Background: Current procedural terminology (CPT) for colon and rectal surgery lacks procedural granularity and misclassification rates are unknown. However, they are used in performance measurement, for example, in surgical site infection (SSI). The objective of this study was to determine whether American College of Surgeons National Surgical Quality Improvement Program (NSQIP) abstraction methods accurately classify types of colorectal operations and, by extension, reported SSI rates.
Materials and methods: This was a retrospective study conducted at a single tertiary care center. The colectomy- and proctectomy-targeted NSQIP database from January 2011 to July 2016 was used to perform a semiautomated reclassification (SAR) of all colectomy and proctectomy cases performed by colorectal surgeons. The primary outcome was the difference in perioperative SSI rates by case classification method.
Results: Thousand sixty-three patients underwent a colectomy or proctectomy during the study period with a mean age of 55.7 (SD = 16.7) years. Use of the NSQIP classification scheme resulted in 849 colectomy and 214 proctectomy cases. Use of the SAR method resulted in 650 colectomy cases and 413 proctectomy cases (P < 0.001), a 23.4% reclassification of colectomy cases. The group of cases classified as colectomy by SAR had a lower rate of deep/organ space infections than those classified as colectomy by NSQIP (4.5% versus 7.1%, P = 0.034).
Conclusions: These findings highlight the challenges of CPT code-based patient classification and subsequent outcomes analysis. Expanding the CPT system to more accurately represent colorectal operations would allow for more representative reported outcomes, thus enabling benchmarking and quality improvement.
Keywords: CPT; NSQIP; Surgical site infection.
Copyright © 2018 Elsevier Inc. All rights reserved.