EUVIMEDEUVIMED entdecken
Open Access

Predicting Out-of-Hospital Cardiac Arrest in the General Population Using Electronic Health Records

Vorhersage von außerklinischem Herzstillstand in der Allgemeinbevölkerung mithilfe elektronischer Gesundheitsakten

Jessica Perry iD, Jennifer A Brody iD, Christine Fong iD, Jacob E Sunshine iD, Vikas N O'Reilly-Shah iD, Michael R Sayre, Thomas D Rea iD, Noah Simon, Ali Shojaie iD, Nona Sotoodehnia iD, Neal A Chatterjee iD

Circulation · 09.07.2024 · 150 (2) · 102-110

DOI: 10.1161/circulationaha.124.069105 ↗

Auf einen Blick

In einer Fall-Kontroll-Studie sagte ein Machine-Learning-Modell den außerklinischen Herzstillstand in der Allgemeinbevölkerung besser voraus als ein Basis-Modell. Wichtige Prädiktoren waren u. a. ein verlängertes korrigiertes QT-Intervall und Suchterkrankungen.

Fragestellung

Kann man mit Daten aus elektronischen Gesundheitsakten das Risiko für außerklinischen Herzstillstand in der Allgemeinbevölkerung identifizieren?

Methode

Populationsbasierte Fall-Kontroll-Studie mit 2366 OHCA-Fällen und 23 660 alters- und geschlechtsgematchten Kontrollen; EHR-Daten zu Komorbiditäten, EKG, Vitalzeichen und Medikation; externe Validierung.

Ergebnisse

In der externen Validierung lag die Diskrimination der Machine-Learning-Modelle bei einer AUC von 0.80–0.85 und damit über dem Basis-Modell mit 0.66. Bei 99 % Spezifität lag der positive prädiktive Wert bei 2.5 % bis 3.1 % vs. 0.8 %.

Schlussfolgerung der Autor:innen

Machine-Learning-Modelle mit leicht verfügbaren EHR-Daten zeigten eine angemessene Diskrimination und Risikoreicherung für OHCA in der Allgemeinbevölkerung; die Vorhersage erfordert die Berücksichtigung vieler kardiovaskulärer und nichtkardiovaskulärer Faktoren.

Kernaussagen

  • AUC der Machine-Learning-Modelle: 0.80–0.85
  • Basis-Modell mit konventionellen Risikofaktoren: AUC 0.66
  • Wichtige Prädiktoren: verlängertes QT, Suchterkrankung, höherer Herzfrequenz

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

Kurz gefasst

Fachartikel in der Zeitschrift Circulation, erschienen am 09.07.2024. Verfasst von Jessica Perry und 10 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biostatistics (J.P., N.S., A.S.), University of Washington, Seattle. Hauptthemen laut MEDLINE: Electronic Health Records, Out-of-Hospital Cardiac Arrest. Die Arbeit wurde bisher 19-mal zitiert (Crossref) und nennt 27 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIGMS NIH HHS.

Auf einen BlickAbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BACKGROUND: The majority of out-of-hospital cardiac arrests (OHCAs) occur among individuals in the general population, for whom there is no established strategy to identify risk. In this study, we assess the use of electronic health record (EHR) data to identify OHCA in the general population and define salient factors contributing to OHCA risk. METHODS: The analytical cohort included 2366 individuals with OHCA and 23 660 age- and sex-matched controls receiving health care at the University of Washington. Comorbidities, electrocardiographic measures, vital signs, and medication prescription were abstracted from the EHR. The primary outcome was OHCA. Secondary outcomes included shockable and nonshockable OHCA. Model performance including area under the receiver operating characteristic curve and positive predictive value were assessed and adjusted for observed rate of OHCA across the health system. RESULTS: There were significant differences in demographic characteristics, vital signs, electrocardiographic measures, comorbidities, and medication distribution between individuals with OHCA and controls. In external validation, discrimination in machine learning models (area under the receiver operating characteristic curve 0.80–0.85) was superior to a baseline model with conventional cardiovascular risk factors (area under the receiver operating characteristic curve 0.66). At a specificity threshold of 99%, correcting for baseline OHCA incidence across the health system, positive predictive value was 2.5% to 3.1% in machine learning models compared with 0.8% for the baseline model. Longer corrected QT interval, substance abuse disorder, fluid and electrolyte disorder, alcohol abuse, and higher heart rate were identified as salient predictors of OHCA risk across all machine learning models. Established cardiovascular risk factors retained predictive importance for shockable OHCA, but demographic characteristics (minority race, single marital status) and noncardiovascular comorbidities (substance abuse disorder) also contributed to risk prediction. For nonshockable OHCA, a range of salient predictors, including comorbidities, habits, vital signs, demographic characteristics, and electrocardiographic measures, were identified. CONCLUSIONS: In a population-based case–control study, machine learning models incorporating readily available EHR data showed reasonable discrimination and risk enrichment for OHCA in the general population. Salient factors associated with OCHA risk were myriad across the cardiovascular and noncardiovascular spectrum. Public health and tailored strategies for OHCA prediction and prevention will require incorporation of this complexity.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Electronic Health RecordsOut-of-Hospital Cardiac ArrestHumansMaleFemaleMiddle AgedAgedRisk FactorsAdultPredictive Value of TestsRisk AssessmentComorbidityElectrocardiographyMachine LearningCase-Control Studies

Stichwörter der Autor:innen

death, sudden, cardiacheart arrestmachine learning

Autor:innen und Institutionen

  • Department of Biostatistics (J.P., N.S., A.S.), University of Washington, Seattle.
  • Cardiovascular Health Research Unit (J.A.B., N.S., T.D.R.), University of Washington, Seattle.
  • Department of Medicine, Department of Anesthesiology and Pain Medicine (C.F., J.E.S., V.N.O.), University of Washington, Seattle.
  • Department of Emergency Medicine (M.R.S.), University of Washington, Seattle.
  • Seattle Fire Department (M.R.S.), WA.
  • King County Emergency Medical Services (T.D.R.), Seattle, WA.

ORCID-Profile: Jessica Perry, Jennifer A Brody, Christine Fong, Jacob E Sunshine, Vikas N O'Reilly-Shah, Thomas D Rea, Ali Shojaie, Nona Sotoodehnia, Neal A Chatterjee

Förderung

  • NIGMS NIH HHS

Zitieren

APA

Perry, J., Brody, J. A., Fong, C., Sunshine, J. E., O'Reilly-Shah, V. N., Sayre, M. R., Rea, T. D., Simon, N., Shojaie, A., Sotoodehnia, N., & Chatterjee, N. A. (2024). Predicting Out-of-Hospital Cardiac Arrest in the General Population Using Electronic Health Records. Circulation, 150(2), 102-110. https://doi.org/10.1161/circulationaha.124.069105

Vancouver

Perry J, Brody JA, Fong C, Sunshine JE, O'Reilly-Shah VN, Sayre MR, et al. Predicting Out-of-Hospital Cardiac Arrest in the General Population Using Electronic Health Records. Circulation. 2024;150(2):102-110. doi:10.1161/circulationaha.124.069105

BibTeX
@article{Perry2024,
  author = {Perry, Jessica and Brody, Jennifer A and Fong, Christine and Sunshine, Jacob E and O'Reilly-Shah, Vikas N and Sayre, Michael R and Rea, Thomas D and Simon, Noah and Shojaie, Ali and Sotoodehnia, Nona and Chatterjee, Neal A},
  title = {Predicting Out-of-Hospital Cardiac Arrest in the General Population Using Electronic Health Records},
  journal = {Circulation},
  year = {2024},
  volume = {150},
  number = {2},
  pages = {102--110},
  doi = {10.1161/circulationaha.124.069105}
}
Kostenlos starten →