination goal.

Subclinical Plasmodium falciparum infections act as year-round reservoir for malaria in the hypoendemic Chittagong Hill districts of Bangladesh

OBJECTIVESnAn analysis of the risk factors and seasonal and spatial distribution of individuals with subclinical malaria in hypoendemic Bangladesh was performed.nnnMETHODSnFrom 2009 to 2012, active malaria surveillance without regard to symptoms was conducted on a random sample (n=3971) and pregnant women (n=589) during a cohort malaria study in a population of 24000.nnnRESULTSnThe overall subclinical Plasmodium falciparum malaria point prevalence was 1.0% (n=35), but was 3.2% (n=18) for pregnant women. The estimated incidence was 39.9 per 1000 person-years for the overall population. Unlike symptomatic malaria, with a marked seasonal pattern, subclinical infections did not show a seasonal increase during the rainy season. Sixty-nine percent of those with subclinical P. falciparum infections reported symptoms commonly associated with malaria compared to 18% without infection. Males, pregnant women, jhum cultivators, and those living closer to forests and at higher elevations had a higher prevalence of subclinical infection.nnnCONCLUSIONSnHypoendemic subclinical malaria infections were associated with a number of household and demographic factors, similar to symptomatic cases. Unlike clinical symptomatic malaria, which is highly seasonal, these actively detected infections were present year-round, made up the vast majority of infections at any given time, and likely acted as reservoirs for continued transmission.

Spatial-Temporal Distribution of Hantavirus Rodent-Borne Infection by Oligoryzomys fulvescens in the Agua Buena Region - Panama

Background Hotspot detection and characterization has played an increasing role in understanding the maintenance and transmission of zoonotic pathogens. Identifying the specific environmental factors (or their correlates) that influence reservoir host abundance help increase understanding of how pathogens are maintained in natural systems and are crucial to identifying disease risk. However, most recent studies are performed at macro-scale and describe broad temporal patterns of population abundances. Few have been conducted at a microscale over short time periods that better capture the dynamical patterns of key populations. These finer resolution studies may better define the likelihood of local pathogen persistence. This study characterizes the landscape distribution and spatio-temporal dynamics of Oligoryzomys fulvescens (O. fulvescens), an important mammalian reservoir in Central America. Methods Information collected in a longitudinal study of rodent populations in the community of Agua Buena in Tonosí, Panama, between April 2006 and December 2009 was analyzed using non-spatial analyses (box plots) and explicit spatial statistical tests (correlograms, SADIE and LISA). A 90 node grid was built (raster format) to design a base map. The area between the nodes was 0.09 km2 and the total study area was 6.43 km2 (2.39 x 2.69 km). The temporal assessment dataset was divided into four periods for each year studied: the dry season, rainy season, and two months-long transitions between seasons (the months of April and December). Results There were heterogeneous patterns in the population densities and degrees of dispersion of O. fulvescens that varied across seasons and among years. The species typically was locally absent during the late transitional months of the season, and re-established locally in subsequent years. These populations re-occurred in the same area during the first three years but subsequently re-established further south in the final year of the study. Spatial autocorrelation analyses indicated local populations encompassed approximately 300–600 m. The borders between suitable and unsuitable habitats were sharply demarcated over short distances. Conclusion Oligoryzomys fulvescens showed a well-defined spatial pattern that evolved over time, and led to a pattern of changing aggregation. Thus, hot spots of abundance showed a general shifting pattern that helps explain the intermittent risk from pathogens transmitted by this species. This variation was associated with seasonality, as well as anthropogenic pressures that occurred with agricultural activities. These factors help define the characteristics of the occurrence, timing, intensity and duration of synanthropic populations affected by human populations and, consequently, possible exposure that local human populations experience.

Estimating the Geographic Distribution of Host-Seeking Adult Amblyomma americanum (Acari: Ixodidae) in Florida

Abstract The lone star tick, Amblyomma americanum (Linneaus) (Acari:Ixodidae), is the most commonly reported human-biting tick in the southeastern United States and is a vector for several human and livestock pathogens. Although it is endemic to Florida, little is known about the ecological preferences and current spatial distribution within the state. Using occurrence records of adult A. americanum collected between August 2015 and September 2016, a logistic regression model was used to estimate environmental associations, as well as to predict the distribution of the tick at a one hectare resolution. Occurrence of adult lone star ticks was associated with land cover and bioclimatic variables, namely the presence of forested areas and precipitation seasonality. The estimated spatial distribution indicated that central and northern regions show greater suitability than the southern half of the state. Furthermore, areas predicted to be suitable for the species decreases from north to south with very little area deemed suitable in the far southern reaches of the state. High heterogeneity in the distribution of suitable habitat has implications for the distribution of tick-borne disease cases in the state.The subcounty resolution of the estimated distribution is an improvement over distributions currently published and may better inform the public and state or federal agencies of potential risk of exposure to A. americanum and its associated pathogens.

Machine learning approaches in GIS-based ecological modeling of the sand fly Phlebotomus papatasi, a vector of zoonotic cutaneous leishmaniasis in Golestan province, Iran

The distribution and abundance of Phlebotomus papatasi, the primary vector of zoonotic cutaneous leishmaniasis in most semi-/arid countries, is a major public health challenge. This study compares several approaches to model the spatial distribution of the species in an endemic region of the disease in Golestan province, northeast of Iran. The intent is to assist decision makers for targeted interventions. We developed a geo-database of the collected Phlebotominae sand flies from different parts of the study region. Sticky paper traps coated with castor oil were used to collect sand flies. In 44 out of 142 sampling sites, Ph. papatasi was present. We also gathered and prepared data on related environmental factors including topography, weather variables, distance to main rivers and remotely sensed data such as normalized difference vegetation cover and land surface temperature (LST) in a GIS framework. Applicability of three classifiers: (vanilla) logistic regression, random forest and support vector machine (SVM) were compared for predicting presence/absence of the vector. Predictive performances were compared using an independent dataset to generate area under the ROC curve (AUC) and Kappa statistics. All three models successfully predicted the presence/absence of the vector, however, the SVM classifier (Accuracyu2009=u20090.906, AUCu2009=u20090.974, Kappau2009=u20090.876) outperformed the other classifiers on predicting accuracy. Moreover, this classifier was the most sensitive (85%), and the most specific (93%) model. Sensitivity analysis of the most accurate model (i.e. SVM) revealed that slope, nighttime LST in October and mean temperature of the wettest quarter were among the most important predictors. The findings suggest that machine learning techniques, especially the SVM classifier, when coupled with GIS and remote sensing data can be a useful and cost-effective way for identifying habitat suitability of the species.

Reply to Shanks and Brundage: Many plausible mechanisms of pandemic mortality disparities

In our recently published paper (1), we find evidence that several metrics of socioeconomic status were associated with influenza and pneumonia mortality during the 1918 pandemic. We agree with Shanks and Brundage (2) that there are many possible causal mechanisms behind this association and appreciate the hypotheses that they add to the several we discussed.nnSecondary bacterial infections, particularly those secondary bacterial infections resulting in pneumonia, have been implicated in a number of deaths during the 1918 pandemic period (3⇓–5). Our dataset includes deaths attributed both to … nn[↵][1]1To whom correspondence should be addressed. Email: datc{at}ufl.edu.nn [1]: #xref-corresp-1-1

A 24-year exploratory spatial data analysis of Lyme disease incidence rate in Connecticut, USA

Despite efforts to control Lyme disease in Connecticut, USA, it remains endemic in many towns, posing a heavy burden. We examined changes in the spatial distribution of significant spatial clusters of Lyme disease incidence rates at the town level from 1991 to 2014 as an approach for targeted interventions. Lyme disease data were grouped into four discrete time periods and incidence rates were smoothed with Empirical Bayes estimation in GeoDa. Local clustering was measured using a local indicator of spatial autocorrelation (LISA). Elliptic spatial scan statistics (SSS) in different shapes and directions were also performed in SaTScan. The accuracy of these two cluster detection methods was assessed and compared for sensitivity, specificity, and overall accuracy. There was significant clustering during each period and significant clusters persisted predominantly in western and eastern parts of the state. Generally, the SSS method was more sensitive, while LISA was more specific with higher overall accuracy in identifying clusters. Even though the location of clusters changed over time, some towns were persistently (across all four periods) identified as clusters in LISA and their neighbouring towns (three of four periods) in SSS suggesting these regions should be prioritized for targeted interventions.