Many Patients Who Undergo Surgery for Colorectal Cancer Receive Surveillance Colonoscopies Earlier Than Recommended by Guidelines

Abstract

Background & Aims Patients treated with surgery for colorectal cancer (CRC) should undergo colonoscopy examinations 1, 4, and 9 years later, to check for cancer recurrence. We investigated the utilization patterns of surveillance colonoscopies among Medicare patients. Methods We used the Surveillance, Epidemiology and End Results (SEER)–Medicare linked database to identify patients who underwent curative surgery for colorectal cancer from 1992 to 2005 and analyzed the timing of the first 3 colonoscopies after surgery. ‘Early surveillance colonoscopy’ was defined as a colonoscopy, for no reason other than surveillance, within 3 months to 2 years after a colonoscopy examination with normal results. Results Approximately 32.1% and 27.3% of patients with normal results from their first and second colonoscopies, respectively, underwent subsequent surveillance colonoscopies within 2 years (earlier than recommended). Of patients who were more than 80 years old at their first colonoscopy, 23.6% underwent a repeat procedure within 2 years for no clear indication. In multivariable analysis, early surveillance colonoscopy was not associated with sex, race, or patients’ level of education. There was significant regional variation in early surveillance colonoscopies among the SEER regions. There was a significant trend toward reduced occurrence of 2nd early surveillance colonoscopies. Conclusion Many Medicare enrollees who have undergone curative resection for colorectal cancer undergo surveillance colonoscopy more frequently than recommended by the guidelines. Reducing overuse could free limited resources for appropriate colonoscopy examinations of inadequately screened populations.

Receipt of recommended surveillance among colorectal cancer survivors: a systematic review

“…Several studies examined early receipt of colonoscopy after colonoscopy with normal results and found that 17–62% of CRC survivors received a second colonoscopy within 18 months of initial colonoscopy [27, 38, 40, 42] and 27% of survivors received a third colonoscopy within 2 years of a second normal colonoscopy [42]. These results were comparable to those obtained in a French sample where 29% of CRC survivors received colonoscopy at a period inconsistent with recommended guidelines [17].…”

“…Twelve [2, 16, 24–27, 29, 34, 39–40, 42, 44] of the 34 studies did not provide data regarding the number of potentially eligible participants. Twenty-three studies [15–23, 30–33, 35–39, 41, 43, 45–47] explicitly reported how sample size was determined; for 11 [2, 24–29, 34, 40, 42, 44] data were inferred (e.g., authors reported that sample was determined from a given administrative database) but not explicitly reported.…”

“…Twelve [2, 16, 24–27, 29, 34, 39–40, 42, 44] of the 34 studies did not provide data regarding the number of potentially eligible participants. Twenty-three studies [15–23, 30–33, 35–39, 41, 43, 45–47] explicitly reported how sample size was determined; for 11 [2, 24–29, 34, 40, 42, 44] data were inferred (e.g., authors reported that sample was determined from a given administrative database) but not explicitly reported. Only 4 studies [15, 31, 37–38] explicitly addressed the issue of confounders, while 28 [2, 16– 18, 20–30, 32–36, 39–43, 45–47] made only implicit reference to potential confounders (e.g., adjusted analyses without specific rationale or justification for doing so), and 2 studies [19, 44] did not address confounders.…”

“…Five studies [23–25, 28, 36] provided only adjusted estimates. Related to generalizability, only 14 [16, 18, 20–22, 27, 34–36, 40, 42, 45–47] of the 34 studies explicitly discussed external validity, while 12 [17, 19, 24–26, 28, 32–33, 38, 41, 43–44] made only implicit reference to the issue.…”

“… ABBREVIATIONS : CRC, colorectal cancer; OV/PV/CV, office visits/physician visits/clinic visits; CE/PE, clinical exam/physical exam; COL, colonoscopy; CEA, carcinoembryonic antigen test; MDT, metastatic disease testing (e.g., x-ray, ultrasound, CT scan, MRI, enzymes); surv, surveillance; PCP, primary care provider; dx, diagnosis. NOTE : +, ≤ three studies report positive associations; ++, > three studies report positive associations; −, ≤ three studies report negative associations; −−, > three studies report negative associations; * ≤ three studies report no effect; and ** > three studies report no effect. a Reflects receipt of any one surveillance test/procedure among studies that reported multiple tests/procedures. b Reflects receipt of overall recommended surveillance as specified by each individual study. c Reflects receipt of early surveillance, defined as colonoscopy within 3 months to 2 years after colonoscopy with normal results [42]. …”