Internist Dr. Missy Scalise can vividly recall the darkest day of her career.
“When I took my son for his regular checkup at age 5, the doctor asked him the typical questions like, 'What do you do at school?' and 'What excites you?' The doctor then asked him, 'What do you play with when I'm with you?' and he replied, 'I work on the computer.' I was absolutely shocked and still feel sad,” Dr Scalise said.
The “computer work” included recording the clinical documentation required for each patient encounter.
After that appointment, Dr. Scalise began looking for ways to spend more time with her family. She realized the only way she could complete all of her clinical documentation and still be with her family was to wake up at 5 a.m. so she could finish before her family woke up. For years, she spent the early morning hours doing administrative tasks, but nine months ago, her health system began leveraging AI assistants to help offload documentation burden from clinicians.
Not surprisingly, Dr. Scalise, who serves as program director for the internal medicine residency program and chair of the clinician benefits committee at Ascension St. Thomas Hospital in Tennessee, believes these AI-enabled documentation tools will become more widespread among health care workers over the next five years, a sentiment echoed by three other health system leaders interviewed this month.
To them, a clinician documentation assistant is one of the most valuable AI tools hospitals can deploy, as it helps both clinicians and patients by reducing burnout and enabling more meaningful interactions.
Ascension Saint Thomas & Suki
Dr. Scalise began using the AI documentation tool earlier this year as part of Suki's pilot program at his health system, and this month, Ascension St. Thomas announced plans to integrate Suki's AI assistant into its residency program as part of a system-wide rollout.
Redwood City, California-based Suki aims to ease the burden of documentation through its AI-powered voice assistant for physicians. By asking Suki's AI assistant, physicians have instant access to important patient information, including a patient's medications, vital signs, allergies, and surgical history. Physicians can also use Suki's tools to dictate clinical notes, view their weekly schedule, assist with ICD-10 coding, and more.
Suki's assistant can also listen to ambient sound during doctor-patient conversations and write clinical notes. These notes are automatically sent back to the patient's EHR, allowing the doctor to review the notes and make any necessary changes before finalizing them.
As a result of using Suki, Dr. Scalise has regained valuable time in her day that she can now spend with her family. Before the technology, she says, it would have taken her hours to manually create clinical notes.
“At the beginning of the appointment, I ask the patient if it's OK to use (the tool). I set the app on my phone aside, turn it on, and it records the entire conversation. At the end of the appointment, I tap to make a note, and it takes a few minutes to calculate the note. I review all my notes at the end of the day, because I'm usually on to the next patient, and they're usually accurate. It probably takes 15 minutes to edit,” Dr. Scalise explained.
What made Suki's tool stand out to her health system, she said, was how easily the technology could be integrated into the Athenahealth EHR: Integrating an AI assistant into a health system's Athenahealth EHR typically takes about five business days.
Dr. Scalise noted that some AI tools require clinicians to copy and paste generated notes into the EHR. Suki allows clinicians to edit and finalize notes directly within the EHR, reducing the number of clicks overall.
Suki also integrates with other EHRs, including Epic, Cerner and Elation Health, according to the company's website.
In an interview last year, Suki CEO Punit Soni said the company's EHR integration is a key part of its overall strategy.
“It's not AI that's going to be successful in healthcare, it's EHR integration that's going to be successful,” he told MedCity News. “The number of engineers working on[EHR integration]at Suki is probably double the number of engineers working on AI.”
Ochsner Health & DeepScribe
Last month, Ochsner Health announced it was rolling out a different documentation tool across its system. The New Orleans-based health system is rolling out DeepScribe's AI assistant to its 4,700 physicians and affiliated physicians.
Like Suki, San Francisco-based DeepScribe also integrates with several EHRs, including Epic and Athenahealth. The company's tool listens to patient-provider interactions from the outside in and provides a transcript of the conversation. It also creates draft clinical notes that clinicians can edit and submit in the EHR.
Ochsner knew it had to do something to ease the enormous documentation burden on doctors, said Jason Hill, M.D., the health system's chief of innovation.
“In most industries, data entry is typically left to individuals who are at the lowest skilled part of the food chain. Healthcare is unique; the most highly skilled people do the data entry. And it's a lot of data entry,” he said.
He added that most physicians would honestly say that documentation is their least favorite part of the day.
Dr. Hill noted that while most physicians dislike the clinical documentation process, they also recognize it is important. Many performance and quality assessments are based on the content of the clinical record, he said.
“We live in an age where we have huge health records that can tell us everything about a patient, but somehow everyone wants to see the doctor's note to know the patient's condition,” he explained. “They've become a huge thing that serves many purposes. As a doctor, you want to show that you're doing your job, and some doctors are better at showing their work than others.”
At the end of a long day, it's usually hard for doctors to remember a specific interaction with one of the dozens of patients they saw that day, Dr. Hill added.
While searching for a tool to solve the problem, Ochsner tried a tool from DeepScribe and another tool that Dr. Hill declined to name.
DeepScribe's technology has seen high adoption and satisfaction rates among the health system's clinicians, Hill said, and he noted that DeepScribe worked with Ochsner during the pilot process and responded well to clinician feedback.
Clinicians often tell him how DeepScribe’s tools have made a positive difference in their work.
“In my career as a clinical informaticist, I've been involved in many projects, but in all of them, I've never had a physician take a video of themselves and tell me how great it is and send it unsolicited to my inbox. In the pilot, the physician took a selfie video and said, 'This is the best technology I've ever had.' Usually it's the other way around, the physician asks, 'What are you doing to me?'” Hill said, referring to other technology that may be burdensome to healthcare workers.
Kaiser Permanente & Abridge
Last week, Kaiser Permanente also announced plans for a large-scale documentation AI rollout: The health system will roll out Abridge's clinical documentation tools to 40 hospitals and more than 600 clinics across eight states.
Abridge was founded in 2018 by a UPMC cardiologist and two Carnegie Mellon University researchers. The company's tool, integrated with Epic, captures visits, transcribes them and generates clinical notes for doctors to edit in their EHRs.
Desiree Gundrup Dupré, senior vice president of healthcare delivery technology services at Kaiser Permanente, said the health system gathered feedback from both doctors and patients during the Abridge pilot.
“Many physicians commented that for the first time in their careers, they were able to complete charting by the end of the session instead of outside of office hours,” she noted.
Gundrup Dupré added that patients report “feeling much more engaged with their doctor.” Before ambient listening was implemented into consultations, patients often felt like their doctor wasn't listening to them because they were sitting at a computer typing throughout the entire appointment, she noted.
Abridge's technology stood out as the best tool for Kaiser to adopt for three main reasons: First, the tool is scalable and can be easily integrated across the health system's enterprise, and second, it received overwhelmingly positive feedback from clinicians during the pilot period, Gandrup-Dupre said.
Additionally, she noted, Abridge's AI also performed well in validation testing at Kaiser Permanente, where the health system's quality assurance team ran a series of tests on the AI so that it could independently assess whether clinicians were satisfied with the quality of the recordings and notes generated.
Ohio State University Wexner Medical Center & Microsoft DAX
An enterprise-wide rollout is also on the horizon at Ohio State University Wexner Medical Center, which announced this week that it has begun piloting the Microsoft Dragon Ambient eXperience (DAX).
Microsoft's DAX tool was born out of the company's acquisition of Nuance in 2022. Integrated with EHRs from the likes of Epic and Cerner, the AI-powered tool listens to patient-provider conversations to create clinical notes and generate structured documentation in real time.
An internal audit found that the tool saves Ohio State clinicians about 30 minutes to an hour per day, said Dr. Harrison Jackson, one of the physicians participating in the program.
To him, this is a welcome benefit, but not the primary goal of the Microsoft DAX implementation: The goal of the Ohio State pilot, he asserted, is not to save clinicians time, but to improve the quality of the doctor-patient relationship.
“We make more eye contact, we change our body posture away from the computer and toward the patient, and we're able to pick up more nonverbal cues,” Dr. Jackson explains. “It becomes much clearer that we're completely focused on the patient, and that has a huge benefit to the quality of the information we get.”
He said Microsoft DAX stood out as the tool of choice for Ohio State because of its ease of use: In his opinion, most clinicians should be able to use the tool after watching a two-minute instructional video.
Dr. Jackson noted that Ohio State has already conducted 5,000 patient consultations through DAX since July 1. He added that the health system has quickly expanded the program from its original 25 users to 500.
In Dr. Jackson's view, every hospital should have clinical documentation assistants as a service to patients.
“And the better we serve our patients, the more satisfied we are with our jobs,” he asserted.
AI tools are proliferating across many areas of healthcare, but when it comes to clinical documentation, the bottom line is that AI will improve physician efficiency and satisfaction, ultimately positively impacting patients.
Photo: megalopp, Getty Images
