Variety and experience: Learning and forgetting in the use of surgical devices

We use a unique, hand-collected data set to examine learning and forgetting in hip replacement surgery as a function of a surgeon’s experience with specific surgical device versions and the time between their repeat uses. We also develop a generalizable method to correct for the left censoring of device-version-specific experience variables that is a common problem in highly granular experience data, using maximum simulated likelihood estimation with simulation over unobservables conditional on observables. Even for experienced surgeons, the first use of certain device versions can result in at least a 32.4% increase in surgery duration, hurting quality and productivity. Furthermore, with the passage of time, surgeons can forget knowledge gained about the use of particular devices. For certain devices, when the time gap between repeat uses increases from its median to its 75th percentile, surgery duration increases by about 3.4%. The high productivity and quality costs associated with device variety suggest that the gain from a new device design needs to be large enough to compensate for the short-term disadvantages of starting up on a new learning curve and of increasing the chances of knowledge depreciation over time.