Diplomate Perspectives

Patients Over Paperwork: How Dr. Eric Jemison Used Artificial Intelligence to Ease the Burden of Electronic Health Records

"AI/ML has brought a more human element back to patient care. I feel it in my visits. I want to make sure I have time for patients, to be their voice and advocate. Being more focused on each individual is so rewarding," said Dr. Jemison.

Dr. Eric Jemison

Electronic Health Records (EHR) have long been considered a necessary evil in the world of medicine. Physicians across the country complete their patients’ EHR in a variety of different ways. Some handle the whole burden of completion themselves, typing or recording during visits and/or completing the paperwork later. Others bring assistants or scribes into the room to help manage the workload.

Now, with recent advancements in the world of artificial intelligence/machine learning (AI/ML), there is a new pathway available for physicians to ease the burden of EHR and hopefully improve patient care at the same time.

For Dr. Eric Jemison, ABFM Diplomate and Washington-based family physician, the use of an AI/ML program has created positive improvements within his practice. He shared this story with ABFM in a recent Performance Improvement activity submission and follow-up interview.

“By using AI/ML capabilities, doctors can quickly produce detailed notes of each patient encounter. This not only ensures that all relevant data is captured, but affords us valuable time with our patients,” Dr. Jemison explained. “Artificial Intelligence can also review patient records and sift through a patient’s history to locate key details for diagnosis.”

Dr. Jemison’s practice improvements began when his clinic implemented the Dragon Ambient Experience (DAX) AI/ML tool. The program has many uses, including audio recording and real time transcription during physician/patient interactions.

“My notes from each patient visit were certainly more efficient with DAX,” he said. “I didn’t have to break up communication with patients to type important data points, but it was still taking hours to complete each day’s EHR.”

The solution for Dr. Jemison came with the release of the DAX Copilot, an enhanced version of the AI/ML program. While every physician’s EHR needs are different, this alternative to the standard DAX program fit Dr. Jemison’s needs and improved his practice.

After each patient encounter, Dr. Jemison documents his physical findings and combines it with the recordings of the DAX Copilot program. Later, when he’s ready for a final review of the encounter, there’s a draft of the entire patient visit already waiting for him to authorize and close the EHR encounter.

“When you walk out of the room, you have a useful draft of the interaction within minutes,” Dr. Jemison said. “I still have to review everything and make sure the data is accurate, but things are so much faster. AI/ML has brought a more human element back to patient care . I feel it in my visits. I want to make sure I have time for patients, to be their voice and advocate. Being more focused on each individual is so rewarding.”

While some of Dr. Jemison’s colleagues have taken up use of either the DAX or Copilot tools, others are more skeptical of relying on this new technology. “The good thing is that you don’t have to use the program for every encounter,” he explained. “There are gradients to how you utilize it. It can be simple documentation translation, or it can be customized with result notes and more personalized communication.”

Despite the possibilities associated with AI/ML tools in a clinical setting, it’s not a perfect system. Dr. Jemison points out that it may misgender patients, for example. “Something like that is an easy fix, as I can tell the program, ‘The patent prefers male or female pronouns, or the patient is non-binary,’” he said, “And it will automatically update with correct data.”

The other concern Dr. Jemison remains cognizant of is the possibility of becoming too reliant on an AI/ML assistant. “That’s a valid consideration,” he said. “But as providers (physicians) we still have to be up on our knowledge and our skills. We still have to complete and finalize our EHR notes. We still have to make our own decisions for each patient.”

While there are concerns to be mindful of, AI/ML has proven to be a useful tool for physicians who embrace its possibilities. No matter how you choose to assist your patients, there are opportunities for new ideas still to be discovered.

Thank you, Dr. Jemison, for sharing your story with ABFM.