Time of day variation in polyp detection rate for colonoscopies performed on a 3-hour shift schedule

Abstract

Background: Recent research suggests that the colonoscopy polyp detection rate (PDR) varies by time of day, possibly because of endoscopist fatigue. Mayo Clinic Rochester (MCR) schedules colonoscopies on 3-hour shifts, which should minimize fatigue.

Objective: To examine PDR variation with the MCR shift schedule.

Design: Retrospective cohort.

Setting: Outpatient tertiary-care center.

Patients: This study involved completed outpatient colonoscopies in 2008. Procedures were excluded for lack of withdrawal time stamps, indications other than average-risk screening, inadequate bowel preparation, fellow participation, or performance by endoscopists with a low number of endoscopies performed.

Intervention: None.

Main outcome measurements: PDR (colonoscopies with ≥1 polyp divided by total number of colonoscopies) by shift of day.

Results: We analyzed 3846 colonoscopies. PDR varied significantly by shift (P = .008) on univariate analysis; results for shifts 1 and 3 were similar (39.0% vs 38.7%, respectively) whereas shift 2 had the highest PDR (44.7%). Mean withdrawal times were stable (P = .92). PDR also varied significantly (P < .0001) by month of year on univariate analysis. On multivariate analysis, patient age (P < .0001), patient gender (P < .0001), endoscopist mean withdrawal time (P < .0001), month of year (P = .0002), endoscopist experience (P = .04), and shift of day (P = .048) significantly predicted PDR.

Limitations: Retrospective study.

Conclusion: MCR's 3-hour shift schedule does not show a decrease in PDR as the day progresses, as seen in other recent studies. Intervention trials at other institutions could determine whether alterations in shift length lead to PDR improvements.