To define the relationship between urology RVUs and measures of surgical complexity and physician workload. Secondary objectives include: 1) identifying procedures with outlying RVU values for their measures of surgical complexity and workload; and 2) calculating projected RVU values for these procedures.
We obtained surgical case data for 71 urology current procedural terminology (CPT) codes from the 2017 American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database. Pearson correlation coefficients were calculated to measure the association between mean total work RVU (mRVU) and operative time, length of hospital stay (LOS), serious adverse events (SAEs), readmissions, and mortality. We developed a multivariable regression model to predict mRVU from these measures. Studentized residuals were used to identify outlying CPT codes for both bivariable and multivariable regression models, and empirically derived RVU values from complexity and work effort metrics were estimated.
We analyzed 71 urology CPT codes encompassing 55,068 cases. RVUs correlated well with median LOS (R = 0.81), median operative time (R = 0.92), SAEs (R = 0.83), and readmissions (R = 0.74). RVUs were poorly correlated with mortality (R = 0.34). Outlying procedures identified using the multivariable model were retroperitoneal lymph node dissection (projected +21.09 RVUs), laparoscopic ureteroneocystotomy (projected -12.34 RVUs), and cystectomy with bilateral pelvic lymphadenectomy (projected +9.37 RVUs).
Urology work RVUs correlate more with operative time than other measures of surgical complexity and physician workload. There exist several significant outlying procedures for various work measures. Incorporating objective work data may improve RVU assignments in the future.

Copyright © 2020. Published by Elsevier Inc.

Author