AI-Driven Medication Reconciliation for Patient Safety
Innovative discharge solutions: Leveraging artificial intelligence for effective medication use and counseling
Patient safety at the point of hospital discharge through effective medication reconciliation and counseling can reduce the risk for medication errors and enhance health outcomes. Unfortunately, the post-hospital discharge median rate of medication errors is 53% and unintentional medication discrepancies is 50% for adult patients. Elderly people on complex medication regimens and taking chronic medications, such as for diabetes and cardiovascular diseases, are most affected by these errors. Medication non-adherence affects as many as 40% to 50% of patients prescribed chronic medications and is associated with at least 100,000 preventable deaths and more than $100 billion in preventable medical costs per year. Of the main factors that influence adherence is the patient's ability to understand medication instructions.
To prevent medication errors and discrepancies, healthcare providers perform medication reconciliation at all stages of care transition. Healthcare providers make clinical decisions based on comparing the patient’s previous, current, and “to be prescribed” medication lists. Afterwards, they communicate the final modified medication list to the appropriate healthcare providers and counsel the patients about it before their discharge.
The use of technology in healthcare industry reduces costs and errors, thus providing better health care at a lower cost. One such technology that is ushering a new era of digital transformation and causing disruptive innovations at an unprecedented rate is Artificial Intelligence (AI). AI may be defined as “a system’s ability to interpret external data correctly, to learn from such data, and to use those learnings to achieve specific goals and tasks through flexible adaptation” (Kaplan and Haenlein, 2019, p. 15). AI has been adopted across various industries, including healthcare to improve business efficiency and create business value. Research scholars have provided insights on pharmacist-driven interventions, such as use of technology in medication reconciliation and counseling, which have shown improvements in medication errors and adherence among patients with more complex regimens including elderly people. However, research lacks understanding the structure and organizational processes behind medication reconciliation and counseling at the hospital discharge point.
To address this gap, this research aims to map out the system around medication reconciliation and patient counseling at the time of hospital discharge using the Systems Engineering Initiative for Patient Safety (SEIPS) model.
The results of this work will help ideate future AI-driven tools using computer vision and LLM-driven interventions to streamline medication reconciliation and counseling to improve patient safety and quality of care. This multidisciplinary research team led by Dr. Asma Ali from the Health Systems Management and Policy Division at the School of Public Health; Dr. Ankur Arora from the department of Management Information Systems at the Fogelman College of Business and Economics and Drs. Haomiao Ni and Xiajung Jiang from the department of Computer Science brings together unique expertise to address the problem from different points and to foster creating innovative solutions.
