Related Experiment Videos
Clinical experience with a decision support computer program using Bayes' theorem to diagnose chest pain patients
1Department of Cardiology, University of Oslo, Ullevâl Hospital, Oslo, Norway.
Cardiology
|March 7, 2000
Summary
A decision support computer program (DSP) improved patient referrals for acute chest pain. This tool enhanced diagnostic accuracy for acute myocardial infarction and unstable angina, optimizing Coronary Care Unit (CCU) resource use.
Area of Science:
- Cardiology
- Medical Informatics
Background:
- Emergency departments frequently manage patients presenting with acute chest pain.
- Accurate and timely diagnosis is crucial for appropriate patient disposition and resource allocation.
Purpose of the Study:
- To evaluate the effectiveness of a decision support computer program (DSP) in guiding emergency room physician referrals for patients with acute chest pain.
- To assess the DSP's impact on diagnostic accuracy and resource utilization in the Coronary Care Unit (CCU).
Main Methods:
- A DSP utilizing Bayes' theorem and 38 clinical variables was employed by 32 physicians over six months for 493 patients.
- Physicians used DSP estimates alongside clinical judgment and ECG findings to refer patients to CCU or general ward.
Main Results:
- The DSP correctly classified 89% of acute ischemic heart disease cases (myocardial infarction or unstable angina).
- DSP use reduced unnecessary CCU referrals from 35% to 19% and misallocated patients from 13% to 10%.
Conclusions:
- The DSP enhances the accuracy of patient referrals from the emergency room for acute chest pain.
- Implementing this DSP can optimize therapy and resource management within the CCU.