Although AI capabilities for EHR systems are limited, integrated delivery networks are working on using AI to make EHRs more flexible and intelligent, three authors write in Harvard Business Review.
"While AI is being applied in EHR systems principally to improve data discovery and extraction and personalize treatment recommendations, it has great potential to make EHRs more user- friendly," the authors wrote. "This is a critical goal, as EHRs are complicated and hard to use and are often cited as contributing to clinician burnout."
The article was written by Thomas Davenport, PhD, president's distinguished professor in management and IT at Babson College in Wellesley, Mass.; Tonya Hongsermeier, MD, vice president and chief medical information officer at Burlington, Mass.-based Lahey Health and Kimberly Alba McCord, PhD candidate at the University Hospital Basel in Switzerland.
Four ways delivery networks are using AI to make EHR systems more flexible:
1. Data extraction from faxes, clinical notes. The authors highlighted several examples of how delivery networks are using AI in patient data. Providers can already pull data from faxes at OneMedical, a membership-based practice focused on improving healthcare accessibility, or by using athenahealth's cloud-based EHR. Healthcare technology and services company Flatiron Health's human "abstractors" review provider notes and extract structured data, using AI to recognize key terms and reveal data insights. Additionally, Amazon Web Services recently launched a cloud-based service where AI pulls out and indexes data from clinical notes.
2. Diagnostic and/or predictive algorithms. Google is teaming up with delivery networks to develop prediction models from big data to alert clinicians to potentially life-threatening conditions such as sepsis and heart failure. Google and several other startups are also creating AI-derived image interpretation algorithms. Healthcare technology company Jvion offers a "clinical success machine" that identifies patients at the highest risk for an adverse clinical event and those most likely to respond to treatment. Each of these tools could be integrated into EHRs to provide decision support and guide treatment strategies, the authors wrote.
3. Clinical documentation and data entry. "Capturing clinical notes with natural language processing allows clinicians to focus on their patients rather than keyboards and screens," the authors wrote. Healthcare solutions company Nuance offers AI-supported tools that integrate with commercial EHRs to support data collection and clinical note composition.
4. Clinical decision support. Decision support was previously generic and rule-based, the authors said. Now, machine-learning solutions that learn from new data and enable more personalized care are coming from various vendors, such as IBM Watson, Change Healthcare and AllScripts.
More articles on EHRs:
5 most-read EHR stories in 2018
16 hospitals, health systems seeking Allscripts, Cerner, Epic, Meditech talent
Jonathan Bush: Running a company with an activist investor is like having a 'gun to your head'