Skip to main content
Artificial intelligence undoubtedly has the potential to transform medicine by improving healthcare practice and outcomes with a variety of tools(1). Predictive analytics tools can use electronic health record (EHR data) to determine patients’ risk for hospitalization or deterioration which can help hospitals effective triage and allocate limited resources. Furthermore, AI can diagnose diseases by detecting patient health variables in wearable devices or analyzing radiographs and pathology images. Ideally, these AI tools can accelerate clinical decision making and improve clinical workflow because they can be more accurate and reliable than humans, but the use and development of AI tools pose numerous ethical concerns, which explains their limited use in clinical practice (1, 2).
As explored in this paper, AI can be used for automation of pre-visit planning (PVP), which can benefit patients and clinicians. PVP on its own is important for helping physicians prioritize patient needs to provide better care while saving time. AI tools can leverage machine learning (ML) techniques to support PVP, such as identifying care gaps, sending appointment reminders, and processing pre-appointment questionnaire. However, these AIML tools for PVP, along with many other AIML tools used in medicine, are not validated in real-world settings and/or published in peer-reviewed journals, making it difficult to assess clinical utility across various healthcare settings (1, 3). Additionally, the use of machine learning that involves collecting patient data poses concerns about patient privacy and “black-box medicine,” in which algorithms generate results based on past data without users understanding how the data are used or how results are produced (4).
Despite the ethical and translational challenges involved in developing and using AI tools, I believe that with proper oversight, AI tools can become commonplace in medicine. Before implementation of AI tools, developers should engage a variety of stakeholders across various healthcare settings ,such as patients, care providers, family members, administrators, and the general public to identify their goals, priorities, and concerns for the use of AI tools in clinical practice (5). To further improve PVP, AI tools could be created to encompass social determinants of health that may affect patients’ health so that physicians can provide more appropriate care for their patients with unique needs. Incorporating the endless amount of patient information available into AI applications in PVP and beyond will improve disease surveillance, treatments, diagnosis, and patient outcomes to improve quality of healthcare.
1. Bjerring JC, Busch J. Artificial Intelligence and Patient-Centered Decision-Making. Philosophy & Technology. 2020.
2. Lin SY, Mahoney MR, Sinsky CA. Ten Ways Artificial Intelligence Will Transform Primary Care. Journal of General Internal Medicine. 2019;34(8):1626-30.
3. Holdsworth LM, Park C, Asch SM, Lin S. Technology-Enabled and Artificial Intelligence Support for Pre-Visit Planning in Ambulatory Care: Findings From an Environmental Scan. The Annals of Family Medicine. 2021;19(5):419-26.
4. Price WN, II. Artificial Intelligence in Health Care: Applications and Legal Implications. The SciTech Lawyer 2017;14(1).
5. Cohen IG, Amarasingham R, Shah A, Xie B, Lo B. The legal and ethical concerns that arise from using complex predictive analytics in health care. Health Aff (Millwood). 2014;33(7):1139-47.