A clinician's time has never been more valuable than now, during the COVID-19 pandemic. Freeing up time for physicians by reducing the administrative burden on them is necessary to drive certain goals, such as decreasing physician burnout and improving both the patient and provider experience. Physicians have more time when Conversational artificial intelligence (AI) and ambient intelligence can be embedded directly in clinician workflows.
During a May 21 virtual event hosted by Becker's Hospital Review and sponsored by 3M, industry leaders discussed how conversational AI can capture the complete patient narrative in electronic health records (EHRs), closing care gaps with proactive physician nudging on CDI and HCC, to virtual assistants helping with routine administrative tasks and delivering an ambient experience.
The speakers were:
- Michael Finke, vice president of M*Modal business for 3M Health Information Systems
- Zach Taft, chief product officer, virtual assistant solutions at 3M Health Information Systems
Here are five ways AI and technology can improve healthcare organizations:
1. Technology needs to be working unobtrusively in the background to understand a patient's medical history — including vitals, lab results, medications — to build a complete and contextual clinical picture of the patient, Mr. Finke said. This is why technology can fall short if it relies solely on language processing, but real-time conversational AI doesn't just recognize words; it understands the content of the narrative in context of the patient's overall medical record to get a complete clinical picture and drive meaningful action. Once there is this type of clinical understanding, different applications can be built with it, including ones that support clinical documentation improvement (CDI), HCC, transfer of care, incidental findings, and even meet changing needs during the COVID-19 pandemic. Using speech recognition, natural language understanding and clinical intelligence, conversational AI can deliver high-value insights and proactive nudges to clinicians to drive action at the point of care. This computer-assisted physician documentation (CAPD) functionality encourages consistency, objectivity and evidence-based medicine while proactively closing care gaps and improving communication.
2. Ambulatory practice physicians spend nearly half of their time completing electronic health records and desk work, compared to 33.1 percent of time spent on direct clinical interactions, according to a 2016 study in Annals of Internal Medicine. Virtual assistants help with documentation tasks, allowing clinicians to spend more time with patients and reducing the probability of burnout, Mr. Finke said. The virtual assistant proactively surfaces insights, responds to direct requests from clinicians, and ambiently discerns and documents information from patient conversations. For example, clinicians can ask 3M's virtual assistant to pull up the results of a patient's recent white blood count or prescribe a Z-Pak. The virtual assistant interacts with information from across the patient encounter, providing rich context to improve patient care as a whole, Mr. Taft said.
3. In January 2017, the CEO and CIO of a large academic site asked M*Modal to leverage virtual assistants to decrease clinician burnout. Starting in August 2017, 50 providers took part in a pilot program. The findings included:
- Note closure rate increased from 50 percent to more than 75 percent
- RVU improved by 14 percent
- Patient volume and satisfaction improved
- Clinicians decreased time spent working at home from three hours to under 30 minutes each day
Today, more than 200 clinicians are using the solution, with a 95 percent adoption rate, while over 150 clinicians are currently using telehealth to support COVID-19 initiatives.
4. Three key stakeholders must buy into this ambient experience: the care team, the patients and the hospital administration/IT department. AI benefits all three of these groups, Mr. Taft said. AI reduces documentation time for clinicians; accurate and timely data can improve patient access to care and patient satisfaction and the solution must also be robust and scalable for an entire healthcare organization.
5. AI should be unremarkable in the sense that one doesn't need to think about it, and it doesn't get in the way, Mr. Finke said. Virtual assistants are there not to just automate efforts, but to help clinicians do the right things more easily, which is why conversational AI and virtual assistants are the next steps in the evolution of transforming the healthcare experience.
To learn more about 3M, click here. To listen to the full event, click here.