Enhancing Physician Satisfaction and Patient Safety Through an Artificial Intelligence–Driven Scheduling System in Anesthesiology

Note: Satisfaction with schedule flexibility and predictability, social support availability, work-life balance, and symptoms of burnout was ranked on a scale of 1 to 5, with 5 representing the highest level of satisfaction.

Note: The Lightning Bolt Scheduling system was implemented in May 2019.

↵^aPreimplementation is the 6-month period (November 2018 through April 2019) prior to implementation of the Lightning Bolt Scheduling system.

↵^bPostimplementation is the pre-COVID period of May 2019 through February 2020 and the post-COVID period of May 2020 through March 2021.

↵^cP<0.001.

↵^dFewer transitions of care was calculated by multiplying the total number of cases postimplementation (85,905) by the preimplementation transition of care rate (10.3%) to estimate handoffs without the Lightning Bolt Scheduling system in an equated case load (8,848) and then subtracting the number of postimplementation transitions of care (7,776): 85,905 × 0.103 = 8,848 – 7,776 = 1,072.

↵^eFewer harm events was calculated by dividing the fewer transitions of care number (1,072) by 15 in accordance with the Jones et al finding that for every 15 patients exposed to a transition of care event, 1 additional patient would be expected to experience a harm event.⁹

↵^fThe hospital-incurred cost of $4,693 per harm event reported in a 2023 study by Haidar et al¹³ was used to calculate the estimated savings resulting from the reduction in harm events: 71.5 × $4,693 = $335,550.