Health News

Evaluating AI digital assistants for primary care physicians

Photo: AAFP

Studies show the majority of U.S. physicians and clinicians feel burned out when going to work, and indicate clerical burdens – including clinical documentation – are a major contributor to burnout.

For primary care physicians, a new class of technologies – AI-powered digital assistants – is improving capacity and capability, while reducing their administrative and cognitive burden. But how can organizations determine which tools most effectively address primary care physicians’ pain points while strengthening patient engagement?

Dr. Steven E. Waldren, vice president and chief medical informatics officer at the American Academy of Family Physicians, will answer this and other questions in a HIMSS22 educational session entitled “Innovation to Improve Physician Burnout: Lessons from AAFP.”

Healthcare IT News sat down with Dr. Waldren to get a sneak peek at his session.

Q. Why is there a need for technologies that improve the capacity of primary care physicians?

A. Physician burnout based on clerical burden is at epidemic levels for primary physicians. Clerical burden requires greater than 50% of the physician’s time. At the same time, they must transform their practices to population-based care and alternative payment models. The associated financial risk threatens to burn down their margins and thus their practices.

EHRs have taught us all that technology can dramatically affect best practices. Primary care should consider technology essential to the optimal practice of medicine and the delivery of care. Over the past two decades, the clerical burden on primary care has grown with increased documentation, reimbursement and reporting requirements.

In a seminal 2017 article in the Annals of Family Medicine, tethered to the EHR, primary care physicians spent more than 50% of their workdays on their EHRs, which averaged 4.5 hours per day in clinic and 1.5 hours after hours per day at home. Nearly a quarter of that time was on EHR documentation tasks.

With this burden, physicians must modify their workflow, which changes their focus and work during the visit, from care delivery to clerical work. Their personal lives also changed from focusing on family and personal pursuits to tackling more clerical work and care tasks squeezed out of the workday.

It is these time and financial pressures on primary care why essential innovations are needed to increase the capacity of primary care. In addition, we have a growing workforce shortage that will continue to increase the demand for primary care. This is compounded with the growing senior population and increases in the incidence of chronic disease, which both increase the need for primary care.

Q. Describe AAFP’s process to assess the value of AI-based primary care tools.

A. We have established the AAFP Innovation Labs to discover, test and support essential innovations within our membership. We have identified existential threats to family medicine and we work to partner with companies that are leveraging proven innovations, like AI and machine learning, to address those threats.

Once a candidate company and product are identified, we establish a set of key performance indicators around effectiveness and “adoptability.” We then establish a proof point in family medicine by interviewing family physicians using the products.

Interviews and data collected from the company/product are used to assess the key performance indicators. We then publish a report of those findings. If we discover that the product was effective and appears adoptable, we move to a lab to test the product across segments in the member.

For this lab, we work to recruit at least 100 family physicians and their practice colleagues to adopt the product in their practice. We again assess effectiveness using the same key performance indicators. We also determine how many decide to adopt the solution after the lab. We then publish a report of our findings.

Should that solution demonstrably optimize family medicine and be adoptable by a cohort of family physicians, then we see that product category as an essential innovation in family medicine for those types of physicians/practices. We then work to disseminate best practices of how to use the class of product in family medicine.

Working with Suki, we have identified an AI Assistant for Documentation as an essential innovation for family physicians with documentation burden where the product is integratable into their EHR and has a mobile offering.

Q. What have been some of the results you’ve achieved during physician pilot programs?

A. We have completed our lab with Suki, which is an AI assistant that uses voice and commands to assist physicians in completing their documentation. In our Suki lab, we had a 60% adoption rate. These adopters saw a 72% reduction in their median documentation time per note.

This resulted in a calculated time savings of 3.3 hours per week per clinician. In addition, participants reported improved satisfaction with their workload and overall with their practice. We had physicians talk about the solution as a “breakthrough.” (Click here for the full report.)

We are completing our discovery phase with Navina, which has an AI assistant that can summarize a patient’s chart into a problem-oriented summary. This helps physicians in chart review and visit preparation.

We have seen significant reductions in chart preparation time and more accurate risk coding. We also have seen the identification of missed diagnoses buried in consultant and diagnostic reports. We have had physicians call the solution a “game changer.”

Dr. Waldren’s session, “Innovation to Improve Physician Burnout: Lessons from AAFP,” will be presented at HIMSS22 in Orlando on March 15 from Noon to 1 p.m. in the Orange County Convention Center in room W230A. His co-presenter is Dr. Kamel Sadek, director of informatics at Village Medical.

Twitter: @SiwickiHealthIT
Email the writer: [email protected]
Healthcare IT News is a HIMSS Media publication.

Source: Read Full Article