Surveillance Systems

Syndromic surveillance can be a useful tool for the early recognition of outbreaks and trends in emergency department (ED) data. In addition, as a more timely data source than traditional disease reporting, syndromic data may also be leveraged to identify individual disease cases, increasing the utility for first time or redundant case recognition.

San Diego County (COSD) performs daily ED syndromic surveillance. In order to assess the utility for early identification of specific conditions of public health interest (e.g., salmonellosis, meningitis, hazardous exposures, heat-related illness), a novel process entitled Priority Infectious Conditions Capture, was developed.

Objective

This paper describes an assessment of an enhanced surveillance process used to identify reportable diseases and conditions of public health importance from ED chief complaint data in COSD.

In February of 2007, the Bureau of Epidemiology (BOE) received a request from Houston Department of Public Works to investigate a possible rise in gastrointestinal (GI) illness associated with complaints about poor water quality in a Northeastern Houston neighborhood. To investigate this complaint, BOE combined case report data with syndromic data from our Real-Time Outbreak Disease Surveillance (RODS). The Houston RODS collects and synthesizes real-time chief complaint data from 34 area hospitals and health facilities, representing approximately 70% coverage of licensed ER beds in Harris County. The system uses a Naïve Bayes Classifier to categorize ER chief complaints into 7 different syndromes, including GI illness.

Objective

To investigate public concern over a possible increase in GI illness associated with water quality complaints in Northeast Houston.

As public health surveillance is becoming more and more prevalent, new sources of data collection are more evident. One such data source is school absenteeism. By monitoring the symptoms of illness recorded when students are absent, health departments ideally can pinpoint potential outbreaks prior to their existence, all at little to no cost. The symptoms reported may not amount to disease, but their increase in incidence may indicate the preliminary spread of illness. This surveillance tool is also used to develop community intervention containment practices.

Objective

This paper describes the application of syndromic surveillance data from area school districts to detect influenza epidemics in a county setting.

The Miami-Dade County Health Department currently utilizes Emergency Department based Syndromic surveillance data, 911 Call Center data, and more recently Public School Absenteeism data. Daily monitoring of school absenteeism data may enhance early outbreak detection in Miami-Dade County in conjunction with the use of other syndromic systems. These systems were employed to detect any possible outbreaks resulting from a large outdoor festival occurring March 11th, 2007. This event had an estimated 1 million visitors and it ended at 7:00 p.m.

Objective

Utility of school absenteeism data to enhance syndromic surveillance activities for unusual public health events or outbreak detection.

The North Carolina Disease Event Tracking and Epidemiologic Collection Tool (NC DETECT) provides early event detection and public health situational awareness to hospital-based and public health users statewide. Authorized users are currently able to view data from emergency departments (n=110), the statewide poison control center, the statewide EMS data system, a regional wildlife center and pilot data from a college veterinary laboratory as well as select urgent care centers. While NC DETECT has over 200 registered users, there are public health officials, hospital clinicians and administrators who do not access the system on a regular basis, but still like to be kept abreast of syndromic trends in their jurisdictions. In order to accommodate this interest and reduce redundant data entry among active users, we developed a summary report that can be easily exported and distributed outside of NC DETECT.

Objective

This paper describes a user driven weekly syndromic report designed and developed to improve the efficiency of sharing syndromic information statewide.

Emergency Department (ED) syndromic surveillance data for influenza-like illness (ILI) have been found to provide timely and representative information about current influenza activity in NYC. DOHMH monitors visits daily from 50 of 61 EDs, capturing about 94% of all ED visits in NYC. Since January 1, 2007, DOHMH has been receiving disposition data (e.g., hospitalized, discharged) from a subset of EDs. Currently, disposition data is received from 37 EDs (approximately 1/3 of all visits by the next day and >60% of all visits within 1 week).

More detailed hospitalization data, including date, demographics, and diagnosis on all NYC hospitalizations are routinely collected by the New York State Department of Health Statewide Planning and Research Cooperative System (SPARCS). SPARCS is subject to a 2-3 year reporting lag, thus limiting its timeliness and prospective use. However, SPARCS data from prior to January 1, 2007 can supplement the ED syndromic data to develop a model for ILI hospitalizations and calculate excess hospitalizations attributable to influenza that can be used in near realtime, particularly in the event of a pandemic.

Objective

To use ED syndromic surveillance data to monitor hospitalizations for ILI and calculate excess hospitalizations attributable to influenza.

Difficulties in timely acquisition and interpretation of accurate data on communicable diseases can impede outbreak detection and control. These limitations are of global importance: they contribute to avoidable morbidity, economic losses, and social disruption; and, in a globalized world, epidemics can spread rapidly to other susceptible populations.

SARS and the potential for an influenza pandemic highlighted the importance of global disease surveillance. Similarly, the World Health Organization’s newly implemented 2005 International Health Regulations require member countries to provide notification of emerging infectious diseases of potential global importance. The challenges arise when Ministries of Health (MoH) in resource-poor countries add these mandates to already over-burdened and under-funded surveillance systems. Appropriately adapted, electronic disease surveillance systems could provide the tools and approaches MOHs need to meet today’s surveillance challenges.

Objective

In this presentation we will discuss the concept of electronic disease surveillance in resource-poor settings, and the issues to be considered during system planning and implementation.

OBJECTIVE

A “whole-system facsimile” recreates a complex automated biosurveillance system running prospectively on real historical datasets. We systematized this approach to compare the performance of otherwise identical surveillance systems that used alternative statistical outbreak detection approaches, those used by CDC’s BioSense syndromic system or a popular scan statistics.

Surveillance of individual data streams is a well-accepted approach to monitor community incidence of infectious diseases such as influenza, and to enable timely detection of outbreaks so that control measures can be applied. However the performance of alerts may be improved by simultaneously monitor a variety of data sources, or multiple streams (eg from different geographic locations) of the same type, rather than monitoring only aggregate data. Rates of influenza-like illness in subtropical settings typically show greater variability than in temperate regions.

Objective

This paper describes the use of time series models for simultaneous monitoring of multiple streams of influenza surveillance data.

In Connecticut (CT), several syndromic surveillance systems have been established by the Department of Public Health (DPH) to detect and monitor potential public health threats. The emergency department syndromic surveillance (EDSS) routinely categorizes chief complaint data into pre-defined syndrome categories, and also has the flexibility to define syndromes in real-time. Thus, DPH can use this system for situational awareness during public health events. Several recent events provided an opportunity to evaluate EDSS for this purpose: 1) two cases of cutaneous anthrax in CT in September 2007; 2) national and local media attention surrounding MRSA infections and published research in October 2007 and 3) the introduction of rotavirus vaccine through the Vaccines for Children Program in July 2006 following its licensing in February 2006.

Objective

To evaluate the performance of the CT EDSS system for situational awareness during specific public health events.