The National Emergency Medical Services Information System (NEMSIS) is the national system used to collect, store, and share data from EMS services in US states and territories. The NEMSIS uniform dataset and database help local, state and national EMS stakeholders more accurately assess EMS needs and performance, as well as support better strategic planning for the EMS systems of tomorrow. Data from NEMSIS are also used to help benchmark performance, determine the effectiveness of clinical interventions, and facilitate cost-benefit analyses. NEMSIS is a program of NHTSA’s Office of EMS and is hosted by the University of Utah.

These data are “event-based” and not “patient-based”. That is, a single patient may be represented in more than one record for a variety of reasons. A patient may request EMS assistance frequently, and therefore, be represented in the dataset more than once. In addition, several agencies may respond to the same event (i.e., one patient) and each submit a patient care record to the National EMS Database. Thus, the dataset is referred to as a registry of “EMS activations.”

The dataset does not contain information that identifies patients, EMS agencies, receiving hospitals, or reporting states. EMS events submitted by states to NEMSIS do not necessarily represent all EMS events occurring within a state. In addition, states may vary in criteria used to determine the types of EMS events submitted to the NEMSIS dataset.

NEMSIS version 3 data are available from 2017 to 2023. Version 2 data are available from 2009 to 2016 but require mapping and translation to version 3 data elements. Users are advised to open a support ticket to discuss their project if they require data from prior to 2017.

Emergency department and hospital discharge status are available for less than 2% of events recorded in the National EMS Information System (NEMSIS) Public Release Research dataset. The purpose of this project was to develop a binary (“dead” vs. “alive”) end-of-event outcome indicator for the NEMSIS dataset. The data dictionary for the Version 3 NEMSIS dataset was evaluated to identify elements and codes providing information about a patient's end-of-event status—defined as the point at which EMS providers stopped providing care for an encountered patient, whether at the scene of the event or the transport destination. Those element and code combinations were then used to test the criteria using the NEMSIS-2017 dataset. After revising the criteria based on the NEMSIS-2017 results, the final criteria were then applied to the 2018 NEMSIS dataset. To assess representativeness, the characteristics of events with a determinable outcome were compared to those of the entire dataset. To assess accuracy, the end-of-event indicator was compared with the final reported outcome for patients with a known emergency department disposition. Eighteen NEMSIS element and code combinations suggest a patient was likely “dead” at the end of EMS care, and 15 combinations suggest a patient was likely “alive” at the end of EMS care. A binary end-of-event outcome indicator could be determined for 13,045,887 (98.6%) of the 13,229,079 NEMSIS-2018 9-1-1 initiated ground EMS responses in which patient contact was established, and for 132,728 (89.1%) of the 148,963 events with documented cardiac arrest. The characteristics of the events with determinable end-of-event outcomes did not differ from those of the full dataset. Among patients with a known outcome, 99.6% of those with an “alive” end-of-event indicator were in fact alive at the time of emergency department disposition. A binary end-of-event outcome indicator can be determined for 98.6% of 9-1-1 initiated ground EMS scene responses and 89.1% of cardiac arrests included in the NEMSIS dataset. The events with a determinable outcome appear representative of the larger dataset and the end-of-event indicators are generally consistent with reported emergency department outcomes.

The National Emergency Medical Services Information System (NEMSIS) is the national database that is used to store EMS data from the U.S. States and Territories. NEMSIS is a universal standard for how patient care information resulting from an emergency 911 call for assistance is collected. NEMSIS is a collaborative system to improve patient care through the standardization, aggregation, and utilization of point of care EMS data at a local, state and national level. NEMSIS is a product of NHTSA’s Office of EMS and in collaboration with the University of Utah is the host of the Technical Assistance Center.

This dataset originates from the National EMS Information System (NEMSIS) database. This dataset has selected severe trauma events where air transport was utilized.

This dataset originates from the National EMS Information System (NEMSIS) database. This dataset has selected trauma events where ground transport was utilized and includes time and rurality elements.

The National Emergency Medical Services Information System (NEMSIS) provides a robust set of data to evaluate prehospital care. However, a major limitation is that the vast majority of the records lack a definitive outcome. This study aimed to evaluate the performance of a recently proposed method (“MLB” method) to impute missing end-of-EMS-event outcomes (“dead” or “alive”) for patient care reports in the NEMSIS public research dataset. This study reproduced the recently published method for patient outcome imputation in the NEMSIS database and replicated the results for years 2017 through 2022 (n = 686,075). We performed statistical analyses leveraging an array of established performance metrics for binary classification from the machine learning literature. Evaluation metrics included overall accuracy, true positive rate, true negative rate, balanced accuracy, precision, F1 score, Cohen’s Kappa coefficient, Matthews’ coefficient, Hamming loss, the Jaccard similarity score, and the receiver operating characteristic/area under the curve. Extended metrics show consistently good imputation performance from year-to-year but reveal weakness in accurately indicating the minority class: e.g., after adjustments for conflicting labels, “dead” prediction accuracy is 77.7% for 2018 and 61.8% over the six-year NEMSIS sub-sample, even though overall accuracy is 98.8%. Slight over-fitting is also present. This study found that the recently published MLB method produced reasonably good “dead” or “alive” indicators. We recommend reporting of True Positive Rate (“dead” prediction accuracy) and True Negative Rate (“alive” prediction accuracy) when applying the imputation method for analyses of NEMSIS data. More attention by EMS clinicians to complete documentation of target NEMSIS elements can further improve the method’s performance.

Characteristics of EMS encounters involving benzodiazepine-treated seizures in the NEMSIS database, 2010–2014.