ISDS Annual Conference Proceedings 2017. This is an Open Access article distributed under the terms of the Creative Commons Attribution-
Noncommercial 3.0 Unported License (http://creativecommons.org/licenses/by-nc/3.0/), permitting all non-commercial use, distribution,  
and reproduction in any medium, provided the original work is properly cited.

ISDS 2016 Conference Abstracts

HAI Surveillance Enhancement within EpiCenter by 
Utilization of Triage Notes
Pinar Erdogdu*, Stella Tsai and Teresa Hamby
New Jersey Department of Health, Trenton, NJ, USA

Objective
Evaluate the usage of triage note data from EpiCenter, a syndromic 

surveillance system utilized by New Jersey Department of Health 
(NJDOH), to enhance Healthcare-Associated Infections (HAIs) 
surveillance for infections following a surgical procedure.

Introduction
In New Jersey, Health Monitoring Systems Inc.’s (HMS) EpiCenter 

collects chief complaint data for syndromic surveillance from 79 of 
80 emergency departments (ED). Using keyword algorithms, these 
visits are classified into syndrome categories for monitoring unusual 
health events.

HAIs are infections that patients acquire while they are receiving 
treatment for a health condition in a health care setting. Following 
the 2014 Ebola outbreak in West Africa, the New Jersey Department 
of Health (NJDOH) Communicable Disease Service (CDS) started 
recruiting EDs to include triage note data in addition to chief 
complaint data to enhance surveillance capability for Ebola and other 
HAIs. Research by the University of North Carolina suggests triage 
note data improve the ability to detect illness of interest by fivefold1. 
Currently, there are three NJ EDs with triage note data in EpiCenter 
along with ICD 10 codes which can be used for comparison.

This pilot study will assess whether infections following a surgical 
procedure can be captured in triage note data along with ICD codes. 
Also, this evaluation will determine if triage note data can be used 
to create HAI custom classifications for syndromic surveillance. 
These classifications can potentially be used by surveillance  
and/or preparedness personnel and local health departments, as well 
as hospitals, to better prepare for detecting and preventing HAIs that 
are a significant cause of morbidity and mortality in the U.S.2

Methods
Three NJ facilities with triage notes information sending to 

EpiCenter were included in this study. ED visits occurred from 
10/23/2015 to 10/29/2015 and from 2/2/2016 to 2/10/2016 in these 
facilities with available ICD 10 codes information in EpiCenter were 
evaluated.

This analysis focused on sepsis and post-surgery infections related 
ICD 10 codes: A400, A401, A402, A403, A408, A409, A410, A411, 
A412, A414, A4150, A4151, A4152, A4158, A418, A419, R571, 
R578, R579, T811, T81.43. The keywords tested in triage notes 
are abdominal pain, redness, fev, fver, pyrexia, temp, elev temp, 
elevated temp, temp elev, hi temp, high temp, temp hi, temp10, temp 
10, feeling hot, feels hot, feel hot, fuo, febr, cloudy fluid, cfluid, 
drainage, abscess, wound, tenderness, swelling, erythema, red, pain, 
post surgery, fever.

The sensitivity, specificity and positive predictive value (PPV) 
of selected keywords applied in the triage notes were evaluated by 
comparing to patient’s ICD 10 codes.

Results
There were 2757 ED visits with triage notes and ICD 10 codes from 

10/23/2015 to 10/29/2015 and from 2/2/2016 to 2/10/2016. During 
these time frames, one ED visit matched with both selected keywords 
and ICD codes, five matched with ICD 10 codes only, 59 matched 

with keywords only, and 2692 did not match with either keywords 
or ICD 10 codes. In Table 1, it indicates that selected keywords 
have a high specificity (97.9 %) but with a relatively low sensitivity  
(16.7 %) and PPV (1.7%).

Conclusions
Selected keywords and ICD 10 codes from facilities sending triage 

notes were used to evaluate the surveillance system on identifying 
infections following a surgical procedure through analysis of ED 
triage note field. We also reviewed all NJ ED data during the same 
study period for other facilities not sending triage notes. It indicated 
that several key ICD codes, e.g. ICD code T81.4, infections following 
a surgical procedure, have been included in many facilities. This 
analysis will be repeated as more EDs participate in EpiCenter 
with triage notes and other data fields to refine the keywords and to 
improve the sensitivity and PPV.

Table 1: Sensitivity, specificity and PPV calculations of selected keywords 
applied in triage notes based on the ICD 10 codes related to infections 
following a surgical procedure.

Keywords
Healthcare-Associated Infections; Surgical Site Infections; EpiCenter; 
New Jersey

Acknowledgments
Elizabeth Kostial, Kristen Weiss and Jason Liggett (HMS).

References
1. Travers DA, Barnett C, Ising A, Waller A. Timeliness of emergency

department diagnoses for syndromic surveillance. Proceedings of the 
American Medical Informatics Association; 2006 :769-773.

2. Healthy People 2020. Washington, DC: U.S. Department of Health
and Human Services, Office of Disease Prevention and Health 
Promotion. Date URL was accessed July 12 2016. Available from:  
https://www.healthypeople.gov/2020/topics-objectives/topic/
healthcare-associated-infections

3. Health Quality Measures of New Zealand. Excess Length of Stay
Associated with Postoperative Sepsis. Date URL was accessed July
12 2016. Available from: http://www.hqmnz.org.nz/library/Excess_
Length_of_Stay_Associated_with_Postoperative_Sepsis

*Pinar Erdogdu
E-mail: pinar.erdogdu@doh.nj.gov

Online Journal of Public Health Informatics * ISSN 1947-2579 * http://ojphi.org * 9(1):e40, 2017