Last update: 2/15/06
March 7-8, 2006, Tucson, Arizona
 
 

Location

All sessions will be held in Room 208 and 208a of McClelland Hall on the campus of The University of Arizona (1130 East Helen Street, Tucson, Arizona, 85721).  See a campus map here.

 
   

Tentative Schedule

Mon., March 6, 2006 

 

Arrival 
 
Tue., March 7, 2006  

7:30 am
8:30 am
12:00 noon
1:00 pm
5:00 pm
6:30 or 7:00 pm

 

 

Breakfast (informal)
Morning session begins
Lunch
Afternoon session begins
Adjournment for day
Dinner (TBA)
Wed., March 8, 2006 7:30 am
8:30 am
12:00 noon
Breakfast (informal)
Morning session begins
Lunch (will also be provided “to go” for those who have to catch a plane)

 

 
   

Speakers and Abstracts

Information from speakers is still being received and will be posted here as it arrives:

 
   

Coleen Burgess
biography

“Pandemic Influenza Modeling in Structured Populations”

The researchers at MathEcology are developing a mathematical model capable of assessing the impact that pandemic influenza may have on specific subpopulations within the United States and abroad. This mathematical model is intended to track disease transmission and its effect on personnel and equipment resources, and offer alerts and intervention recommendations appropriate to the specific subpopulation and setting.

 Eileen Koski1, Greg Kinne2, Patrick Tellez3, Colin Goodall4, Simon Tse5, Jake Geller1
1Quest Diagnostics Incorporated, Lyndhurst, NJ, 2AT&T Government Solutions, Vienna, VA, 3MedPlus, A Quest Diagnostics Company, Mason, OH, 4AT&T Labs-Research, Middletown, NJ, 5AT&T Labs-Research, Florham Park, NJ

“National Biosurveillance and Electronic Laboratory Data”

Events of recent years, ranging from the emergence of West Nile Encephalitis in the United States, to the anthrax attacks that followed 9/11 and, most recently, to the emergence of human cases of influenza H5N1 to Europe, have heightened awareness of the challenges to existing disease surveillance systems.  These challenges include the need for near real-time data that is sufficiently sensitive and specific to detect conditions of concern and that provides representative national coverage.  The system must be secure and scalable, with respect to both quantity of data and speed of analysis.  Analytic capabilities must include statistical anomaly detection, alerting and notification, as well as case management.  Furthermore, the system must be multi-modal since infectious disease outbreaks, bioterrorism or environmental events, and pandemics all have distinct and different characteristics.  Our collaborative work at Quest Diagnostics Incorporated, AT&T Government Solutions and AT&T Labs-Research, uses laboratory test orders and results in combination with a deep and broad array of statistical and mathematical techniques, as well as sophisticated case management technologies, in a biosurveillance system that is differentiated by how we address these challenges.

 
   

Kathleen Carley
biography

“BioDefense through City Level Multi-Agent Modeling of Bio and Chemical Threats"

A basic model for assessing the potential impact of epidemics and biological attacks on cities in the United States is described as are implications for disease spread.  The models is a multi-agent dynamic-network model at the city level that can be used for examining the impact of epidemics, weaponized biological attacks, and chemical spills on American cities.  This model facilitates city level defense planning and training for various bio- and chemical scenarios.  It is configured using real data streams such as census information, military base boundaries, hospital and other staging area locations, and weather information.  Agents in the model engage in various activities such as going to work, the doctor, the emergency room, restaurants, sporting events, pharmacies, getting infected, becoming contagious, becoming ill, communicating, etc.  Behavior is keyed off of socio-demographic characteristics, social network influences, dynamic health status, time of day, day of week, month, year, weather, etc., and infectious disease transmission between agents is constrained by the time-variable spatial proximity of infectious agents to susceptible agents.  Simulations have been run for five cities, including San Diego and Norfolk, with baseline data on 61 diseases including various weaponized and potentially epidemic pathogens such as anthrax, smallpox, and SARS.

 
   

Mary P. Derby, RN, MS, MPH
biography

"Public Health Surveillance of Foodborne Illnesses Using a Novel Dataset"

Background:  Poison Control Centers (PCCs) provide real-time data that could enhance public health surveillance systems for foodborne disease outbreaks including those produced by deliberate contamination.
Methods:  A search of a PCC database identified callers with gastrointestinal symptoms due to suspected foodborne illnesses with reported illnesses during January 1—March 31, 2000.    For each call, the PCC call data were reviewed to determine if study criteria for a predefined diarrhea/gastroenteritis syndrome were met, and then compared to the PCC’s coding of the call. PCC calls also were evaluated using zip code, age, gender and date of onset of symptoms to determine if they matched any laboratory-confirmed cases reported to a county health department.
Results:  There was high agreement between an independent review of call data and the PCC’s classification of the call. When calls and cases were compared, there was only one potential match. 
Conclusion:  While no confirmatory diagnostic information was available, PCC calls and county health department cases appear to be two independent data sets.  PCCs may provide a useful addition to syndromic surveillance data reported to public health agencies for the early detection of foodborne disease outbreaks. Prospective surveillance will now be developed collaboratively.

 
   
Dan Desmond
biography

"BioPortal - Two Years Later: The Loomis Project - Regional Response and Preparedness"

Two years after the original BioPortal effort for the NSF, a number of government, non-government, and private sector organizations have used the basic concepts and constructs of the BioPortal product in demonstration projects, commercial systems, and surveillance and response pilots with great success.

The result of these activities is "The Loomis Project" - a proposed regional pilot to provide near real-time situational awareness of pandemic response, with dynamic analytical database updates and "reach-back" capability. The Loomis Project is intended to improve response in real-time across local, regional, and state partners, as well as federal organizations/resources, such as HHS, CDC, Indian Health, Veterans Affairs, and Military Health Services.  Planned data sources include mass casualty, in-transit triage, patient/resource tracking, ER capacity/diversion, lab capacity/diversion, vaccine/prophylaxis and adverse events/treatment follow up.

This presentation will provide a quick overview of significant advances in BioPortal architecture from the original NSF project, beginning with the HIMSS/IHE-2004 Public Health Interoperability Demonstration and including current implementations of Service-Oriented Architectures and distributed, web-based systems for clinical response. New Technologies included in The Loomis Project will be overviewed, including "Analytical/Virtual Databases," "Adaptive Awareness," and "Situational-Sensitive, Context-Aware, Role-Based Access Groups" working in both connected and disconnected models across multiple vendor platforms.
 
   

Melissa R. Finley, Susan A. Caskey, Troy D. Ross, Bradly D. De Groot, Kathleen M. Lowe, Flint Taylor,
Billy Dictson, Jeff Witte
biography

"The Rapid Syndrome Validation Project – Animals: Monitoring Animal Health"

The Rapid Syndromic Validation Project – Animal (RSVP-A) is an animal health surveillance system designed to facilitate rapid communication between practicing veterinarians and national, state and regional veterinary service organizations.  RSVP-A is currently being tested in Kansas and New Mexico as a means to monitor and improve the health of the national cattle herd using practicing veterinarians as data collectors.  The syndromes that RSVP-A currently tracks are defined to exclude common diseases and production problems, but include the signs of less common disease presentations that an exotic disease outbreak in cattle might mimic.  Practicing veterinarians report these syndromes into the RSVP-A system through personal data assistants (PDAs) or a web-based interface.  RSVP-A presents syndrome data spatially and temporally on the web interface with geographic information systems (GIS) technology.  Other data sets are included and can be layered on the GIS display.  Advisories from state veterinary diagnostic labs or other veterinary epidemiologists are displayed in a textual manner.  The system in its current configuration does not use Global Positioning System (GPS) because this level of information could compromise the confidentiality of beef and dairy producers. 

 
   

J. H. Kaufman and D. A. Ford
biography

"Protecting Public Health through advanced SpatioTemporal Epidemiological Modeling"

The rise of global economies in the 21st century, and the increased reliance by developed countries on global transportation and trade, increases the risk of national and worldwide disaster due to infectious disease. Such epidemics may be the result of global climate change, vector-borne diseases, food-borne illness, new naturally occurring pathogens, or bio-terrorist attacks.  The threat is most severe for rapidly communicable diseases.  The coincidence of rapidly spreading infectious disease along with the rapid transportation, propagation, and dissemination of the pathogens and vectors for infection poses the risk of new and dangerous pandemics. The anthrax attacks in the United States, which occurred shortly after the events of September 11th, demonstrated the subtle nature of biological attacks and their effectiveness in spreading terror.

An equally significant threat to public health is posed by the emergence of new infectious disease by cross species infection. This process resulted in the world wide AIDs epidemic. A topical example is emerging in Asia with the transmission of Avian Influenza A (H5N1) from birds to Humans.  In 1997, 6 people died in Hong Kong after being infected with this virus. The number of cases world wide has been rapidly increasing since 1997 and, more recently, several patients in Thailand have died. In at least one instance this new virus has demonstrated the ability to effectively transfer between humans.  If the virus acquires the ability to rapidly spread by the upper respiratory system, the consequences for the world's Human population could be dramatic. The “Spanish Flu” pandemic of 1918-1919 was caused by a mutated flu virus and caused an estimated 30 to 50 million deaths world-wide. The full dynamic consequences of the transfer of a disease from one species to another is critical to understand. Such transfers can be a disaster for the receiving species as it often has little natural resistance to new pathogens.

Reacting to infectious disease, whether naturally occurring or the result of bio-terrorist attack requires a strategic understanding of the dynamics of the spread of the disease.  In this talk we will describe the IBM Research Spatiotemporal Epidemiological Modeler (STEM). STEM is a tool designed to help scientists and public health officials create and use spatial and temporal models of emerging infectious diseases.  Policymakers responsible for creating strategies to contain diseases and prevent epidemics need an accurate understanding of disease dynamics and the likely outcomes of preventative actions. In an increasingly connected world with extremely efficient global transportation links, the vectors of infection can be quite complex. STEM facilitates the creation of flexible models involving multiple populations (species) and interactions between diseases.

 
   

Dr. Chwan-Chuen King, Dr.PH.
biography

Timely and sensitive outbreak detection is the number one priority of all surveillance systems worldwide. Early detection should enable officials to quickly focus limited public health resources, decrease mortality, and provide efficient communications to minimize the risk of acquiring the infection or the disease. The system has to be simple, flexible, and, most importantly, acceptable to providers who like to participate without extra work load.

To face future challenges of EID, the integration of laboratory results, clinical data and epidemiological investigation is necessary. Furthermore, using HL-7 to exchange information for currently operating different surveillance systems (sentinel physician surveillance, school-based absenteeism surveillance, healthcare worker fever surveillance, nursing-home fever surveillance) and obtaining chief complains in the emergency Dept. to establish a “surveillance net” will be future efforts.  We have applied such timely automatic syndromic surveillance to successfully detect enterovirus 71 and human influenza virus activities and outbreak detections in much earlier time. In conclusion, continuous developing new surveillance system and evaluating different currently available surveillance systems will be very helpful to minimize the public health threats by avian influenza and other emerging infectious diseases.

 
   

Kenneth Komatsu, MPH
biography

"Biosurveillance in Arizona"

With new efforts and resources devoted to homeland security and public health preparedness, new methods and approaches to detect infectious disease outbreaks are needed.  This brief presentation will highlight the purpose, possible data sources and some of the biosurveillance issues from an epidemiologist’s perspective for the Arizona Department of Health Services.

 
   

James Kvach
biography

“International Infectious Disease Information Technology:  What’s Needed and What’s Possible”

There has never been a greater need for international infectious disease situational awareness as evidenced by the current epidemic of highly pathogenic H5NI avian influenza in wild and domestic birds.  Despite the availability of information technology a worldwide IT system capable of providing spatial-temporal situational awareness of endemic and epidemic animal, plant and human diseases is unlikely to become a reality any time soon

 
   

Bill Lober, MD
biography

"A Web Services Framework for Population Health Monitoring"

Since 2000, the Clinical Informatics Research Group has built several syndromic surveillance systems in the Pacific Northwest.  Yet, we have found it challenging to rapidly build and deploy new systems, to compare clustering or detection algorithms, and to replicate our work outside our region.  Our work-in-progress report will describe a new web services framework to segregate and structure components for data acquisition and management, text classification and clustering, event detection, and graphic display and visualization.  The presentation will describe the Shoki framework for population health monitoring, and an application of Shoki to rapidly create a syndromic surveillance system for the Integrating the Healthcare Enterprise Showcase at HIMSS 2006.

 
   

Cecil Lynch
biography

"Public Health Data Standards and Ontologies"

The domain of infectious diseases, and the knowledge associated with effective surveillance activities surrounding outbreaks of infectious disease has reached a level of complexity, and quantity that effective means of storing and retrieving knowledge has exceeded the capacity of the traditional paper-based world of public health.  Public health has struggled with the adoption of standards for vocabulary and messaging infrastructure that can be generalized to the entire public health community.  The centers for disease control, has initiated several efforts surrounding the uniform implementation of infectious disease surveillance systems and has provided guidance in the form of message structures and vocabulary.

In my talk, I will go over the issues associated with the development of a foundational model for infectious disease surveillance and the translation of that model to an ontological framework that allows distribution of standards-based vocabulary and message segments, and that instantiates the semantics of a broader information model into discrete reusable fragments.  We will discuss the challenges associated with standardizing surveillance instruments and discuss ways of capturing well defined and structured data at the source of an investigation.

 
   

Mark Thurmond DVM, PhD
biography

"Global FMD risk surveillance"

Foot-and-mouth disease (FMD) is considered one of the most contagious infectious animal diseases in the world, where affected countries typically must endure severe and far-reaching economic and social consequences.  Currently only 34% of the 167 member countries of the OIE are considered ‘free’ where vaccination is not practiced, and many recent efforts to eradicate FMD in many areas of the world have met with little success.  While the United States is currently FMD-free, the disease posses a significant threat to the sustainability of U.S. animal agriculture, and direct and indirect costs of FMD could well likely exceed $50 billion.  The U.S. is particularly vulnerable to FMD, once it enters the country, because the highly efficient system of animal movement will quickly spread the disease throughout the country.  In addition to the expanding trade and travel between the U.S. and countries with FMD, the escalating threats of biowarfare and agroterrorism increase the likelihood that, in the not too distant future, FMD will be introduced into the U.S.  Thus, in order to protect the U.S. against an incursion of FMD, it will be necessary that we understand the changing complexion of FMD throughout the world and how these changes could affect the threat of FMD for the U.S.

The overall goal of the FMD Lab has been to develop models and systems for the global surveillance of FMD.  Mathematical and statistical models have been developed to understand the global epidemiology of FMD and to predict future occurrences and risks of FMD throughout the world.  These models are particularly important in assessing risks of FMD in regions and countries where disease reporting is not undertaken and other data typically useful in predicting disease are not available.  In these situations, models use surrogate and remote-sensed data, which provide some correlated value with disease risk.  The repertoire of models includes local or country-level models utilizing region-specific data, a global model that predicts FMD risk in 50 km2 cells, spread models that projects temporal-spatial movement of FMD from region to region, anomaly detection models, and phylogenetic models that predict temporal-spatial evolutionary changes in the virus.  In collaboration with the AI Lab (U. Arizona) programs are being developed to incorporate data, models, and projections into a coherent information web-based routing system, referred to as the FMD BioPortal, that permits real-time access and analysis of the global FMD data.  A new initiative is being proposed for development of a global network of countries that will collaborate in providing and evaluating real-time FMD data, via the FMD BioPortal, relating initially to predicting evolutionary changes in the virus.

 
   

Xiaohhui Zhang, Ph.D.
biography

"Development of A Bio-Intelligence System to Over-The-Counter Pharmaceutical Sales Data To Identify Potential Disease Outbreaks"

This presentation introduces the development and application of a bio-intelligence system, which systematically developed a dynamic system model, and integrated advanced information technology and with public health knowledge, to identify potential disease outbreaks and estimate the community health status through monitoring over-the-counter (OTC) pharmaceutical sales data.  The developed information system features automated knowledge acquisition, the spatial and temporal analysis and decision-making.  A successful pilot application of this system was executed for the New Hampshire Department of Health and Human Services (NH DHHS). Since its inception, the system has successfully supported the identification of gastrointestinal illness and influenza outbreaks.

 

   
Copyright (C) 2006 Artificial Intelligence Lab, The University of Arizona