The future of voice recognition in healthcare — key thoughts from 8 CMIOs

Eight chief medical information officers discuss voice recognition technology and how it could impact the healthcare industry, from clinical documentation to decision support tools.

Shelly Nash, MD, CMIO at Adventist Health System (Altamonte Springs, Fla.): Voice recognition technology, which has improved vastly over the last couple of years in the ability to understand medical terminology and regional accents, has been a great success for us. Any mobile tool that can be accessed from a phone, and not just a laptop or desktop, is a huge win as far as patient and physician satisfaction.

To read the full interview, click here. 

William Hanson III, MD, CMIO at UPenn Health System (Philadelphia): While Alexa, [Apple's] Siri and Google Assistant don’t yet have real, scalable applications in healthcare, I think voice recognition, when paired with artificial intelligence, may soon become a common ‘background’ technology during clinical encounters, acting as a scribe, clinical documentation assistant and potentially, a decision support tool.

To read the full interview, click here. 

Robert Budman, MD, CMIO at Piedmont Healthcare (Atlanta): I foresee great potential for Amazon or Google voice recognition technologies in healthcare, but most of that still seems to be in the formative stage and garnering a lot of headlines. However, companies like Nuance and M*Modal are currently using voice recognition technology with various algorithms and artificial intelligence embedded in EHRs now. For example, we currently use Nuance’s Dragon Medical One coupled with Dragon Medical Advisor to drive thoroughness and specificity of documentation in our providers' notes. On top of that, I previewed Nuance's in-room assistant that can help take notes, search the chart and place orders while listening to the physician-patient interaction. It holds great promise and will only grow in its abilities and usefulness. Cost, implementation, training and support beyond just the development of the tools all are likely to push this as a mainstream product many years into the future.

To read the full interview, click here. 

Ryan Walsh, MD, CMIO and Vice President at The University of Texas Health Science Center (Houston): I'm a hunt-and-peck typist. I've used some form of voice recognition software since I graduated residency, and I don't know how I'd live without it. Ask my helpdesk what happens when mine breaks.

With respect to Amazon and Google, it seems like they are very solidly into voice recognition, artificial intelligence and machine learning, but I'm not sure at what level they are infusing it into healthcare IT.

As far as having a virtual assistant in the EMR, I need to see more data, or hands-on work, regarding what the workflow looks like. For example, can my brain process dictate text and computer commands at the same time? Would a mouse click be faster? Will the technology facilitate me being away from my workstation entirely? There is definitely a place for it, and the technology is definitely exciting. It needs to be well thought out and done right or we'll wind up with the next wonder-gadget my colleagues yell at me about.

To read the full interview, click here. 

Sam Bagchi, MD, CMIO at Christus Health (Irving, Texas): With the rapid development of voice recognition and semantic understanding/AI, we are approaching a more joyful clinician experience in some ways. Ambient listening devices for clinical encounters will transform the modern practice of medicine, and we [Christus Health] look forward to piloting this in the next 12 to 18 months.

To read the full interview, click here. 

Craig Joseph, MD, interim CMIO at El Camino Hospital (Mountain View, Calif.): Physicians are more excited about ambient listening devices than anything I've seen in my career. The idea that physicians will be able to ask, 'What was Bob's last ejection fraction?' or, 'Have we ever tried Betty on an ACE inhibitor?' is astounding. Soon these devices will be listening for documentation points or potential orders during office visits, all of which can be queued up for the physician to act on when he or she is ready. This is the future that clinicians have been waiting for. This is the technology that the EHR has been lacking.

To read the full interview, click here. 

Patrick Woodard, MD, associate chief medical officer for IT at Renown Health (Reno, Nev.): This is such an exciting time. Voice recognition will definitely make its way into the EHR, as it should, but not until it is consistently more accurate. A good 10 percent of the time, my bathroom lights turn on when I want my bedroom lights to. That's fine in my house, but not fine when we need to know the difference between Inderal, a blood pressure medication, and Adderall. As the technology matures, voice recognition will ultimately be able to make great strides in provider satisfaction with the EMR and greatly improve convenience and clinician-patient relationships.

To read the full interview, click here. 

Paul Dolan, MD, CMIO at Benefis Health System (Great Falls, Mont.): When referring to a voice recognition assistant helping a provider navigate within an EHR, as opposed to voice recognition transcription, my first reaction was that I didn't think I saw a place for it, but the more I have thought about it I do see a potential for it going forward. I can foresee a provider asking for a risk calculation to be performed, an order set to be pulled up, or asking for what test — based on the input of the patient's current history, symptoms, exam and testing — is best to rule out a certain condition based on pretest probability. That would involve the intersection of the voice activated assistant, the medical record and an artificial intelligence platform. 'Alexa, what is the current differential diagnosis and based on the patient's genetic profile and testing thus far, what is the next test indicated?'

Relative to voice recognition technology in the generation of a visit note, we are on the cusp of dramatic changes. It is anticipated that a system will listen to the conversation between a patient and provider and after a provider has answered some key questions, the system will generate a note capturing the encounter. That will be a game changer if, in particular, the notes generated paint a clear picture of the patient history and exam.

To read the full interview, click here.

To learn more about clinical informatics and health IT, register for the Becker's Hospital Review 2nd Annual Health IT + Clinical Leadership Conference May 2-4, 2019 in Chicago. Click here to learn more and register.

To participate in future Becker's Q&As, contact Jackie Drees at jdrees@beckershealthcare.com.

More articles on health IT:
CIOs are losing decision-making power, Black Book finds
Columbia University, IBM launch startup accelerators for blockchain projects
Why clinical teams drive EHR success: Q&A with SSM Health CIO Dr. Philip Loftus

© Copyright ASC COMMUNICATIONS 2019. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.

 

Top 40 Articles from the Past 6 Months