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

Going Beyond Chief Complaints to Identify  
Opioid-Related Emergency Department Visits
Andrew Walsh*
Health Monitoring, Pittsburgh, PA, USA

Objective
To identify heroin- and opioid-related emergency department visits 

using pre-diagnositc data. To demonstrate the value of clinical notes 
to public health surveillance and situational awareness.

Introduction
Overdoses of heroin and prescription opioids are a growing 

cause of mortality in the United States. Deaths from opioids have 
contributed to a rise in the overall mortality rate of middle-aged white 
males during an era when other demographics are experiencing life 
expectancy gains.1 A successful public health intervention to reverse 
this mortality trend requires a detailed understanding of which 
populations are most affected and where those populations live. While 
mortality is the most relevant metric for this emerging challenge, 
increased burden on laboratory facilities can create significant 
delays in obtaining confirmation of which patients died from opioid 
overdoses.

Emergency department visits for opioid overdoses can provide a 
more timely proxy measure of overall opioid use. Unfortunately, chief 
complaints do not always contain an indication of opioid involvement. 
Overdose patients are not always conscious at registration which 
limits the amount of information they can provide. Menu-driven 
registration systems can lump all overdoses together regardless of 
substance. A more complete record of the emergency department 
interaction, such as that provided by triage notes, could provide the 
information necessary to differentiate opioid-related visits from other 
overdoses.

Methods
Emergency department registration data was collected from 

hospitals via the EpiCenter syndromic surveillance system. This 
data included chief complaints, triage notes, discharge disposition, 
and preliminary diagnosis codes. Data elements were linked across a 
given visit using patient identifiers and visit numbers as appropriate.

Heroin- and opioid-related indicators were identified in chief 
complaints and triage notes using regular expressions. These were 
separated into three categories: visits with an indication of overdose, 
visits for withdrawal symptoms, and visits where opioids were 
mentioned in some other context such as history of use. These 
categories were designed to be mutually exclusive.

Regular expression classification results were compared to 
classifications based on opioid-related diagnosis codes.

Results
A total of 2,934,610 ED registrations with triage notes and diagnosis 

codes were collected from 82 hospitals between January 1, 2015 and 
August 21, 2016. Of these encounters, 24,012 referenced opioid use 
in some way; 16,718 mentioned heroin specifically; 3,663 mentioned 
fentanyl specifically; and 5,350 mentioned opioids generically.

Table 1 shows the distribution of heroin-related ED visits across 
categories and source of the indicator. Column totals are not the sum 
of individual row amounts; they have been adjusted so that a given 
registration is only counted once.

Table 2 shows the overlap of heroin-related ED visits between 
sources of indicators. Triage notes showed the least overlap with the 
other two sources, while chief complaints showed the most.

Conclusions
While it is possible to find indicators of opioid use or overdose 

in chief complaint data, that field alone does not provide total 
information about which ED visits are related to opioids. Triage 
notes in particular indicate opioid involvement in a large number 
of visits not identified by other data sources. While many of these 
are simply mentions of opioids, possibly indicating past history of 
use or even in some cases just that questions about opioid use were 
asked, a substantial number of visits with overdose indicators were 
also detected solely from triage note data. These results suggest that 
triage notes can be a valuable additional data source for more complex 
health concerns such as opioid drug use.

Table 1: Heroin-Related ED Visits By Indicator Source and Category

Table 2: Overlap of Heroin-Related ED Visits between Indicators

Keywords
opioid; heroin; overdose; triage note; fentanyl

Acknowledgments
We wish to thank our public health customers for funding support and 
data for this work.

References
1. Case A, Deaton A. Rising morbidity and mortality in midlife among

white non-Hispanic Americans in the 21st century. PNAS 2015 
December; 112 (49).

*Andrew Walsh
E-mail: andy.walsh@hmsinc.com

Online Journal of Public Health Informatics * ISSN 1947-2579 * http://ojphi.org * 9(1):e39, 2017