Passive surveillance systems are widely used to monitor diseases occurrence over wide spatial areas due to their cost-effectiveness and integration into broadly distributed healthcare systems. However, such systems are generally associated with imperfect ascertainment of disease cases and with heterogeneous capture probabilities arising from factors such as differential access to care. Augmenting passive surveillance systems with other surveillance efforts provides a way to estimate the true number of incident cases. We develop a hierarchical modeling framework for analyzing data from multiple surveillance systems that allows for individual-level covariate-dependent heterogeneous capture probabilities, and borrows information across surveillance sites to improve estimation of the true number of incident cases. Inference is carried out via a two-stage Bayesian procedure. Simulation studies illustrated superior performance of the proposed approach with respect to bias, root mean square error, and coverage compared to a model that does not borrow information across sites. We applied the proposed model to data from three surveillance systems reporting pulmonary tuberculosis (PTB) cases in a major center of ongoing transmission in China. The analysis yielded bias-corrected estimates of PTB cases from the passive system and led to the identification of risk factors associated with PTB rates, as well as factors influencing the operating characteristics of the implemented surveillance systems.

Extrapolating landscape regression models for use in assessing vector-borne disease risk and other applications requires thoughtful evaluation of fundamental model choice issues. To examine implications of such choices, an analysis was conducted to explore the extent to which disparate landscape models agree in their epidemiological and entomological risk predictions when extrapolated to new regions. Agreement between six literature-drawn landscape models was examined by comparing predicted county-level distributions of either Lyme disease or Ixodes scapularis vector using Spearman ranked correlation. AUC analyses and multinomial logistic regression were used to assess the ability of these extrapolated landscape models to predict observed national data. Three models based on measures of vegetation, habitat patch characteristics, and herbaceous landcover emerged as effective predictors of observed disease and vector distribution. An ensemble model containing these three models improved precision and predictive ability over individual models. A priori assessment of qualitative model characteristics effectively identified models that subsequently emerged as better predictors in quantitative analysis. Both a methodology for quantitative model comparison and a checklist for qualitative assessment of candidate models for extrapolation are provided; both tools aim to improve collaboration between those producing models and those interested in applying them to new areas and research questions.

BACKGROUND:Associations between ambient particulate matter < 2.5 μm (PM2.5) and asthma morbidity have been suggested in previous epidemiologic studies but results are inconsistent for areas with lower PM2.5 levels. We estimated the associations between early-life short-term PM2.5 exposure and the risk of asthma or wheeze clinical encounters among Massachusetts children in the innovative Pregnancy to Early Life Longitudinal (PELL) cohort data linkage system. METHODS:We used a semi-bidirectional case-crossover study design with short-term exposure lags for asthma exacerbation using data from the PELL system. Cases included children up to 9 years of age who had a hospitalization, observational stay, or emergency department visit for asthma or wheeze between January 2001 and September 2009 (n = 33,387). Daily PM2.5 concentrations were estimated at a 4-km resolution using satellite remote sensing, land use, and meteorological data. We applied conditional logistic regression models to estimate adjusted odds ratios (ORs) and 95% confidence intervals (CI). We also stratified by potential effect modifiers. RESULTS:The median PM2.5 concentration among participants was 7.8 μg/m3 with an interquartile range of 5.9 μg/m3. Overall, associations between PM2.5 exposure and asthma clinical encounters among children at lags 0, 1 and 2 were close to the null value of OR = 1.0. Evidence of effect modification was observed by birthweight for lags 0, 1 and 2 (p < 0.05), and season of clinical encounter for lags 0 and 1 (p < 0.05). Children with low birthweight (LBW) (< 2500 g) had increased odds of having an asthma clinical encounter due to higher PM2.5 exposure for lag 1 (OR: 1.08 per interquartile range (IQR) increase in PM2.5; 95% CI: 1.01, 1.15). CONCLUSION:Asthma or wheeze exacerbations among LBW children were associated with short-term increases in PM2.5 concentrations at low levels in Massachusetts.

After publication of the article [1], it was brought to our attention that a number in Table 1 is incorrect.

Mass gatherings exacerbate infectious disease risks by creating crowded, high-contact conditions and straining the capacity of local infrastructure. While mass gatherings have been extensively studied in the context of epidemic disease transmission, the role of gatherings in incidence of high-burden, endemic infections has not been previously studied. Here, we examine diarrheal incidence among 17 communities in Esmeraldas, Ecuador, in relation to recurrent gatherings characterized using ethnographic data collected during and after the epidemiologic surveillance period (2004-2007). Using distributed-lag generalized estimating equations, adjusted for seasonality, trend, and heavy rainfall events, we found significant increases in diarrhea risk in host villages, peaking 2 weeks after an event's conclusion (incidence rate ratio, 1.21; confidence interval, adjusted for false coverage rate of ≤0.05: 1.02, 1.43). Stratified analysis revealed heightened risks associated with events where crowding and travel were most likely (2-week-lag incidence rate ratio, 1.51; confidence interval, adjusted for false coverage rate of ≤0.05: 1.09, 2.10). Our findings suggest that community-scale mass gatherings might play an important role in endemic diarrheal disease transmission and could be an important focus for interventions to improve community health in low-resource settings.

Poikilothermic disease vectors can respond to altered climates through spatial changes in both population size and phenology. Quantitative descriptors to characterize, analyze and visualize these dynamic responses are lacking, particularly across large spatial domains. In order to demonstrate the value of a spatially explicit, dynamic modeling approach, we assessed spatial changes in the population dynamics of Ixodes scapularis, the Lyme disease vector, using a temperature-forced population model simulated across a grid of 4 × 4 km cells covering the eastern United States, using both modeled (Weather Research and Forecasting (WRF) 3.2.1) baseline/current (2001-2004) and projected (Representative Concentration Pathway (RCP) 4.5 and RCP 8.5; 2057-2059) climate data. Ten dynamic population features (DPFs) were derived from simulated populations and analyzed spatially to characterize the regional population response to current and future climate across the domain. Each DPF under the current climate was assessed for its ability to discriminate observed Lyme disease risk and known vector presence/absence, using data from the US Centers for Disease Control and Prevention. Peak vector population and month of peak vector population were the DPFs that performed best as predictors of current Lyme disease risk. When examined under baseline and projected climate scenarios, the spatial and temporal distributions of DPFs shift and the seasonal cycle of key questing life stages is compressed under some scenarios. Our results demonstrate the utility of spatial characterization, analysis and visualization of dynamic population responses-including altered phenology-of disease vectors to altered climate.

The relationship between rainfall, especially extreme rainfall, and increases in waterborne infectious diseases is widely reported in the literature. Most of this research, however, has not formally considered the impact of exposure measurement error contributed by the limited spatiotemporal fidelity of precipitation data. Here, we evaluate bias in effect estimates associated with exposure misclassification due to precipitation data fidelity, using extreme rainfall as an example. We accomplished this via a simulation study, followed by analysis of extreme rainfall and incident diarrheal disease in an epidemiologic study in Ecuador. We found that the limited fidelity typical of spatiotemporal rainfall data sets biases effect estimates towards the null. Use of spatial interpolations of rain-gauge data or satellite data biased estimated health effects due to extreme rainfall (occurrence) and wet conditions (accumulated totals) downwards by 35%-45%. Similar biases were evident in the Ecuadorian case study analysis, where spatial incompatibility between exposed populations and rain gauges resulted in the association between extreme rainfall and diarrheal disease incidence being approximately halved. These findings suggest that investigators should pay greater attention to limitations in using spatially heterogeneous environmental data sets to assign exposures in epidemiologic research.

A typical challenge in air pollution epidemiology is to perform detailed exposure assessment for individuals for which health data are available. To address this problem, in the last few years, substantial research efforts have been placed in developing statistical methods or machine learning techniques to generate estimates of air pollution at fine spatial and temporal scales (daily, usually) with complete coverage. However, it is not clear how much the predicted exposures yielded by the various methods differ, and which method generates more reliable estimates. In this paper, we aim to address this gap by evaluating a variety of exposure modeling approaches, comparing their predictive performance. Using PM2.5 in year 2011 over the continental U.S. as a case study, we generate national maps of ambient PM2.5 concentration using: (i) ordinary least squares and inverse distance weighting; (ii) kriging; (iii) statistical downscaling models, that is, spatial statistical models that use the information contained in air quality model outputs; (iv) land use regression, that is, linear regression modeling approaches that leverage the information in Geographical Information System (GIS) covariates; and (v) machine learning methods, such as neural networks, random forests and support vector regression. We examine the various methods' predictive performance via cross-validation using Root Mean Squared Error, Mean Absolute Deviation, Pearson correlation, and Mean Spatial Pearson Correlation. Additionally, we evaluated whether factors such as, season, urbanicty, and levels of PM2.5 concentration (low, medium or high) affected the performance of the different methods. Overall, statistical methods that explicitly modeled the spatial correlation, e.g. universal kriging and the downscaler model, outperform all the other exposure assessment approaches regardless of season, urbanicity and PM2.5 concentration level. We posit that the better predictive performance of spatial statistical models over machine learning methods is due to the fact that they explicitly account for spatial dependence, thus borrowing information from neighboring observations. In light of our findings, we suggest that future exposure assessment methods for regional PM2.5 incorporate information from neighboring sites when deriving predictions at unsampled locations or attempt to account for spatial dependence.

Disease surveillance data are important for monitoring disease burden and occurrence, and for informing a wide range of efforts to improve population health. Surveillance for infectious diseases may be conducted passively, relying on reports from healthcare facilities, or actively, involving surveys of the population at risk. Passive surveillance typically provides wide spatial coverage, but is subject to biases arising from differences in care-seeking behavior, diagnostic practices, and under-reporting. Active surveillance minimizes these biases, but is typically constrained to small areas and subpopulations due to resource limitations. Methods based on linkage of individual records between passive and active surveillance datasets provide a means to estimate and correct for the biases of each system, leveraging the size and coverage of passive surveillance and the quality of data in active surveillance. We develop a spatial Bayesian hierarchical model for bias-correcting data from both systems to yield an improved estimate of disease measures after adjusting for under-ascertainment. We apply the framework to data from a passive and an active surveillance system for pulmonary tuberculosis (PTB) in Sichuan, China, and estimate the average sensitivity of the active surveillance system at 70% (95% credible interval: 62%, 78%), and the passive system at 30% (95% CI: 24%, 35%). Passive surveillance sensitivity exhibited considerable spatial variability, and was positively associated with a site's gross domestic product per capita. Bias-corrected estimates of county-level PTB prevalence in the province in 2010 identified regions in the southeast with the highest PTB burden, yielding different geographic priorities than previous reports.

Climate exerts complex influences on leptospirosis transmission, affecting human behavior, zoonotic host population dynamics, and survival of the pathogen in the environment. Here, we describe the spatiotemporal distribution of leptospirosis incidence reported to China’s National Infectious Disease Surveillance System from 2004–2014 in an endemic region in western China, and employ distributed lag models at annual and sub-annual scales to analyze its association with hydroclimatic risk factors and explore evidence for the potential role of a soil reservoir in the transmission of Leptospira spp. More than 97% of the 2,934 reported leptospirosis cases occurred during the harvest season between August and October, and most commonly affected farmers (83%). Using a distributed lag Poisson regression framework, we characterized incidence rate ratios (IRRs) associated with interquartile range increases in precipitation of 3.45 (95% confidence interval 2.57–4.64) over 0-1-year lags, and 1.90 (1.18–3.06) over 0-15-week lags. Adjusting for soil moisture decreased IRRs for precipitation at both timescales (yearly adjusted IRR: 1.05, 0.74–1.49; weekly adjusted IRR: 1.36, 0.72–2.57), suggesting precipitation effects may be mediated through soil moisture. Increased soil moisture was positively associated with leptospirosis at both timescales, suggesting that the survival of pathogenic Leptospira spp. in moist soils may be a critical control on harvest-associated leptospirosis transmission in the study region. These results support the hypothesis that soils may serve as an environmental reservoir and may play a significant yet underrecognized role in leptospirosis transmission.