Syndromic Surveillance Using Ambulatory Electronic Health Records

From Clinfowiki
Revision as of 04:27, 30 November 2015 by Nneka.nwaeme (Talk | contribs)

Jump to: navigation, search

Introduction

Syndromic Surveillance is the monitoring of specific syndromes, such as the flu or gastrointestinal disease. Data sources for syndromic surveillance include emergency department chief complaints, and more recently with their wide implementation, Electronic Health Records (EHRs). This study focuses specifically on the syndromes of flu and gastrointestinal disease.

Methods

Two syndromic surveillance systems were used in the Institute for Family Health (IFH), which is comprised of 13 community health centers in Manhattan and Bronx, NY. One of the systems consisted of searching for structured (specifically ICD-10 coded) data in health records, while the second one analyzed narrative text. The data obtained was compared to the NYC Dept. of Health and Mental Hygiene Emergency Dept. Chief Complaint Surveillance System and also to NYC WHO data for positive influenza isolate results. WHO data was not available for gastrointestinal disease.

Results

The authors found that a syndromic surveillance system using structured EHR data had a cross-correlation of 0.89 to Emergency Department (ED) data for influenza like illness (ILI) and 0.81 for gastrointestinal infectious disease (GID). When the system used structured data, there was a 0.84 cross-correlation to ED data for ILI, but 0.47 to ED data for GID.

Discussion

The study design had many flaws:

-EHR variables had to be mapped to terms in the syndrome definitions, requiring engineers experienced with the IFH Epic system.

-The degree of system tailoring did not allow for conclusions about structured v. narrative data. Goals and context should be considered when choosing one over the other.

-Only E.D. chief complaint data was available to compare for GID results from the surveillance systems.

My Comments

There were some parts of this article that were difficult to follow, examples of the ED chief complaint data would have greatly aided in understanding the comparison between the syndromic surveillance systems data and ED data. However, I agree that the authors did succeed in showing that EHRs are a good data source for syndromic surveillance.[1]


References

  1. Hripcsak, G., Soulakis, N.O., Li, L., Morrison, F.P., Lai, A.M., Friedman, C., ..., Mostashari, F.(2009). Syndromic Surveillance Using Ambulatory Electronic Health Records. Journal of American Medical Informatics Association, 16. DOI 10.1197/jamia.M2922