Stephen H. Liebowitz

Since the early 2000s, artificial intelligence (AI) has raised concerns regarding its use in healthcare to manage vast amounts of patient data, ensure proper handling, and maintain robust security measures. Nevertheless, contemporary healthcare organizations are exploring ways AI can safely enhance operational efficiency and support their patient populations. Successful, evidence-based utilization relies on a well-defined ambulatory strategy, and operational efficiency must be foundational to that strategy. Patient no-shows and appointment compliance, especially in the context of social determinants of health such as access, present inherent obstacles to patient and provider satisfaction, continuity of care, practice productivity, and the financial sustainability of an organization. To address these obstacles, Berkeley Research Group has been working with Phoebe Physician Group. Their shared objective is twofold: enhance patient encounter volume and the associated revenue. This article provides insights into the steps taken to integrate AI and machine learning to mitigate the problem of no-shows by automatically double-booking appointments for patients with a high probability of not showing up. A glimpse into the outcomes achieved and lessons learned throughout the process also is presented.