Improving Inferences Regarding Patient Events Using Emergency Medical Services Response-Based Data - 01/04/26

Doi : 10.1016/j.annemergmed.2026.02.018

Christopher N. Morrison, PhD ^a, Brady R. Bushover, MPH ^a, Remle P. Crowe, PhD ^b, Carrie W. Mills, PhD, MPH ^c, Alexander X. Lo, MD, PhD ^d,^e, Andrew G. Rundle, DrPH, FNYAM ^a,^⁎

^a Department of Epidemiology, Mailman School of Public Health, Columbia University, New York, NY

^b Department of Research, ESO Solutions, Austin, TX

^c Global Medical Affairs, Emergent BioSolutions, Gaithersburg, MD

^d Department of Emergency Medicine, Northwestern University Feinberg School of Medicine, Chicago, IL

^e Center for Health Services and Outcomes Research, Northwestern University Feinberg School of Medicine, Chicago, IL

^∗ Corresponding Author.

Sous presse. Épreuves corrigées par l'auteur. Disponible en ligne depuis le Wednesday 01 April 2026

Abstract

Study objective

The National Emergency Medical Services Information System public release data set is an important tool for researching emergency medical services (EMS) responses. However, multiple EMS units might attend to a single patient and the data are organized by EMS response, presenting challenges to inferences about patient-level events. We test whether data on time of the 911 call and patient characteristics can be used to screen for multiple EMS records that reflect a single patient encounter.

Methods

In this cross-sectional study, we used data on EMS responses to assaults in New York City in 2024 to identify EMS responses that had identical data for 911 call time, patient age, sex, race/ethnicity, and longitude-latitude where the patient was encountered. EMS responses with identical data for all of these variables were assumed to have attended to a single patient. We assessed the validity of matching on 911 call time, patient age, sex, and race/ethnicity to identify instances where separate EMS responses matched for these variables plus location.

Results

Of 32,202 EMS responses, 5,143 responses matched other responses for all variables, suggesting that there were 26,451 patients encounters. Matching on permutations of variables for time of 911 call, patient age, sex, race, and ethnicity had 100% sensitivity and a high specificity (range 91.3% to 98.6%) for identifying responses that matched on all of these variables plus longitude-latitude.

Conclusion

Data available in the National Emergency Medical Services Information System public release data set may potentially be used to screen for duplicate EMS responses improving inferences about patient-level events.

Le texte complet de cet article est disponible en PDF.

Keywords : NEMSIS, Research methods, EMS

Plan

Introduction

Background

Importance

Goals of This Investigation

Methods

Study Design and Setting

Selection of Participants

	Please see page XX for the Editor’s Capsule Summary of this article.
	Supervising editor: Theodore R. Delbridge, MD, MPH. Specific detailed information about possible conflict of interest for individual editors is available at editors .
	Author contributions : CNM conceptualized the analyses plan, co-wrote first draft of the manuscript and edited drafts. BRB conducted data analyses. RPC helped conceptualize the research question, edited drafts of the manuscript. CWM helped conceptualize the research question, edited drafts of the manuscript. AXL helped conceptualize the research question, edited drafts of the manuscript. AGR conceptualized the research question, conceptualized the analyses plan, co-wrote first draft of the manuscript and edited drafts. AGR takes responsibility for the paper as a whole.
	Data sharing statement : The deidentified dataset and data dictionary for this research are available upon request, from the date of article publication by contacting Christopher N. Morrison, PhD, via email at christopher.morrison@yale.edu .
	All authors attest to meeting the four ICMJE.org authorship criteria: (1) Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; AND (2) Drafting the work or revising it critically for important intellectual content; AND (3) Final approval of the version to be published; AND (4) Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
	Funding and support: By Annals’ policy, all authors are required to disclose any and all commercial, financial, and other relationships in any way related to the subject of this article as per ICMJE conflict of interest guidelines (see www.icmje.org ). Drs. Morrison and Rundle were supported in part by Emergent BioSolutions through a contract with Columbia University. Drs. Morrison and Rundle were also provided support by Columbia Center for Injury Science and Prevention, Centers for Disease Control and Prevention (CDC) grant no. R49CE003555. Dr. Crowe is a current employee of ESO Solutions. Dr. Mills is a current employee of Emergent BioSolutions and holds equity in the company. The other authors have stated that no such relationships exist.
	Publication dates: Received for publication September 4, 2025. Revisions received January 13, 2026; February 10, 2026. Accepted for publication February 16, 2026.