ISDS Conference

Syndromic surveillance of emergency department (ED) visit data is often based on computer algorithms which assign patient chief complaints (CC) and ICD code data to syndromes. The triage nurse note (NN) has also been used for surveillance. Previously we developed an “NGram” classifier for syndromic surveillance of ED CC in Italian for detection of natural outbreaks and bioterrorism. The classifier is developed from a set of ED visits for which both the ICD diagnosis code and CC are available by measuring the associations of text fragments within the CC (e.g. 3 characters for a “3-gram”) with a syndromic group of ICD codes. We found good correlation between daily volumes by the ICD10 classifier and estimated by NGrams. However, because the CC was limited to 23 options based on the pick list, it might be possible to obtain results as good as the NGram method or better using a simpler probabilistic approach. Also, in addition to the CC, the Italian data included a free-text NN note. We might be able achieve improved performance by applying the n-gram method to the NN or the CC supplemented by the NN.

Objective

Our objective was to compare the performance of the NGram CC classifier to two discrete classifiers based on probabilistic associations with the CC pick list items. Also, we wished to determine the performance of the NGram method applied to CC alone, NN alone, and CC plus NN.

Syndromic surveillance of emergency department (ED) visit data is often based on computer algorithms which assign patient chief complaints (CC) to syndromes. ICD9 code data may also be used to develop visit classifiers for syndromic surveillance but the ICD9 code is generally not available immediately, thus limiting its utility. However, ICD9 has the advantages that ICD9 classifiers may be created rapidly and precisely as a subset of existing ICD9 codes and that the ICD9 codes are independent of the spoken language. If a classifier based on ICD9 codes could be used to automatically create the code for a chief-complaint assignment algorithm then CC algorithms could be created and updated more rapidly and with less labor. They could also be created in multiple spoken languages. We had developed a method for doing this based on an “ngram” text processing program adapted from business research technology (AT&T Labs). The method applies the ICD9 classifier to a training set of ED visits for which both the CC and ICD9 code are known. A computerized method is used to automatically generate a collection of CC substrings with associated probabilities, and then generate a CC classifier program. The method includes specialized selection techniques and model pruning to automatically create a compact and efficient classifier.

Objective

Our objective was to determine how closely the performance of an ngram CC classifier for the gastrointestinal syndrome matched the performance of the ICD9 classifier.

The New York State Department of Health (NYSDOH) Syndromic Surveillance System consists of five components: 1. Emergency Department (ED) Phone Call System monitors unusual events or clusters of illnesses in the EDs of participating hospitals; 2. Electronic ED Surveillance System monitors ED chief complaint data; 3. Medicaid data system monitors Medicaid-paid over-the-counter and prescription medica-tions; 4. National Retail Data Monitor/Real-time Outbreak and Disease Surveillance System monitors OTC data; 5. CDC’s BioSense application monitors Department of Defense and Veterans Administration outpatient care clinical data (ICD-9-CM diag-noses and CPT procedure codes), and LabCorp test order data.

Objective

This poster presentation provides an overview of the NYSDOH Syndromic Surveillance System, including data sources, analytic algorithms, and resulting reports that are posted on the NYSDOH Secure Health Commerce System for access by state, regional, county, and hospital users.

In addition to monitoring Emergency Department chief complaint data and pharmacy sales as indicators of outbreaks, the New York State Department of Health (NYSDOH) Syndromic Surveillance System also monitors information from the CDC’s Early Event Detection and Situational Awareness System, BioSense. BioSense includes Department of Defense (DOD) and Veterans Affairs (VA) outpatient clinical data (ICD-9-CM diagnoses and CPT procedure codes), and LabCorp test order data. Within NYS excluding New York City, there are a total of 7 DOD and 60 VA hospitals and/or clinics reporting to the BioSense system, located across 41 of 57 counties.

BioSense includes a Sentinel Alert system, which monitors for diagnoses of CDC-classified Category A, B, and C diseases that have been reported from DOD and VA facilities. Sentinel Alerts are issued for single disease records, and can be followed up at local discretion to assess for public health significance and to determine whether the source of the disease might be intentional.

Objective

To describe the NYSDOH's experience with the monitoring of Sentinel Alerts generated for NYS within the CDC’s BioSense application, following up each alert with local health department staff to determine case resolution, and providing user-level feedback to the CDC to effect system improvements.

The development of a real time surveillance system for Forces on duty areas is one of the 5 initiatives of the November 2002 Prague’s NATO meeting. The French Military Health Service has decided to implement a military demonstrator within Forces in operations in a tropical area. This military prototype has three main objectives : i) to study the feasability of real time surveillance system within Forces in operations ii) to evaluate the benefit of such a system and iii) to develop a interoperable system for NATO. This French real time system has been developped by a multidisciplinary team, with military people but also with civilian experts from Pasteur Institute and Mediterranean University of Marseille.

Objective

This paper describes the new real time surveillance system, which has been installed within the French Forces in French Guiana.

On August 29, 2005, Hurricane Katrina made landfall just east of New Orleans, LA at 6:10AM CST and again at the LA/MS border at 10:00AM CST as a Category 3 hurricane, causing mass destruction along their coastlines. The devastation in LA and MS forced many residents to evacuate. Outside of the hurricane affected areas of LA, MS, and AL, GA received the second largest number of evacuees (approximately 125,000).

Objective

To describe the victims of Hurricane Katrina who evacuated to GA and to assess their impact on emergency departments enrolled in GA’s syndromic surveillance system.

Clinician reporting of notifiable diseases has historically been slow, labor intensive, and incomplete. Manual and electronic laboratory reporting (ELR) systems have increased the timeliness, efficiency, and completeness of notifiable disease reporting but cannot provide full demographic information about patients, integrate an array of pertinent lab tests to yield a diagnosis, describe patient signs and symptoms, pregnancy status, treatment rendered, or differentiate a new diagnosis or from follow-up of a known old diagnosis. Electronic medical record (EMR) systems are a promising resource to combine the timeliness and completeness of ELR systems with the clinical perspective of clinician initiated reporting. We describe an operational system that detects and reports patients with notifiable diseases to the state health department using EMR data.

Objective

To leverage EMR systems to improve the timeliness, completeness, and clinical detail of notifiable disease reporting.

On 27 April 2005, a simulated bioterrorist event—the aerosolized release of Francisella tularensis in the men’s room of luxury box seats at a sports stadium—was used to exercise the disease surveillance capability of the National Capital Region (NCR). The objective of this exercise was to permit all of the health departments in the NCR to exercise inter-jurisdictional epidemiological investigations using an advanced disease surveillance system. Actual system data could not be used for the exercise as it both is proprietary and contains protected, though de-identified, health information about real people; nor is there much historical data describing how such an outbreak would manifest itself in normal syndromic data. Thus, it was essential to develop methods to generate virtual health care records that met specific requirements and represented both ‘normal’ endemic visits (the background) as well as outbreak-specific records (the injects).

Objective

This paper describes a flexible modeling and simulation process that can create realistic, virtual syndromic data for exercising electronic biosurveillance systems.

Real-time disease surveillance is critical for early detection of the covert release of a biological threat agent (BTA). Numerous software applications have been developed to detect emerging disease clusters resulting from either naturally occurring phenomena or from occult acts of bioterrorism. However, these do not focus adequately on the diagnosis of BTA infection in proportion to the potential risk to public health.

GUARDIAN is a real-time, scalable, extensible, automated, knowledge-based BTA detection and diagnosis system. GUARDIAN conducts real-time analysis of multiple pre-diagnostic parameters from records already being collected within an emergency department. The goal of this system is to move from simple trend anomaly detection to an infectious disease specific expert system in order to assist clinicians in detecting potential BTAs as quickly and effectively as possible. GUARDIAN improves the diagnostic process for BTA infection through the capture and automated application of associated clinical expertise. The automated application of this knowledge provides the focus and accuracy necessary for effective BTA infection diagnosis. The continuity of this process improves the efficiency by which diagnoses of BTA infections can be made.

Complex, highly parameterized data models are often used to detect syndromic outbreaks. Unfortunately, such models can pose greater maintenance challenges as parameter variations increase. As such, our work focuses on whether day-of-the-week (DoW) effects may (or may not) show little variation across hospitals.

Objective

This paper investigates the existence of the DoW effect across twenty-six hospitals within the Indiana Public Health Emergency Surveillance System. We will consider both the impact of each DoW and the impact of individual hospitals.