Article

Managing the ECG Data Deluge by Working Smarter, Not Harder

In 2022, experts predict that more than 300 million health wearables will be distributed across the globe—many of them smartwatches that perform basic functions such as counting steps, tracking stress, and assessing heart rate.1

The proliferation of these devices invites a deluge of data that often goes unused in the clinical setting. Many clinicians have become understandably disinclined to consider wearable data, not only because of accuracy and privacy concerns but also due to the challenges inherent in making those insights more actionable.2

If a patient comes in after a smartwatch alert about heart rate, for example, that insight may have little bearing on the subsequent exam and workup. Why bother deciphering the patient's personal data when you're going to run a 12-lead ECG?

But despite the reluctance to use wearable data, it can meaningfully inform decision-making in the right contexts and with the right tools. Stories abound of smartwatches discovering Afib that might not have been detected otherwise.3 And many cardiologists have, themselves, discussed the promise of wearables, noting the transition in digital health from disease management to detection and prevention.4

Even so, wearables require clinicians to contend with the massive data those devices generate, which often takes time away from busy practitioners. Moreover, wearable users tend to skew younger, toward people who perhaps haven't experienced cardiac issues just yet. As they age, their data will inevitably pile atop this growing mountain of insights to be used in the cardiology setting.

At some point, the wave of wearable data will become too large to ignore. To improve outcomes everywhere, now and in the future, physicians and healthcare stakeholders need to work smarter—not necessarily harder—as they manage ECG data from various sources.

The Potential Impact of ECG Integration

What if a cardiologist could more easily access wearables' single-lead ECG trendlines? This could be possible if the data were integrated into the clinician's workflow so that it became less burdensome and more immediately beneficial.

Such is the premise underlying ECG integration, which consolidates inputs from external and internal sources for a standardized dataset (accessible from the EMR) that speaks a common language. With this connectivity, the question of data integration becomes less about how advanced the devices are and more about the intelligence of the overall system processing the data.

Predicting this need, GE Healthcare has been designing a data integration solution supported by AI and machine learning to bring consolidated models into practice. The company has worked to collect ECG information across different origin points and make it accessible in one spot for clinicians, following many of the lessons learned from broader data interoperability initiatives. From there, algorithmic models can help to standardize and optimize interpretation of that combined data to help providers feel more assured of the insights' reliability, and more confident in their own ability to spot patterns, trends, and events of concern.

Evolving Toward Criteria-Based Routing

Even though a systematized analysis by AI stands to streamline ECG interpretation—especially given the confluence of data both inside and outside the hospital—clinical judgment still requires human oversight.

What computerized interpretations can do is support criteria-based routing. This model routes patients to recommended interventions by their individualized ECG readings, as opposed to probabilistic routing, which establishes care planning for everyone based on the aggregate. Criteria routing has already shown promise across disciplines such as radiology and sepsis detection.

In a future scenario, for example, the AI might dually assess data from the patient's smartwatch with previous ECG baselines from the clinic. Taking what is known about that person's history and recent readings, it could detect a significant change in the underlying rhythm and alert the physician via the EMR to issue a Holter monitor.

Researchers are just scratching the surface of what's possible in these and other applications, but the technology undoubtedly offers potential—if not to diagnose, then at least to detect abnormalities that indicate a diagnosis may be needed.

The Role of Alerting and Escalation

Another practice that could simplify the overflow of data is escalation alerting. As one cardiologist suggests, adopting EMR workflows that deliver automated alerts to medical assistants, nurse practitioners, or physicians could help to prioritize and stratify new data streams of particular concern. The author adds that his own center established such a system with alerts color-coded between red, yellow, and green and found them successful in tracking the large volume of people receiving remote physiological monitoring.4

Similar telemonitoring efforts that embrace varying degrees of automation have been deployed. In one blood pressure monitoring study, patients' at-home devices remotely sent readings straight to the EMR, though those readings were then escalated manually by nursing staff.5

Stay on top of cardiology trends and best practices by browsing our Diagnostic ECG Clinical Insights Center.

Risks and Rewards of Data Integration for DEI

Wearable data integration may also promote digital tool applications in support of diversity, equity, and inclusion (DEI) efforts. When data gets submitted and centralized in the EMR—and considered in tandem with the rest of the patient's chart—patients may receive a more accessible healthcare experience, free of biases known to complicate marginalized groups' care plans.

However, wearables don't serve everyone. Many experts have cautioned against overreliance on wearable data when the people who use these devices are predominantly more privileged or may have conditions such as movement disorders that affect their accuracy.6,7

Policy such as the Centers for Medicare & Medicaid Services (CMS) to cover remote devices, as well as broader 5G, will expand the net cast of wearables and their associated data, but practitioners will still want to approach patient cases in a more nuanced and contextualized manner, as they always have.8,9

Managing the Deluge of Data With Efficiency in Mind

Data scientists often follow a four-part framework concerning data management and stewardship: the FAIR Principles, which indicate that data should be findable, accessible, interoperable, and reusable.10 With millions of wearables teeming with clinically relevant data in circulation, healthcare organizations and clinicians would do well to refer to those principles as they confront the challenges ahead.

Fortunately, many opportunities exist to check off all four criteria and help providers work more efficiently without compromising patient care—whether through ECG integration, alerting tools, or other systems. After all, if consumers are to embrace wearables, clinicians should also embrace opportunities to make the resulting data less daunting to use.

References:

  1. Loucks J, Stewart D, Bucaille A, Crossan G. Wearable Technology in Health Care: Getting Better All the Time. Deloitte Insights. https://www2.deloitte.com/us/en/insights/industry/technology/technology-media-and-telecom-predictions/2022/wearable-technology-healthcare.html. Accessed September 7, 2022.
  2. American Health Information Management Association. Healthcare Data: Navigating the Untapped Clinical Resource. Journal of AHIMA. https://journal.ahima.org/page/healthcare-data-navigating-the-untapped-clinical-resource-1. Accessed September 7, 2022.
  3. Perlow J. How Apple Watch Saved My Lfe. ZDNET. https://www.zdnet.com/article/how-apple-watch-saved-my-life/. Accessed September 7, 2022.
  4. American College of Cardiology. Wearable Technology in 2021: Five Burning Questions Cardiologists Are Asking. ACC.org. https://www.acc.org/latest-in-cardiology/articles/2021/03/01/01/42/wearable-technology-in-2021-five-burning-questions-cardiologists-are-asking. Accessed September 7, 2022.
  5. American Heart Association. Remote Blood Pressure Monitoring Beneficial for Stroke Survivors in Under-Resourced Areas. Heart.org. https://newsroom.heart.org/news/remote-blood-pressure-monitoring-beneficial-for-stroke-survivors-in-under-resourced-areas. Accessed September 7, 2022.
  6. Jercich K. Access to Wearables Could Become a Social Determinant of Health, Researchers Warn. Healthcare IT News. https://www.healthcareitnews.com/news/access-wearables-could-become-social-determinant-health-researchers-warn. Accessed September 7, 2022.
  7. Wendel N, Macpherson CE, Webber K, et al. Accuracy of Activity Trackers in Parkinson Disease: Should We Prescribe Them? Physical Therapy. 2018;98(8):705-714. doi:10.1093/ptj/pzy054
  8. Wicklund E. CMS Expands Medicare Coverage for Remote Patient Monitoring. HealthLeaders. https://www.healthleadersmedia.com/innovation/cms-expands-medicare-coverage-remote-patient-monitoring. Accessed September 7, 2022.
  9. Alsever J. With 5G, Wearable Devices are Expected to Become Even More Sci-Fi. Fortune. https://fortune.com/2020/03/24/5g-wearable-devices/. Accessed September 7, 2022.
  10. GO Fair. FAIR Principles. Go-FAIR.org. https://www.go-fair.org/fair-principles/. Accessed September 7, 2022.

‌