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

Predicting 30-Day Pneumonia Readmissions Using Electronic Health Record Data.

Anil N Makam1,2, Oanh Kieu Nguyen1,2, Christopher Clark3

  • 1Department of Internal Medicine, University of Texas Southwestern Medical Center, Dallas, Texas, USA.

Journal of Hospital Medicine
|April 16, 2017
PubMed
Summary

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

Complexity of Diagnosing Volume Overload.

JAMA·2026
Same author

Disentangling Fatigue from Depression among Survivors of Severe COVID-19.

medRxiv : the preprint server for health sciences·2026
Same author

Evidence Over Eminence: Rethinking Specialty Guidelines Amid the ATS-IDSA Debate.

Clinical infectious diseases : an official publication of the Infectious Diseases Society of America·2026
Same author

Navigating Postacute Care Options for Patients After Hospital Discharge: A Review.

JAMA internal medicine·2026
Same author

The effectiveness and value of sonpiretigene isteparvovec for the treatment of advanced retinitis pigmentosa.

Journal of managed care & specialty pharmacy·2025
Same author

Did Site-Neutral Payment for Long-Term Care Hospitals Reduce Financial Incentives to Time Discharge?

Journal of the American Medical Directors Association·2025

Electronic health record data from the entire hospital stay can accurately predict pneumonia readmission risk. This full-stay model significantly outperforms existing prediction tools, improving patient risk stratification.

Area of Science:

  • Clinical Medicine
  • Health Services Research
  • Predictive Analytics

Background:

  • Pneumonia readmissions are a significant clinical challenge, with existing risk-prediction models demonstrating limited accuracy.
  • Electronic health records (EHR) offer a rich data source that may enhance the prediction of hospital readmissions.

Purpose of the Study:

  • To develop and validate pneumonia-specific risk-prediction models for 30-day readmissions.
  • To compare the predictive performance of models utilizing EHR data from the first day versus the entire hospital stay.

Main Methods:

  • An observational cohort study was conducted using data from 1463 pneumonia hospitalizations across six diverse hospitals.
  • Stepwise-backward selection and cross-validation techniques were employed to develop predictive models.

Related Experiment Videos

  • Models were evaluated based on discrimination (C statistic) and reclassification ability (net reclassification index).

    • Main Results:

    • The full-stay pneumonia-specific model, incorporating discharge disposition and vital sign stability, achieved a C statistic of 0.731.
    • This full-stay model demonstrated superior performance compared to a first-day model (C statistic 0.695) and other established readmission risk scores.
    • The full-stay model showed significantly better discrimination and reclassification of patients by their true readmission risk.

    Conclusions:

    • Utilizing comprehensive EHR data throughout the entire hospitalization provides accurate prediction of pneumonia readmission risk.
    • The developed full-stay model offers improved predictive capability over existing pneumonia-specific and general readmission risk assessment tools.
    • This approach facilitates better identification of high-risk patients, potentially enabling targeted interventions to reduce readmissions.