I have never had another person in the exam room with my doctor and me during a physician visit after childhood. With the introduction of electronic medical records many people complained there was hardly even a doctor in the room listening to them. Health insurers, and public health advocates, all want a ton of documentation from each physician visit. Electronic medical records are a way to capture data on physician visits, but someone has to enter the data. This may sound simple, but a doctor could, at least in theory, spend as much time writing information on a patient’s chart as they spent talking to the patient. Doctors too complained about having their faces buried in a computer screen detracted from the patient visit, searching for a pulldown screen, looking for a diagnosis code that corresponds with your patients’ condition. Documenting patient encounters is an inconvenience that can detract from the visit.
The most common way to capture data is for the doctor to take a moment after the visit to record a quick observation with is transcribed by someone else into the patient medical record. Another solution to the data capture problem during physician visits is to have a stenographer in the room with doctor and patient. However, most patients do not want to describe their hemorrhoids, erectile dysfunction or OB/GYN concerns to anyone other than their doctors.
Increasingly there is another party eavesdropping on your doctor’s visit. An artificial intelligence (AI) interface passively recording and processing the visit discussion and turning the conversation into a coherent medical record. The following is from the Wall Street Journal:
A fast-growing technology known as ambient listening is taking over an onerous but necessary task in healthcare: documenting what happens in the doctor-patient encounter.
Already gaining traction for outpatient medical visits, the AI-powered systems are also moving into hospital rooms and emergency departments to capture discussions at the bedside, update medical records, draft care plans and create discharge instructions. Healthcare systems nationwide, including Stanford Health Care, Mass General Brigham, University of Michigan Health and Ardent Health, are adopting the technologies widely referred to as AI scribes.
The AI interface is capable of more than merely turning patient Chatty Cathy into a list of health complaints with her physician’s comments. It holds the promise of even helping physicians with diagnosis or important medical observations.
“We are just scratching the surface of what this technology can do,” says Dr. Lance Owens, regional chief medical information officer at University of Michigan Health… “I see it being able to provide insights about the patient that the human mind just can’t do in a reasonable time.” By connecting older data with new information in the medical record, for instance, the technology could help make sure that an incidental finding years ago was followed up on.
It is easy to see how the mere assistance of entering information into a medical record would free up the doctor to see more patients during a workday. However, the next step is freeing the doctor from mentally having to connect the dots with visits long past. Having to stop and think more about what the patient said, or did not say is mentally taxing. I have long criticized the typical physician visit of today because few professionals can solve a complicated problem in a 15-minute interview unless that problem is simple.
Ambient-listening software runs on devices from desktops to mobile phones and tablets, using speech recognition and AI language models to capture and process conversations between a clinician and a patient during a visit. The AI, from companies including Microsoft, Ambience Healthcare, Abridge, Onpoint Healthcare Partners and Nabla, can recognize pertinent medical dialogue, distinguish voices and filter out casual chitchat, such as talk about the weather or sports.
By the time a doctor’s appointment is done, an AI scribe can generate a comprehensive note for the electronic medical record, create a concise after-visit summary for the patient and provide data for coding and billing purposes.
New features are on the way, including diagnostic aid tools that suggest conditions and more sophisticated multilingual capabilities.
The initial responses to ambient listening AI tools are promising. Doctors report reducing the time spent documenting patient encounters by two thirds. Nearly all (79%) report the technology allowed them to focus more attention on their patient, while 60% said they are more likely to push back retirement due to less stress and less burnout from balancing patient care with medical transcription needs. Ambient listening is likely the first step in a line of future decision-support tools that make specialties like primary care more accessible.
Read more at WSJ: AI-Powered ‘Ambient Listening’ Is Coming to Your Doctor Visit