ESSENCE

ESSENCE is a web-based syndromic surveillance system utilized by DHMH to detect and track outbreaks, suspicious patterns of illness, public health emergencies, and biological threats. ESSENCE ED chief complaint data is collected daily from 47 emergency departments in Maryland (all 45 acute care hospitals and 2 freestanding emergency medical facilities). A chief complaint in ESSENCE is a free text field that lists the patient’s reason for the ED visit upon arrival at the hospital. Chief complaint data may be comprehensive or abbreviated and may include a single reason or multiple reasons for the ED visit. Medical history may be included in chief complaint data, which can create low specificity (false positive cases). Chief complaint data alone may yield less accurate modeling and lower outbreak detection sensitivity. This analysis evaluates whether counts of chief complaints are appropriate indicators of disease burden for several specific illnesses, by comparing chief complaints to their corresponding discharge diagnoses.

Objective

The state of Maryland has incorporated chief complaint data from 100% of its acute care emergency departments (ED) into the Electronic Surveillance System for the Early Notification of Community-Based Epidemics (ESSENCE). In 2012, the Maryland Department of Health and Mental Hygiene (DHMH) began using this statewide disease surveillance system to track several specific disease measures including certain chronic diseases. The validity of using ESSENCE ED data to track and analyze these health outcomes was evaluated.

The Electronic Surveillance System for the Early Notification of Community-based Epidemics in Florida (ESSENCE-FL) is a web-based application for use by public health professionals within the Florida Department of Health (FDOH). The main source of data for ESSENCE-FL is emergency department (ED) data. Ten hospitals in Hillsborough County, Florida send their data to the ESSENCE-FL server. ESSENCE-FL requires only a limited data set to be sent by the hospital which includes patient chief complaint (CC) and discharge diagnosis (DD). These fields can be searched individually, in separate queries, to identify possible records of interest. These two fields have been concatenated to create the single chief complaint and discharge diagnosis (CCDD) field, allowing both fields to be searched with a single query.

Objective

While syndromic surveillance systems were originally designed for the detection of outbreaks and clusters of illness, they have been found to be useful at identifying unreported conditions of public health importance. Within the Florida Department of Health in Hillsborough County (FDOH-Hillsborough), these conditions of public health importance have primarily focused on the reportable diseases and conditions that fall under the responsibility of the Epidemiology Program and have not included tuberculosis. A specific query has been developed to search for and identify possible tuberculosis patients and exposed contacts. This study is designed to determine the usefulness of specific-term chief complaint and discharge diagnosis (CCDD) queries in identifying tuberculosis patients and exposed contacts.

One criterion for evaluating the effectiveness of a surveillance system is the system’s positive predictive value. To our knowledge few studies have described the positive predictive value of syndromic surveillance signals for naturally occurring conditions of public health importance.

Objective

We evaluated the positive predictive value of signals detected by our syndromic surveillance system.

In addition to utilizing syndromic surveillance data to respond to public health threats and prepare for major incidents, local health departments can utilize the data to examine patient volumes in the emergency departments (EDs) of local hospitals. The information obtained may be valuable to hospital and clinic administrators who are charged with allocating resources. 

Indianapolis represents 92% of Marion County’s population. The county’s public hospital and clinic network provide care for 1 in 3 county residents who are Medicaid enrollees or uninsured. To assess the need for extended hours at eight public primary care clinics in Marion County, Indiana, this study examined the hospital’s ED volume. We hypothesize that

changes in the ED volume trends that corresponded to the start or end of usual clinic hours (8am-5pm) would be evidence of clinic hours’ impact on ED use.

Objective

This paper highlights the use of syndromic surveillance data to examine daily trends in ED volume at an urban public hospital.

The United States Environmental Protection Agency (U.S. EPA) has developed a prototype contamination warning system (CWS) for drinking water in response to Homeland Security Presidential Directive 9 (HSPD9). The goal of HSPD9 and the CWS is to expedite contamination containment and emergency response, thereby minimizing public health and economic impacts.

U.S. EPA’s conceptual CWS system, named WaterSentinel, is currently being pilot tested by U.S. EPA and its research partners. WaterSentinel is a multi-faceted approach involving water quality monitoring at optimal locations throughout the drinking water distribution system, enhanced security monitoring at key water utility infrastructure assets, consumer complaint surveillance, and innovative uses of public health surveillance data streams.

Objective

This paper summarizes the use and evaluation of various types of public health surveillance data for the early detection of chemical and biological contamination of drinking water.

The threat of pandemic and seasonal influenza has drawn attention to syndromic surveillance systems for early detection of influenza-like illness. Since 2005, the Miami-Dade County Health Department has implemented ESSENCE (Electronic Surveillance System for the Early Notification of Community-based Epidemics) to monitor emergency department data for influenza-like Illness (ILI) using chief complaint information. This study evaluates the ability of the ESSENCE ILI chief complaint grouping for identifying true ICD-9 diagnosed influenza.

Objective

Previous studies have examined the utility of different methods of syndromic grouping. This study evaluates the utility of ESSENCE for ILI surveillance.

A major goal of biosurveillance is the timely detection of an infectious disease outbreak. Once a disease has been identified, another very important goal is to find all known cases of the disease to assist public health investigators. Natural language processing (NLP) systems may be able to assist in identifying epidemiological variables and decrease time-consuming manual review of records.

Objective

To identify epidemiologically important factors such as infectious disease exposure history, travel or specific variables from unstructured data using NLP methods.

Although Electronic Surveillance System for the Early Notification of Community Based Epidemics (ESSENCE) provides tools to detect a significant alert regarding an unusual public health event, combining that information with other surveillance data, such as 911 calls, school absenteeism and poison control records, has proved to be more sensitive in detecting an outbreak. On Monday, June 16, Florida Poison Information Network, which takes after-hours and weekend calls for Miami-Dade County Health Department (MDCHD), contacted the Office of Epidemiology and Disease Control about five homeless persons that visited the same hospital simultaneously with gastrointestinal symptoms on Saturday, June 14. Poison control staff asked MDCHD to investigate further to determine whether it was an outbreak.

Objective

To illustrate how MDCHD utilized ESSENCE in order to track a gastrointestinal outbreak in a homeless shelter.

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.