Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Computerized coding of injury narrative data from the National Health Interview Survey.

Helen M Wellman1, Mark R Lehto, Gary S Sorock

  • 1Liberty Mutual Research Institute for Safety, 71 Frankland Road, Hopkinton, MA 01748, USA. helen.wellman@libertymutual.com

Accident; Analysis and Prevention
|December 4, 2003
PubMed
Summary

A computerized method accurately classifies injury narratives into external-cause-of-injury and poisoning (E-code) categories using Fuzzy Bayes logic. This approach improves coding efficiency by filtering cases for manual review.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Impact of schedule V controlled substance classification of gabapentin on adult gabapentin-involved overdose rates, West Virginia, 2016-2019: A controlled time series analysis.

Drug and alcohol dependence·2026
Same author

The Association Between Receiving Medications for Opioid Use Disorder and Human Immunodeficiency Virus Testing: Findings From the Rural Opioid Initiative.

Journal of addiction medicine·2026
Same author

Patterns of Felt Stigma Among Rural-Dwelling People Who Use Drugs: A Latent Class Analysis.

Substance use & misuse·2025
Same author

Higher mortality following SARS-CoV-2 infection in rural versus urban dwellers persists for two years post-infection.

Nature communications·2025
Same author

National Estimates of Work-Related Emergency Department-Treated Finger, Hand, and Wrist Injuries, U.S. 2015-2022.

American journal of preventive medicine·2025
Same author

Analysis of Severity of Finger, Hand, and Wrist Injuries Among Department of Air Force Workers.

Journal of occupational and environmental medicine·2025

Area of Science:

  • Public Health
  • Medical Informatics
  • Injury Surveillance

Background:

  • Accurate classification of injury narratives is crucial for public health surveillance.
  • Manual coding of external-cause-of-injury (E-code) data is time-consuming and resource-intensive.

Purpose of the Study:

  • To evaluate the accuracy of a computerized method for classifying injury narratives into E-code categories.
  • To assess the performance of a Fuzzy Bayesian model in this classification task.

Main Methods:

  • Utilized injury narratives and expert-assigned E-codes from the US National Health Interview Survey (NHIS).
  • Employed a Fuzzy Bayesian model to categorize narratives into 13 E-code categories.
  • Measured sensitivity, specificity, and positive predictive value against expert classifications.

Related Experiment Videos

Main Results:

  • The computer program achieved 82.7% accuracy in classifying 5677 injury narratives.
  • Using multiple-word keywords improved the sensitivity and specificity of the automated coding.
  • The model can identify narratives requiring manual review (approx. 33%) based on probability thresholds.

Conclusions:

  • A Fuzzy Bayes-based computer program accurately categorizes E-codes from injury narratives.
  • The ability to filter cases for manual coding enhances the overall utility and efficiency of the process.