Using an at-risk salary model to improve throughput in academic medical center operating rooms.

STUDY OBJECTIVE: The objective was to analyze if an at-risk salary model for faculty anesthesiologists could improve on-time first case starts (FCSs) and case turnovers (TOs) in an academic hospital inpatient operating room (IOR) and ambulatory surgery center (ASC). Organizational goals were 65% and 70% on-time FCS and case TO times for IOR and ASC, respectively.

DESIGN: This was a retrospective study.

SETTING: Surgical cases performed at both the IORs and the ASCs at a large academic medical center were included.

INTERVENTIONS: We converted 5% to 7% (academic rank dependent) of anesthesiologist salary from guaranteed to an at-risk salary model. Salary was earned back on a case-by-case basis by starting cases on time or by documenting a valid reason for case delay in the anesthesia record. On-time first case and goal TO times were determined using American Association of Clinical Directors standard definitions.

MEASUREMENTS: Data were reviewed for 1 year prior to implementation of the at-risk salary model and for 1 year after the implementation. Monthly average on-time FCS and TO times were compared between the preimplementation and postimplementation time frames. Data were analyzed using analysis of variance for repeated measures.

MAIN RESULTS: After the implementation of the at-risk salary model, the organization experienced a 33% and 86% improvement in on-time FCSs (P< .01) in the inpatient and ambulatory operating rooms, respectively. A 41% (IOR) and 44% (ASC) improvement in timely case TOs (P< .01) was also seen.

CONCLUSIONS: Anesthesiologists can drive efficiency in an operating room setting. By incentivizing on-time FCS and timely case TO with an at-risk salary model for faculty anesthesiologists, we were able to demonstrate a sustained significant improvement for these metrics. In both an inpatient and an ambulatory setting, operating room efficiency may be best served by aligning provider financial incentives with desired outcome metrics.