How AI Assistants Streamline EHR Documentation, Clinician Workload
- Clinicians at OrthoIndy Hospital, Indiana’s first orthopedics specialty hospital and one of the largest of its kind in the country, were spending up to three hours after work and even over the weekend to complete EHR-related tasks. However, after the practice implemented a mobile artificial intelligence (AI) assistant into its EHR, after-hours charting and documentation was a thing of the past.
“When I first started in practice 30 years ago, the medical record was to remind the doctor of when the patient came in for last time, what the direction was, and what the plan is going forward,” Timothy Dicke, MD, president and CEO of OrthoIndy, said in an interview with EHRIntelligence.
“It’s turned into something completely different. It turned into a report card for the government. It turned into a billing measure for the insurance record. If you didn’t dot all the Is, and cross all the Ts, and have all the different templated blocks filled in, you did not have a note.”
Dicke explained the EHR did nothing to improve the patient encounter, but in reality, it just added more time to the clinician workday.
“The note could be absolutely worthless, five pages long, all full of templated information, which may or may not be accurate, and that was supposed to be a better note and a better encounter, but it was not a better encounter,” Dicke said.
“If you wanted to keep that same encounter quality, which most physicians did, it just added to their time, instead of being able to simply jot down a note and have a record, which was meaningful to the physician, it became something else.”
Dicke mentioned that after two decades, a breaking point was hit with clinicians across the country, and OrthoIndy aimed to simplify the process, rather than make it more complex. Up until recently, clinician burden was tolerated by providers and invisible for the most part.
“We don’t utilize an EHR to its full capacity,” Dicke explained. “I don’t think anybody does. It’s kind of like our brains, where we use 10 percent by some measure.”
“With that, is there an easier way to input and get the job done, save time, and use technology to our benefit? Whether it’s an overlay that puts information in, instead of the physician or provider doing it live or after the fact, through either using manual scribes, automated scribes, recording devices, or the voice recognition software.”
After attempting different methods, the overall negative impact of clinician burden still needed to be reduced at OrthoIndy.
“That is something that is a very huge thing, that can lead to the loss of quality time with yourself, your family, and increased burnout rate,” Dicke noted. “It’s been one of those burdens, which is part of that process.”
As a response, Dicke and health IT leaders at OrthoIndy tapped Saykara’s mobile AI assistant, Kara, to aid EHR documentation and restore a sense of work-life balance for its users.
“Kara was not our first attempt of trying to simplify this equation,” Dicke noted. “We have had success, and we have had movement of trying to ease the burden. I think many providers have tried different types of voice recognition versus scribes, versus other programs.”
Dicke also said the hospital attempted to reduce burden by leveraging its patients to input their own data into their patient portal, while the provider was used as a secretary to manually input the rest of the data.
“This EHR-implemented tool is an improved dictation system, where it has a little more ingenuity, and a little more AI in there,” Dicke explained. “The user can actually just act as if he’s dictating into an app, but it goes into the note in a correct fashion.”
While some EHR-implemented dictation tools can put notes into the incorrect place, Dicke said this well-designed tool is easy to implement and the notes are documented in a clean, easy to read interface in the EHR.
“It has that ability of separating the different parts of note structure,” Dicke stated. “You can also use it beyond just the transcription, while adding it to the thought pattern.”
“If you want to order a test or order a surgery, you have a fairly standard approach to that, ‘MRI of the left wrist,’ that can be instantly translated into that order without having to do a separate input from the provider side,” Dicke continued. “Before, it would take several clicks to the order segment of the EHR. Having that automated artificial intelligence aspect to dictation is where this really has been a boost to us.”
Right away, Dicke and OrthoIndy clinicians saw the tool as an immediate time saver. Documentation was now occurring in real-time, compared to waiting an hour or two to document the visit. And while many EHR-implemented tools come with a steep learning curve, that was not the case here.
“With that, there has been a significant load reduction on our workforce,” Dicke said. “Now that the provider is relieved of this time burden, there has been an incentive to actually produce more. The clinician can be more productive in the same time frame.”
“You can also get to a point there that’s a diminishing return, but if that means you could see another two, three, or four patients in that day without an additional burden, that’s all net new, and without an additional time burden to do such,” Dicke concluded. “Being done quicker, as well as being able to document quicker, gives you a boost to perform better.”
Discover Past Posts