Skip to main content

DeLisle Sylvain

A comprehensive electronic medical record (EMR) represents a rich source of information that can be harnessed for epidemic surveillance. At this time, however, we do not know how EMR-based data elements should be combined to improve the performance of surveillance systems. In a manual EMR review of over 15 000 outpatient encounters, we observed that two-thirds of the cases with an acute respiratory infection (ARI) were seen in the emergency room or other urgent care areas, but that these areas received only 15% of total outpatient visits. Because of this seemingly favorable signal-to-noise ratio, we hypothesized that an ARI surveillance system that focused on urgent visits would outperform one that monitored all outpatient visits.

Referenced File
Submitted by hparton on

Microorganisms resistant to antibiotics (ABX) increase the mortality, morbidity and costs of infections. In the absence of a drug development pipeline that can keep pace with the emerging resistancemechanisms, these organisms are expected to threaten public health for years to come. Because exposure to ABX promotes the development of bacterial resistance, health care providers have long been urged to avoid using antibiotics to treat conditions that they are unlikely to improve, including many uncomplicated acute respiratory infections. We asked if interposing clinical decision support software at the time of electronic order entry could adjust ABX utilization toward consensus guidelines for these conditions. 

Referenced File
Submitted by hparton on

Measures aimed at controlling epidemics of infectious diseases critically benefit from early outbreak recognition. Through a manual electronic medical record (EMR) review of 5,127 outpatient encounters at the Veterans Administration health system (VA), we previously developed single-case detection algorithms (CDAs) aimed at uncovering individuals with influenza-like illness (ILI). In this work, we evaluate the impact of using CDAs of varying statistical performance on the time and workload required to find a community-wide influenza outbreak through a VA-based syndromic surveillance system (SSS). The CDAs utilize various logical arrangements of EMR data, including ICD-9 codes, structured clinical parameters, and/or an automated analysis of the free-text of the full clinical note. The 18 ILI CDAs used here are limited to the most successful representatives of ICD-9-only and EMR-based case detectors.

 

Objective

This work uses a mathematical model of a plausible influenza epidemic to begin to test the influence of CDAs on the performance of a SSS.

Referenced File
Submitted by elamb on

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.

Referenced File
Submitted by elamb on

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.

Referenced File
Submitted by elamb on

Objective

There were two objectives of this analysis. First, apply text-processing methods to free-text clinician notes extracted from the VA electronic medical record for automated detection of Influenza-Like-Illness. Secondly, determine if use of data from free-text clinical documents can be used to enhance the predictive ability of case detection models based on coded data.

Referenced File
Submitted by elamb on

Measures aimed at controlling epidemics of infectious diseases critically benefit from early outbreak recognition [1]. SSS seek early detection by focusing on pre-diagnostic symptoms that by themselves may not alarm clinicians. We have previously determined the performance of various Case Detector (CD) algorithms at finding cases of influenza-like illness (ILI) recorded in the electronic medical record of the Veterans Administration (VA) health system. In this work, we measure the impact of using CDs of increasing sensitivity but decreasing specificity on the time it takes a VA-based SSS to identify a modeled community-wide influenza outbreak. Objective This work uses a mathematical model of a plausible influenza epidemic to test the influence of different case-detection algorithms on the performance of a real-world syndromic surveillance system (SSS).

Referenced File
Submitted by elamb on

To understand the types of false positive cases identified by an Influenza-like illness (ILI) text classifier by measuring the prevalence of ILI-related concepts that are negated, hypothetical, include explicit mention of temporality, experienced by someone other than the patient, or described in templated text that is difficult to process.

Referenced File
Submitted by elamb on

Communicable diseases are underreported by physicians, especially diseases without laboratory tests. The goals of our study were to determine reporting levels for clinical chickenpox, describe clinical data elements common to chickenpox, and assess ability of an electronic syndromic surveillance system, BioSense, to capture chickenpox cases.

Referenced File
Submitted by elamb on
Subscribe to DeLisle Sylvain