Catálogo de publicaciones - libros

A significant proportion of human diseases, spanning the gamut from viral respiratory disease to arthropod-borne macroparasitic infections of the blood, exhibit distinct and stable seasonal patterns of incidence. Traditional statistical methods for the evaluation of seasonal time-series data emphasize the removal of these seasonal variations to be able to examine non-periodic, and therefore unexpected, or ‘excess’, incidence. Here, the authors present an alternate methodology emphasizing the retention and quantification of exactly these seasonal fluctuations, explicitly examining the changes in severity and timing of the expected seasonal outbreaks over several years. Using a PCRconfirmed Influenza time series as a case study, the authors provide an example of this type of analysis and discuss the potential uses of this method, including the comparison of differing sources of surveillance data. The requirements for statistical and practical validity, and considerations of data collection, reporting and analysis involved in the appropriate applications of the methods proposed are also discussed in detail.

This paper reviews the results of a study into combining evidence from multiple streams of surveillance data in order to improve timeliness and specificity of detection of bio-events. In the experiments we used three streams of real food- and agriculture-safety related data that is being routinely collected at slaughter houses across the nation, and which carry mutually complementary information about potential outbreaks of bio-events. The results indicate that: (1) Non-specific aggregation of p-values produced by event detectors set on individual streams of data can lead to superior detection power over that of the individual detectors, and (2) Design of multi-stream detectors tailored to the particular characteristics of the events of interest can further improve timeliness and specificity of detection. In a practical setup, we recommend combining a set of specific multi-stream detectors focused on individual types of predictable and definable scenarios of interest, with non-specific multi-stream detectors, to account for both anticipated and emerging types of bio-events.

To better support the surveillance of infectious disease and epidemic outbreaks by public health professionals, we design and implement BioPortal, an advanced Web-based system for cross-jurisdictional information sharing and integration. In this paper, we report two empirical studies that evaluate the outcomes, task performance efficiency, user information satisfaction, and usability associated with BioPortal. Overall, our results suggest that the use of BioPortal can improve users’ surveillance performance as measured by analysis accuracy and efficiency (i.e., the amount of time required to complete an analysis task). Our subjects were highly satisfied with the information support of BioPortal and considered it reasonably usable. Our evaluation findings show the effectiveness and value of BioPortal and, at the same time, shed light on several areas where its design can further improve.

The negative impact of infectious disease on contemporary society has the potential to be considerably greater than in decades past due to the growing interdependence among nations of the world. In the absence of worldwide public health standards-based networks, the ability to monitor and respond quickly to such outbreaks is limited. In order to tackle such threats, IBM Haifa Research Lab and IBM Almaden Research Lab developed a PHAD implementation which consists of an information technology infrastructure for the public health community leveraging the Integrating the Healthcare Enterprise (IHE) initiative and important standards. This system enables sharing of data generated at clinical and public health institutions across proprietary systems and political borders. The ability to share public health data electronically paves the way for sophisticated and advanced analysis tools to visualize the population health, detect outbreaks, determine the effectiveness of policy, and perform forecast modeling.

The Influenza Data Summary (IDS) is a tool that provides a unified view of influenza activity in the United States. It currently incorporates data from portions of the U.S. Influenza Surveillance System and BioSense. The IDS allows users to customize dashboards, interactive maps, and graphs from each of these data sources. The purpose of this paper is to provide an overview of the IDS and to discuss current features and future plans for improvement.

The paper presents a description of the FMD BioPortal biosurveillance system ( http://fmd.ucdavis.edu/bioportal/ ) that is currently operating to capture, analyze, and disseminate global data on foot-and-mouth (FMD) disease. The FMD BioPortal makes available to users world-wide FMD-related data from the Institute for Animal Health at Pirbright, England. The system’s tools include those for tabulating and graphing data, performing spatio-temporal cluster analysis of outbreak cases of FMD, and analyzing genomic changes in FMD viruses. The FMD BioPortal also includes the FMD News ( http://fmd.ucdavis. edu/index.php?id=1 ), which is a near real time web search to identify and capture FMD-related news items appearing worldwide. Major systems components include a communication backbone for secure, real-time data transfer, a data analysis module that can run analytical programs to assess spatial-temporal clustering, and an interactive visualization tool for integrated analysis and display of epidemiological and genomic data.

By providing useful measures of the outcome of an influenza pandemic, the utilization of predictive mathematical models can be an extremely valuable tool within crisis preparedness exercises. We discuss our experiences with developing and implementing such a simulation model for use within a regional crisis preparedness exercise, and make recommendations for maximizing the utility of predictive mathematical epidemiological models in general for future exercises.

Researchers have studied many potential sources of data for biosurveillance but have tended to focus on ambulatory visits and over-the-counter pharmaceutical sales. Data from electronic prescribing (e-prescribing) systems in an ambulatory setting have not been evaluated critically, but they may provide valuable data for surveillance. In this paper we evaluate the utility of e-prescribing data for surveillance of respiratory infections. Demographic data were analyzed to determine the differences between patients in an e-prescribing system and the general population. Correlation analysis was performed on the time-series for common respiratory tract antibiotics and the time-series for respiratory tract infection incidence. Demographic data showed a strong bias towards older people in the e-prescribing system when compared to the general population. The analysis also showed that a subset of antibiotics are highly correlated with respiratory tract indications (0.84, p<0.0001, 95% CI 0.73-0.90). The over-representation of higher age groups in the electronic prescribing system suggest that data from such systems may be suitable for observing trends in chronic conditions or infectious conditions more common in the elderly. The results also suggest that a set of antibiotics can be identified that reflect the incidence of respiratory tract infections.

We have combined two methods to detect anomalies in a time series – in this case in emergency department visit data. The n-gram method applies an existing ICD classifier to a set of emergency department (ED) visits for which both the chief complaint (CC) and ICD code are known. A collection of CC substrings (or n-grams), with associated probabilities, are automatically generated from the training data. This information becomes a CC classifier which is then used to find a classification probability for each patient. The output of this classifier can be used to build volume predictions for a syndromic group or can be combined with a selected threshold to provide syndromic determinations on a per-patient basis. Once the daily volume predictions have been calculated using the n-grams, the HWR anomaly detection algorithm is applied, which alerts both for unusual values and for changes in the overall behavior of the time series in question. The earliest alert was generated by the series of volume predicted by flu n-grams as a proportion of total daily visits.

Syndromic Surveillance is typically a system used for early detection of bioterrorism attacks, pandemic flu or other emerging diseases, which monitors symptoms of outpatients or is conducted in the Emergency Department. However, if we monitor symptoms of inpatients, we can apply Syndromic Surveillance to early detection of nosocomial infection. To test this possibility, we constructed and are performing a Syndromic Surveillance System for inpatients who have fever, respiratory symptoms, diarrhea, vomiting or rash. We will then evaluate its statistical properties and its usefulness.