What role is nephrology likely to play in the new age of digital medicine? What impact will predictive modeling and machine learning have on improving patient care? How much training is needed to prepare nephrologists for a more digitized landscape of medicine?
These are just some of the questions ASN Executive Vice President Tod Ibrahim and Girish N. Nadkarni, MD, MPH, pondered in a recent podcast about augmented intelligence and its future in health care and nephrology.
Dr. Nadkarni, who chairs ASN's new augmented intelligence and digital health task force, is the Irene and Dr. Arthur M. Fishberg Professor of Medicine at the Icahn School of Medicine at Mount Sinai. A principal and co-investigator for several grants funded by the National Institutes of Health focusing on informatics, data science, and precision medicine, Nadkarni was recently awarded the American Nephrologists of Indian Origin Rising Star award for 2021.
Recognizing that augmented intelligence can help physicians see the bigger picture and make informed clinical decisions, Nadkarni pointed out that physicians need to go beyond just using electronic health records. If we only rely on them, “we are just looking at pixels in picture,” he said. “We are not seeing the whole picture. We are looking at individual pixels and trying to guess what the whole picture is. But if you partner with Apple Health and partner with CVS and partner with community organizations, then people get more health care, not sick care. You're getting a wholistic picture of this individual patient throughout the health cycle.”
Given that nephrologists often deal with the predictability of numbers, Nadkarni believes the specialty is well-positioned to lead the way in a path forward for augmented intelligence and digital health. Yet investment in training will be paramount.
“I think physicians now are very under-prepared for the digitization of medicine,” he acknowledged. “The only exposure we have to augmented health and digitized intelligence is the electronic health record, which not many of us like because it takes up a significant portion of our time. We have no introduction to probabilistic thinking, we have no introduction to the basics of predictive models, we have no introduction to how digital health can be used as an ally, and, more importantly, how it can help to strengthen the physician-patient relationship.”
Training needs to be built into medical education at an early stage, according to Nadkarni. “Both didactic and experiential training programs starting from medical students to residents to nephrology fellows to faculty—in order to prepare them for what's coming in the next 5 to 10 to 15 years, when medicine will be predominantly more digitized with augmented intelligence and predictive models—this will be a key part of the success of health care.”
Although there is plenty of excitement about the potential of augmented intelligence and digital health, Nadkarni underscored the importance of testing and reliability. “It is important that all of these models be validated to ensure they are as accurate as you say they are, and they need to be tested and randomized in trials to make sure they are doing what you expect them to do and not cause harm, so that they can benefit the patients and health care providers.”
As chair of ASN's new augmented intelligence and digital health task force, Nadkarni hopes to develop best practices, guidelines, and recommendations for augmented intelligence and digital health in nephrology moving forward. “Nephrology is my first love,” he said. “This is exciting work. It’s a huge honor and huge responsibility and I look forward to getting input from the ASN community and the larger nephrology community on what we should be doing and not be doing, and where we should be focusing our attention.”