Clinicians May Be Unprepared for Widespread CDS Algorithm Integration
Researchers posit that clinicians need more training on how to use clinical decision support algorithms before they can be integrated into medical practice.
Source: Getty Images
By - In a perspective article recently published in the New England Journal of Medicine, researchers from the University of Maryland School of Medicine (UMSOM) and Beth Israel Deaconess Medical Center argued that clinicians may not be adequately prepared for the integration of clinical decision support (CDS) algorithms due to a lack of training.
The authors noted that while the United States government has recently worked to ensure that CDS tools are safe for use in clinical settings, the next step in this process is ensuring that clinicians are educated on how to effectively leverage these algorithms.
The success of CDS tools depends on how well clinicians can interpret and act upon the algorithm’s risk predictions, they indicate, but many healthcare providers lack the skills to do so.
“These new technologies have the potential to significantly impact patient care, but doctors need to first learn how machines think and work before they can incorporate algorithms into their medical practice,” said Daniel Morgan, MD, MS, co-author of the perspective who serves as professor of Epidemiology & Public Health at UMSOM, in a news release.
The authors explained that while CDS technologies are already integrated into some EHR systems, clinicians often find these tools difficult or cumbersome to use in their daily practice.
“Doctors don’t need to be math or computer experts, but they do need to have a baseline understanding of what an algorithm does in terms of probability and risk adjustment, but most have never been trained in those skills,” said co-author of the perspective Katherine Goodman, JD, PhD, assistant professor of Epidemiology & Public Health at UMSOM.
To bridge this gap, the researchers posited that clinical training and medical education must integrate explicit coverage of probabilistic reasoning specific to CDS tools into existing curricula.
In particular, the authors recommended that trainees improve their probabilistic skills by learning foundational aspects of probability and uncertainty, which may be enhanced by using visualization tools to make these concepts more intuitive.
The researchers also suggested that clinicians be taught how to effectively incorporate algorithmic outputs into their decision-making through critical evaluations of CDS predictions. This education would involve increasing trainees’ understanding of the context in which these tools operate, an algorithm’s limitations, and how to consider relevant patient factors that CDS tools can miss.
Finally, the authors indicated that medical students and clinicians should be able to participate in practice-based learning to help them better interpret CDS predictions. This training would involve applying algorithms to individual patients and investigating how various inputs impact a CDS tool’s predictions. Additionally, the researchers noted that trainees should also be taught to communicate with patients regarding CDS-guided decision-making.
This perspective article reflects UMSOM’s ongoing efforts to leverage advanced technologies in clinical care and educate providers on these tools. To support this work, the school recently launched its Institute for Health Computing (IHC).
“Probability and risk analysis is foundational to the practice of evidence-based medicine, so improving physicians’ probabilistic skills can provide advantages that extend beyond the use of CDS algorithms,” said UMSOM dean Mark T. Gladwin, MD, vice president for Medical Affairs, University of Maryland, Baltimore, and the John Z. and Akiko K. Bowers Distinguished Professor. “We’re entering a transformative era of medicine where new initiatives like our Institute for Health Computing will integrate vast troves of data into machine learning systems to personalize care for the individual patient.”
