ISDS Conference

Within the syndromic surveillance literature there are acknowledged gaps with respect to penetration of syndromic surveillance systems and standard or promising practices for response. The lack of adequate data and evidence-based policy recommendations on response is especially concerning because syndromic surveillance systems are only as useful as the timely pubic health response launched after aberration detection. We undertook the first step of a multi-phase study, with the global objectives of describing existing infrastructure in responding to alerts generated by syndromic surveillance systems and creating response guidance materials for public health practitioners. The preliminary findings contained here describe syndromic surveillance systems in use throughout the United States, future plans related to the use of such systems, and basic information regarding how outbreak response is initiated. This cataloging of systems complements work currently underway by the International Society for Disease Surveillance directed towards developing a comprehensive registry of syndromic surveillance systems.

Objective

We aim to describe current syndromic surveillance systems in use throughout the U.S. and approaches to initiating an outbreak response as reported by survey participants.

Although syndromic surveillance cannot serve its intended purpose without the timely public health response launched after aberration detection, the literature is very limited with respect to response to syndromic surveillance systems alerts and related guidance for public health practitioners. Literature reviews reveal an absence of uniform approaches to developing and evaluating response protocols. The one published study that aimed to inform the development of written protocols was based on experience with a single system, ESSENCE, and concluded that careful development of an evaluation and response framework should be undertaken.

Objective

To develop a framework for public health departments to use for developing and enhancing response protocols to syndromic surveillance system alerts.

The Connecticut Department of Public Health (DPH), like all public health agencies, is constantly challenged by new health threats and emerging diseases. A major responsibility of these agencies is the rapid and effective communication of information on emerging threats to members of the public who may be potentially exposed. This responsibility for effective risk communication is critical when the public perception of risk is high. The September 11, 2001 terrorist attacks and subsequent anthrax mail attacks (Amerithrax) resulted in a new era of public risk perception and concern. Many new and advanced surveillance systems, developed in response to these events, have increased the need for effective risk communication. For example, the DPH developed its first syndromic surveillance system in September 2001 to monitor for possible bioterrorism events and emerging infections. This resulted in the implementation of a number of risk communication and response protocols. These and other protocols were tested in responding to the recent anthrax contamination of a drum maker’s residence and a multistate rash outbreak.

Objective

This paper describes various risk communications techniques used in Connecticut to provide health information to the public following surveillance signal alerts. The use of hotlines and contemporary social networking systems to quickly communicate with targeted populations are compared to the use of news releases and other traditional approaches.

Over the past seven years, the number of Lyme Disease (LD) cases in Anne Arundel County has more than doubled, from 84 in 2000 to 196 in 2007, which correlates to CDC findings. It is endemic in 10 states, including Maryland, and Anne Arundel County has the second highest number of LD cases in the state. Despite the increasing prevalence and growing public concern, there is no definitive evidence regarding efficacy of personal preventive measures and environmental interventions. Other county-level studies have investigated risk factors, but none have included the investigation as a part of routine surveillance or narrowed the study population to cases with a known exposure date range.

Objective

In order to respond to the increase of reports of LD to local health departments and the limited utility of routine LD surveillance, active surveillance activities were focused on collecting exposure data from LD cases with a reasonably narrow date range of exposure.

With increased penetration of clinical information system products and increased interest in clinical data exchange, a variety of clinician’s notes are becoming available for surveillance. Chief complaints have been studied extensively, and emergency department notes have received attention, but narrative clinic visit notes have gotten little attention.

Objective

To assess the performance of an unmodified, general purpose natural language processing system to detect fever, and to assess the feasibility of parsing visit notes for syndromic surveillance.

Emergency Department surveillance methods currently rely on identification of acute illness by tracking chief complaint or ICD9 discharge codes. Newer generation electronic medical records are now capturing additional  information such as vital signs. These data have the potential for identifying disease syndromes earlier than the traditional methods.

Objective

This paper describes the temporal relationship between numbers of cases of fever, recorded as discrete vital sign data in an electronic medical record, and ICD9 Influenza Like Illnesses in the Emergency Department at the University of Wisconsin Hospital.

The spatial scan statistic [1] detects significant spatial clusters of disease by maximizing a likelihood ratio statistic over a large set of spatial regions. Typical spatial scan approaches either constrain the search regions to a given shape, reducing power to detect patterns that do not correspond to this shape, or perform a heuristic search over a larger set of irregular regions, in which case they may not find the most relevant clusters. In either case, computation time is a serious issue when searching over complex region shapeso r when analyzing a large amount of data. Analternative approach might be to search over all possible subsets of the data to find the  most relevant pat-terns, but since there are exponentially many subsets, an exhaustive search is computationally infeasible.

Objective

We present a new method of "linear-time subset scanning" and apply this technique to various spatial outbreak detection scenarios, making it computationally feasible (and very fast) to perform spatial scans over huge numbers of search regions.

The semantic web is an emerging technology for expressing rich descriptions of a problem domain in the form  of  ontologies.    An  ontology  provides  a  domain  specific  knowledge  base  for  the  communication  and  sharing  of  knowledge  between  various  human  and  computer agents [1].   Many  public  health  organizations  have  adopted  syndromic surveillance systems but criteria for the selection  of  appropriate  data  sources,  syndrome  definitions,   and   applicable   outbreak   detection   methods   have not been established [2].  Application of semantic web technology to the field of syndromic surveil-lance has been seen to be successful in an experimental  environment  through  the  BioSTORM  project  at  the  SMI  labs  at  the  Stanford  University  School  of  Medicine  [3].    The  semantic  web  shows  promise  for  providing  a  universal  problem  description  layer  that  will allow for easier integration between heterogeneous data sources and problem solving techniques. 

Objective

A syndromic surveillance system which uses a semantic web description layer is more extensible than existing systems. This will be shown through the application of appropriate software metrics, as well as a case based review that targets three major system design components.

Early detection of new diseases such as bovine spongiform encephalopathy is the subject of great interest (Gibbens et al., 2008). Understanding whether a disease is infectious or sporadic becomes essential for the application of control measures. Consistent and robust ways to the assessment of temporal trends are required to help in the elucidation of this question. Clustering of cases in space, or time and space, is also relevant in the understanding of the aetiology of a new disease. This paper presents a third approach: knowledge by comparison, either of diseases, surveillance sources or both. We applied this approach to the current debate about the nature of atypical scrapie, a fatal neurological animal disease, by comparing the spatial distribution of this form of scrapie with that of classical scrapie. A similar spatio-temporal distribution of these two diseases would indicate shared environmental disease determinants and help in the generation of hypotheses about the aetiology of atypical scrapie.

A Quest Diagnostics Incorporated – CDC collaboration in 2000  pioneered  exploration  of  test  ordering data to enhance infectious diseasessurveillance1. This  year’s  unexpected shortage of vaccine and reports of human illness caused by avian influenza  A  (H5N1)  in  Asia2  heightened concern about  influenza and focused attention on moving toward more complete, real time surveillance. We extended our previous collaboration to explore the use of  the Quest Diagnostics Corporate Informatics Data Warehouse (QIDW) as a tool for surveillance of influenza.

Objective

To explore the potential of a large commercial data warehouse for influenza surveillance.