ISDS Conference

The Centers for Disease Control and Prevention BioSense project has developed chief complaint (CC) and ICD9 sub-syndrome classifiers for the major syndromes for early event detection and situational awareness. This has the potential to expand the usefulness of syndromic surveillance, but little data exists evaluating this approach. The overall performance of classifiers can differ significantly among syndromes, and presumably among subsyndromes as well. Also, we had previously found that the seasonal pattern of diarrhea was different for patients < 60 months of age (younger) and for patients > 60 months of age (older).

Objective

Using chart review as the criterion standard to estimate the sensitivity, specificity, positive predictive value and negative predictive value of New York State hospital emergency department CC classifiers for patients < 60 months of age and > 60 months of age for the gastrointestinal (GI) syndrome and the following GI sub-syndromes: “abdominal pain”, “nausea-vomiting” and “diarrhea”.

Ideal anomaly detection algorithms should detect both sudden and gradual changes, while keeping the background false positive alert rate at a tolerable level. Further, the algorithm needs to perform well when the need is to detect small outbreaks in low-incidence diseases. For example, when surveillance is done based on the specific ICD9 diagnosis of flu rather than a larger syndromic grouping, the baseline counts will generally be low, in the range of 0 or 1 per day even in a large sample of EDs. 

Objective

Our goal was to determine the sensitivity of detection of various inserted outbreak sizes and shapes using a modified Holt-Winters detection algorithm applied to daily flu count data before the flu season and after its peak. We compare our results to C3 of EARS.

On December 14th, 2006, a severe windstorm in western Washington caused hundreds of thousands of residents to lose power. On December 15, 2006, there was a surge in emergency department (ED) visits for patients presenting with signs of acute carbon monoxide (CO) poisoning. A Public Health investigation was initiated following the storm to determine the extent of CO poisoning due to the windstorm. A retrospective analysis was later undertaken to evaluate how well our syndromic surveillance system was able to identify patients who presented to area EDs with carbon monoxide poisoning.

Objective

We evaluated the performance of our ED syndromic data for detecting visits associated with CO poisoning.

To recognize outbreaks so that early interventions can be applied, BioSense uses a modification of the EARS C2 method, stratifying days used to calculate the expected value by weekend vs weekday, and including a rate-based method that accounts for total visits. These modifications produce lower residuals (observed minus expected counts), but their effect on sensitivity has not been studied.

Objective

To evaluate several variations of a commonlyused control chart method for detecting injected signals in 2 BioSense System datasets.

The Georgia Power Corporation (GPC) provides power to 155 (97.5%) of the 159 counties in Georgia (GA), and employs 9,600 people throughout the state. GPC is engaged in preparing for pandemic influenza, and committed to protecting the critical infrastructure and ensuring its continuity of operations. The GPC employee “Crisis Absence Reporting Tool” (CART) was designed to provide the Georgia Syndromic

Surveillance (GA SS) Program with employee absentee/ reason to inform Public Health and GPC leadership about health events occurring in their employees statewide.

The GA SS Program has been implemented in 13 (72%) of the 18 Health Districts. In each of these locations, data are transferred from an ED, ambulatory care center, or school district to the Georgia Division of Public Health (GDPH) for analysis and dissemination of results to all stakeholders. GDPH wanted to collaborate with a large corporation with a statewide employee base to conduct absentee and reason for absence SS to provide an additional perspective to the existing data streams used by GA SS.

In GA, the LHD are responsible for organizing pandemic planning committees comprised of community partners to discuss continuity of basic services and maintenance of the critical infrastructure at the local level during an influenza pandemic. Increasing SS capacity is an important component of Local Health District (LHD) pandemic planning strategies in GA.

Objective

To create a non-traditional partnership between the GPC and the GDPH to aid in adverse health event detection and response activities during an influenza pandemic or other health emergency. This will include augmenting CART with SS data from the GA SS Program. These data will be analyzed by GA SS and results disseminated to LHDs, who monitor and respond to SS data in their jurisdictions. Analyses will also be provided to GPC to aid in resource allocation to ensure the continuity of services in GA during emergencies.

Irregularly shaped cluster finders frequently end up with a solution consisting of a large zone z spreading through the map, which is merely a collection of the highest valued regions, but not a geographically sound cluster. One way to amenize this problem is to introduce penalty functions to avoid the excessive freedom of shape of z. The compactness penalty K(z) is a function used to reduce the scan value of irregularly shaped clusters, based on its geometric shape. Another penalty is the cohesion function C(z), a measure of the absence of weak links, or underpopulated regions within the cluster which disconnect it when removed. It was mentioned in that such weak links could be responsible for a diminished power of detection in cluster finder algorithms. Methods using those penalty functions present better performance. The geometric  compactness is not entirely satisfactory, although, because it has a tendency to avoid potentially interesting irregularly shaped clusters, acting as a low-pass filter. The cohesion function penalty method, although, has slightly less specificity.

Objective

We introduce a novel spatial scan algorithm for finding irregularly shaped disease clusters maximizing simultaneously two objectives: the regularity of shape and the internal cohesion of the cluster.

A number of different methods are currently used to classify patients into syndromic groups based on the patient’s chief complaint (CC). We previously reported results using an “Ngram” text processing program for building classifiers (adapted from business research technology at 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 (or Ngrams), with associated probabilities, from the training data. We then generate a CC classifier from the collection of Ngrams and use it to find a classification probability for each patient. Previously, we presented data showing good correlation between daily volumes as measured by the Ngram and ICD9 classifiers.

Objective

Our objective was to determine the optimized values for the sensitivity and specificity of the Ngram CC classifier for individual visits using a ROC curve analysis. Points on the ROC curve correspond to different classification probability cutoffs.

In 2004, the Indiana State Department of Health (ISDH) partnered with the Regenstrief Institute to begin collecting syndromic data from 14 ED’s to monitor bioterrorism-related events and other public health emergencies. Today, Indiana’s public health emergency surveillance system (PHESS) receives approximately 5,000 daily ED visits as real-time HL7 formatted surveillance data from 55 hospitals. The ISDH analyzes these data using ESSENCE and initiates field investigations when human review deems necessary.1 The Marion County Health Department, located in the state’s capitol and most populous county, is the first local health department in Indiana using ESSENCE.

Objective

This paper describes how local and state stakeholders interact with Indiana’s operational PHESS, including resources allocated to syndromic surveillance activities and methods for managing surveillance data flow. We also describe early successes of the system.

Real-time Outbreak and Disease Surveillance (RODS), a syndromic surveillance system created by the University of Pittsburgh has been used in Ohio by the state and local health departments since late 2003. There are currently 133 health care facilities providing 88% coverage of emergency department visits statewide to the RODS system managed by Health Monitoring Systems Inc. (HMS). The system automatically alerts health department jurisdictions when various syndromic thresholds are exceeded.

As part of response protocols, investigators export a case listing in a comma-separated values file which typically includes thousands of lines with each row containing: date admitted, age, gender, zip code, hospital name, visit number, chief complaint, and syndrome. The HMS-RODS web site provides basic graphs and maps, yet lacks the flexibility afforded by ad hoc queries, cross tabulation, and portability enabling off-line analysis.

Objective

This paper describes the integration of open source applications as portable, customizable tools for epidemiologists to provide rapid analysis, visualization, and reporting during surveillance investigations.

Free-text emergency department triage chief complaints (CCs) are a popular data source used by many syndromic surveillance systems because of their timeliness, availability, and relevance. The lack of standardization of CC vocabulary poses a major technical challenge to any automatic CC classification approach. This challenge can be partially addressed by several methods, for example, medical thesaurus, spelling check, manually-created synonym list, and supervised machine learning techniques that directly operate on free text. Current approaches, however, ignore the fact that medical terms appearing in CCs are often semantically related. Our research exploits such semantic relations through a medical ontology in the context of automatic CC classification for syndromic surveillance.

Objective

This paper presents a novel approach of using a medical ontology to classify free-text CCs into syndrome categories.