Eliminating Human Error with Integrated, Tech-Enabled ECG Systems

GE Healthcare

By Sarah Handzel, BSN, RN

Human error remains a significant factor in mortality rates among patients nationwide. While mortality rates have generally fallen over time, a number of adverse events can still be attributed to clinician error. In many cases, delayed diagnosis, misdiagnosis, or other diagnostic errors occur due to a healthcare professional's lack of training or inability to interpret standard diagnostic tests, such as ECG, correctly.

To improve the situation, focus has shifted to the integration of computerized learning systems that support collection, analysis, and storage of medical information. New ECG management systems incorporate a variety of features that can assist clinicians with ECG interpretation, protect patient data, and streamline workflows. Such systems have become critical tools for diagnosing cardiovascular issues quickly and accurately, planning appropriate treatment, and monitoring patient response to therapy.

The Cost of Human Error

Healthcare professionals and organizations generally recognize two specific types of medical errors: errors of omission, which occur as a result of actions not taken, and errors of commission, which are due to wrong actions being taken. According to information published in StatPearls, as many as 400,000 patients experience some type of preventable harm each year, which costs the healthcare system around $20 billion annually.

Many medical errors involve misuse of equipment such as ECG, and harmful events are not limited to one specific practice area. For example, incomplete preoperative assessments resulting from failure to note abnormal ECGs may be a key risk factor for preventable surgical errors. But primary diagnostic errors are also common—in many cases, these mistakes result from a clinician's inability to cross-reference newly collected medical data with old data from the same patient.

A recent study published in the Journal of Emergency Nursing shows that ECG misinterpretation by emergency medical service (EMS) providers leads to one in every three myocardial infarctions being undertriaged upon the patient's arrival at the hospital. After obtaining a 10 second 12-lead ECG for each STEMI-suspected patient, EMS providers classified individuals as high or low risk for STEMI. High risk ECGs were immediately transmitted to the receiving hospital for physician interpretation/confirmation and possible activation of the CCL. These high risk ECGs were also premanently stored within the hospital's EMR system and were available for comparison review at any time. This was especially beneficial, given that up to 20% of prehospital ischemic changes spontaneously resolve before ED arrival, according to the study authors.

However, low risk ECGs were not transmitted and were only temporarily stored in the memory of EMS monitors. These low risk ECGs were also not integrated into any type of cardiology management system for future review. According to the study authors, the strongest predictors of MI misclassification was ECG misinterpretation coupled with no previous patient history of coronary revascularization. But because this information was not stored, there was no way to compare prehospital ECG with ED ECG obtained upon arrival.

The same study concludes that hospital providers, such as nurses and physicians, should always reinterpret ECGs for subtle changes after the patient is admitted to the facility. However, factors such as emergency department overcrowding, increased complexity of each patient's healthcare needs, and inadequate staffing levels may prevent clinicians from performing secondary ECG review. Additionally, these factors may prevent compliance with the American Heart Association's 10 minute door-to-ECG benchmark, which may cause further diagnostic and treatment delays.

The Benefits of Integrated ECG Systems

Today's ECG information management systems are designed to provide physicians with pertinent patient information and integrate with a hospital's existing patient records and information technology systems easily. Unlike older systems, which may still use outdated technology like faxes or scanners to communicate patient information, modern cardiology management systems allow clinicians to view real-time data as it is collected. This information is especially beneficial for patients with suspected cardiovascular events, as research in Heart indicates prehospital ECGs offer survival advantages by reducing time-to-ECG and time-to-treatment.

These systems can store prehospital and hospital ECGs, display multiple waveforms at the same time, offer graphical event summaries, help to identify trends, and link various providers and departments across the healthcare system. Many also allow comparison of multiple ECG recordings for a single patient on a one screen using statements, waveforms, and measurements to help physicians accurately interpret data and improve diagnostic accuracy. This is especially important, as outdated systems may not alert the user to any previously recorded ECG data.

With the advantage of integrated ECG management systems, information from multiple monitoring sources is combined with historical patient data to present a clear and complete picture of each patient's unique healthcare needs. Automated cardiovascular management systems help facilitate the diagnostic process by offering fast, easy access to previously recorded ECGs.


To learn more about the power of the ECG in today's clinical landscape, browse our Diagnostic ECG Clinical Insights Center.


How AI Can Help Reduce ECG Interpretation Error

In many cases, these systems' interpretive algorithms assist physicians with ECG analysis so that treatment decisions can be made quickly. These algorithms are evolving as technology improves, and many cardiovascular management systems currently rely on neural networks and machine learning software to improve diagnostic capabilities and track cardiovascular events over time. AI integration with ECG information systems is a particularly prominent trend to watch for in diagnostic ECG.

According to a recent study published in Circulation: Cardiovascular Quality and Outcomes, many of the machine learning methods being applied actually match or outperform interpretation by practicing physicians. Proper interpretation of ECG data is vital to ensuring diagnostic accuracy and predicting patient outcomes, so there is enormous potential in using machine learning systems to interpret large quantities of patient data over time.

While it may be impossible to prevent human error in medicine completely, modern cardiovascular information management systems can help physicians better interpret ECG recordings to make accurate diagnoses and develop effective treatment plans.


Sarah Handzel, BSN, RN has been writing professionally since 2016 after spending over nine years in clinical practice in various specialties.

The opinions, beliefs, and viewpoints expressed in this article are solely those of the author and do not necessarily reflect the opinions, beliefs, and viewpoints of GE Healthcare. The author is a paid consultant for GE Healthcare and was compensated for creation of this article.