Hospital learning curves for robot-assisted surgeries: a population-based analysis

Area of Science:

• Surgical Innovation
• Health Services Research
• Medical Technology Assessment

Background:

• Robot-assisted surgery (RAS) adoption is increasing rapidly.
• Defining the learning curve is crucial for training, credentialing, and patient safety.
• Understanding hospital-level learning curves is essential for widespread RAS implementation.

Purpose of the Study:

• To characterize the hospital learning curve for common robot-assisted procedures.
• To evaluate the association between cumulative hospital volume and patient outcomes.
• To inform credentialing and training programs for robotic surgery.

Main Methods:

• Cohort study using administrative health data from Ontario, Canada (2010-2021).
• Included adult patients undergoing robot-assisted radical prostatectomy (RARP), total robotic hysterectomy (TRH), robot-assisted partial nephrectomy (RAPN), or robotic portal lobectomy (RPL-4).
• Multivariable logistic models assessed the relationship between hospital volume and major complications, adjusting for patient factors and hospital clustering.

Main Results:

• No association found between cumulative hospital volume and major complications.
• Visual analysis showed transient worsening followed by improvement in outcomes with experience.
• Operative times decreased and plateaued with increasing volume (e.g., ~300 RARPs, ~75 TRHs).
• Increased volume correlated with decreased odds of prolonged length of stay for RARP and RPL-4.

Conclusions:

• Hospitals can adopt RAS without significantly increasing patient risk early in the learning curve.
• Increasing hospital experience with RAS is associated with improved efficiency and outcomes.
• The findings support the safe integration of RAS into surgical practice with appropriate volume progression.