Medical documentation by practicing physicians has been a challenge since the creation of the medical chart. During his years of practice, Dr. Will Carracino, chief medical information officer at Florida-based not-for-profit health system Lee Health, recalls struggling to keep up with documentation from the first day he became a physician.

THE PROBLEM

“It is challenging to put into the chart exactly what will convey the feeling the patient is experiencing,” Carracino explained. “While I was in practice, I would buy a very nice fountain pen to entice myself to spend additional time charting, convincing myself I’d enjoy charting more with this addition.

“Medical doctors appreciate the time spent serving and supporting each patient individually,” he continued. “Charting is time-consuming, which takes away from the limited office time a physician has with patients. The process affects the quality of patient-physician interactions, as well. Note-taking, often with a computer, can feel distracting and intrusive for both the physician and the patient.”

Though the process is at times burdensome, clinical documentation is a vital part of the role of a practicing physician. It is the record keeping of the patient experience. The information and summary offered, accessible to the patient as well, supports the improved health and wellness of each individual.

PROPOSAL

“Historically, we’ve looked at a variety of solutions for medical documentation,” Carracino noted. “An opportunity using generative AI presented as a solution that could enhance the patient-physician experience together. Previously, using human transcription was a tested solution, but this option proved both costly and included a risk of human error.

“More than a decade ago, we started using voice recognition and natural language processing,” he continued. “The initial iterations were cumbersome. I used to explain that the natural language processing voice recognition software is like trying to train a spouse. There are things you can change over time, but for other things, you just need to work within the structure of the relationship.”

Generative AI promised to require less training, restore the physician’s ability to interact with the patient without technology affecting the conversation, and give the physician valuable time back both for patient experiences and personal opportunities, helping prevent burnout, he added.

MEETING THE CHALLENGE

The new generative AI functionality is a game-changer, Carracino stated.

“Simply by having a listening device in the room after receiving consent from the patient, the conversation ensues much like you might have one in a coffee shop, except with medical concerns,” he explained. “Gone are the times of the patient watching as the doctor taps away on a keyboard to document their experience. The artificial intelligence focuses on the medical concerns, extracts the social niceties, and organizes the note into an easy-to-read, well-formatted medical document.

“We started with a pilot group of 20 providers,” he continued. “The profile of the providers included internal medicine, cardiology, pain management and pediatrics. Our plan was to find providers who were capable of new technology yet had the potential to improve their document time or patient throughput.”

Lee Health’s AI transcription technology, Abridge, integrates seamlessly into the Epic electronic health record. Providers find the process much less burdensome than the previous documentation techniques.

RESULTS

Lee Health still is in the process of validating exact outcomes data, but Carracino outlined five beneficial results so far:

  • Improved note completion, now occurring in the same day for more than half of the physicians. Improved provider efficiency and satisfaction; this is vital to supporting providers and combating burnout. Based on the pilot group’s pre/post survey questions, 74% of clinicians spent less time working outside of work hours, which improved provider satisfaction and work-life balance.

  • 86% of physicians now feel they have adequate time to document. Lee Health is monitoring the impact of this to improve access for patients by allowing the providers to add an additional patient per day to their schedule. This allows the patients to schedule in a more expeditious time frame, already adding two patients per week on average. With time to document during their regularly scheduled work, physicians are able to complete their work.

  • Reduction in documentation errors. Enhanced time for charting ensures doctors can input detailed and accurate patient information, decreasing the likelihood of mistakes.

  • Patients benefit from more attentive care and detailed consultations, leading to higher overall satisfaction with their healthcare experience.

  • Lee Health anticipates improvement in physician retention rates. With decreased burnout, fewer doctors feel a need to leave the profession, which helps maintain a stable and experienced medical workforce, further contributing to the overall quality of care. This improvement in workflow efficiency allows doctors to maintain better work-life balance, reducing stress and improving job satisfaction.

ADVICE FOR OTHERS

“I don’t think anyone can deny that generative artificial intelligence is going to change medical care in our country,” Carracino stated. “The question is, How can it occur in a safe, productive and effective fashion?

“We have an artificial intelligence governance committee at our organization and this committee is made up of both operational and clinical leaders,” he advised others. “The committee looks at the value of implementing new technology and opportunities to analyze what it can bring to the patient and the provider.”

While every product has a purpose, not all of them will align with Lee Health’s mission and vision as a health system.

“This is going to be an exciting segment of our medical history, and we do need to embrace it appropriately with both the excitement vision casting it deserves as well as using our classical techniques of product evaluation for value and safety,” he concluded.

The HIMSS AI in Healthcare Forum is scheduled to take place September 5-6 in Boston. Learn more and register.

Follow Bill’s HIT coverage on LinkedIn: Bill Siwicki
Email him: bsiwicki@himss.org
Healthcare IT News is a HIMSS Media publication.



Source link

Leave a Reply

Your email address will not be published. Required fields are marked *

Related Posts