AI and Clinical Practice—the Potential to Reduce Clinician Burden and Streamline Health Systems
Oct 4, 2023
auto_awesome
JAMA Editor in Chief and a professor of pediatrics discuss how AI can reduce clinician burden and streamline health systems. Topics include AI in the back office of healthcare, types of studies needed in AI and clinical practice, equitable considerations in applying new technologies, and transforming electronic health records with AI.
AI implementation in the back office functions can reduce clinician burden and streamline health systems, offering benefits such as clear messages, improved education, and reduced cognitive burden.
To ensure responsible AI adoption in healthcare, a code of conduct is needed, along with studies focusing on algorithmic fairness, explainable AI models, and equity impact assessment.
Deep dives
AI in the Back Office and Its Impact on Clinicians
AI in the back office refers to the integration of artificial intelligence in various healthcare functions beyond clinical decision support and patient messages. It encompasses areas such as scheduling, prior authorization, and improving patient understanding of healthcare messages. By focusing on the back office functions first, AI implementation can be tested and refined without causing additional burden on clinicians. The potential benefits include clear and concise messages, improved education, and reduced cognitive burden. However, careful considerations must be made to ensure equity, avoid biases, and address disparities in funding and access to technology.
Moving from Electronic Health Records to Smarter Approaches
The electronic health record (EHR) has been a double-edged sword, introducing both benefits and challenges to the healthcare system. To move towards smarter approaches, there is a need to develop a metric for cognitive support, ensuring that EHR tools not only minimize cognitive burden but also effectively serve patients and providers without causing staffing issues. Properly designed studies, including clinical trials, are crucial to assess the clinical effectiveness and user experience of AI technologies. Usability studies, qualitative assessments, and stakeholder involvement, particularly from marginalized communities, can contribute to a holistic understanding of AI's impact on cognitive burden and patient care.
Guardrails for Responsible AI Adoption in Healthcare
As AI continues to rapidly evolve, there is a need for a code of conduct that sets up guardrails and ensures responsible AI adoption in healthcare. This code of conduct should apply to various stakeholders in the healthcare ecosystem, including computer scientists, health systems, and patients. To address concerns about biases, disparities, and policy regulations, studies focusing on algorithmic fairness, developing explainable AI models, and assessing the equity impact of AI technologies are needed. Additionally, collaboration between researchers, policy-makers, and industry experts is crucial to iteratively learn, refine, and respond to the challenges and potential ethical dilemmas introduced by AI in medicine.
In this Q&A, JAMA Editor in Chief Kirsten Bibbins-Domingo, PhD, MD, MAS, and Kevin B. Johnson, MD, MS, the David L. Cohen and Penn Integrates Knowledge University Professor of Pediatrics, Informatics, Engineering, and Communication at the University of Pennsylvania, discuss how AI can reduce clinician burden and streamline health system functions. Related Content: