Getting Back to Face Time: AI Tools that Help Reduce Physicians’ Computer Use

Just as computers have propelled major advances in modern medicine, their presence has also become a significant burden. That’s because the amount of time physicians now spend on the computer — reviewing clinical data and completing required documentation in electronic health records (EHRs) — has soared. It is estimated that physicians spend roughly half of their day on documentation and less than a third with their patients. Put another way, for each hour of face-to-face interaction with patients, physicians log almost two hours on administrative computer tasks. Moreover, many physicians spend time at home, typically one to two hours each night, to stay abreast of their computer work.

Click here to watch Disruptive Dozen: 12 AI Technologies That Will Reinvent Care

The upshot? Physicians often report feeling dissatisfied with their work lives. A 2016 survey of more than 6,000 U.S. doctors revealed that the vast majority used EHRs, and that this group was frustrated with how much time they spend on the computer. The survey also reported that these physicians were more prone to professional burnout.

Despite the gravity of the problem, straightforward solutions are elusive, particularly as missing, inaccurate, or difficult-to-interpret documentation represents a major source of revenue loss for hospitals and other health care organizations. Moreover, EHRs, even with their flaws, hold remarkable promise for improving clinical care.

Nevertheless, several research teams and organizations are now working on effective ways to reduce physicians’ screen time. Multiple efforts are focused on improving the infrastructure of EHRs: for example, more user-friendly interfaces, better alignment with clinicians’ actual workflows, and enhanced interoperability between systems. In addition, researchers are also turning to artificial intelligence (AI) and machine learning to help lighten physicians’ computer workload by automating tasks they now perform by hand. If Netflix can choose the next television series you’ll want to binge-watch, why can’t AI anticipate the information doctors need to submit digitally?

For example, a team based in Santa Monica, California is now working on creating AI-based tools that will help physicians reclaim valuable time with their patients and improve professional satisfaction. Their approach involves scanning patient records for relevant medical information, proposing a likely diagnosis, and creating the necessary documentation for downstream services, including billing. The team likens their AI system to a digital “co-pilot” that mimics how physicians make decisions and takes on the most mundane, administrative tasks. The ultimate goal of these and other AI-based efforts focused on clinical documentation is to relieve doctors of repetitive and time consuming computer work — which is ideally suited for machines anyway — and allow them more time for doing what they do best: caring for patients.

For more information about Dr. Landman’s research, please contact Partners HealthCare Innovation by clicking here.