Defining adenoma detection rate benchmarks in average-risk male veterans

Abstract

Background and Aims Veterans have higher prevalence of colorectal neoplasia than non-veterans; however, it is not known whether specific Veterans Affairs (VA) adenoma detection rate (ADR) benchmarks are required. We compared ADRs of a group of endoscopists for colonoscopies performed at a VA to their ADRs at a non-VA academic medical center.

Methods This was a retrospective review of screening colonoscopies performed by endoscopists who practice at the Indianapolis VA and Indiana University (IU). Patients were average-risk males aged 50 years or older. ADR, proximal adenoma detection rate, advanced adenoma detection rate, and adenomas per colonoscopy were compared between IU and the VA groups.

Results Six endoscopists performed screening colonoscopies at both locations during the study period (470 at IU vs 608 at the VA). The overall ADR was not significantly different between IU and the VA (58% vs 61%; p =0.21). Advanced neoplasia detection rate (13% vs 17%; p=0.46), proximal adenoma detection rate (46% vs 47%; p=0.31), and adenoma per colonoscopy (1.59 vs 1.84; p=0.24) were not significantly different. There were no significant differences in cecal intubation rate (100% vs 99%; p=0.13) or withdrawal time (10.9 vs 11.1 min; p=0.28).

In regression analysis, there was significant correlation between the attending-specific ADRs at IU and the VA (p=0.041, r-square=0.69).

Conclusions In this study of average-risk males undergoing screening colonoscopies by the same group of endoscopists, the ADRs of VA and non-VA colonoscopies were not significantly different. This suggests that a VA-specific ADR target is not required for endoscopists with high ADR.