These digital tools may save some time and perhaps lighten a clinician’s cognitive load, but one may also have to contend with more errors in the patient’s record and the potential to adversely impact decision making skills.
By John Halamka, M.D., M.S., Dwight and Dian Diercks President, Mayo Clinic Platform and Paul Cerrato, MA, senior research analyst and communications specialist, Mayo Clinic Platform
A growing number of healthcare systems are investing in AI scribes, which can sit in the background, record a conversation between clinician and patient, and then generate a carefully worded summary of the encounter. Among the vendors in this domain are Abridge, Nuance, and Ambience Healthcare. Ambient AI scribes are popular among many physicians in the U.S. By one estimate, 70% of doctors at the University of California San Francisco use them, and at Kaiser Permanente, over 7,000 physicians used them in over 2.5 million patient visits in a 14-month period.
Statistics on the value of AI scribes vary widely. The American Medical Association cites a study that found “these generative AI scribes not only saved physicians an estimated 15,791 hours of documentation time—equal to 1,794 eight-hour workdays—but also improved patient-physician interactions and enhanced doctor satisfaction.” The AMA report also stated that “Compared to non-users, physicians who adopted AI scribes recorded significant reductions in pajama time, work outside 7 a.m. to 7 p.m., time in notes, and time in orders.” A recent JAMA study, however, was less enthusiastic. It included five U.S. institutions and over 8,500 clinicians during a two-year period and found: “AI scribe adoption was associated with 13.4 fewer minutes of EHR time, 16.0 … fewer minutes of documentation time, and 0.49 … additional weekly visits delivered. Electronic health record time outside work hours did not change significantly.”
On the other hand, a multisite study that measured administrative burden and professional burnout over a 30-day period among six U.S. health systems found “use of an ambient AI scribe platform was associated with a significant reduction in burnout, cognitive task load, and time spent documenting, as well as the perception that it could improve patient access to care and increase attention on patient concerns in an ambulatory environment.” An independent evaluationconducted by British investigators likewise found AI scribes eased cognitive load, sped up documentation, and improved work-life balance. But on a negative note, there were more errors in the EHR, the notes were too long, and they were poorly integrated into the patient’s medical record.
There’s also concern among clinicians that relying on ambient AI scribes may weaken their clinical reasoning skills. One physician, writing about her experience using these digital tools, states: “The danger is not only that A.I. might write something wrong. It is that, in writing something fluid and plausible, A.I. may make it easier for us to offload some of the reasoning medicine requires, to do less of the painstaking work of sorting through uncertainty ourselves. That loss, I fear, will eventually show up in the clinical care patients receive.”
That criticism, however, has to be weighed against evidence to suggest that ambient AI scribes are not only capable of reducing the time clinicians need to do documentation. If implemented correctly into their daily routine, it can also reduce the workflow disruption that often occurs when physicians have to document patient care. Easing clinicians’ cognitive burden in this way may offset any impact AI scribes may have on the clinical reasoning process.
One of the most fascinating aspects of their evolution over the past six quarters is that the technology has become so commonplace that EHR vendors are offering ambient listening as part of their products for no additional cost. The pace of technology change is so fast currently that product life cycles are short. We have never seen a class of products evolve from proprietary/expensive to commonplace/free in six quarters. Likely these products will evolve to provide additional functions such as revenue cycle and supply chain integration, making them even more useful in daily workflow. As the products change, the debate about the strengths and weaknesses will continue for some time.
Share this:
