Original article—alimentary tract
Endoscopist Specialty Is Associated With Incident Colorectal Cancer After a Negative Colonoscopy

https://doi.org/10.1016/j.cgh.2009.10.022Get rights and content

Background & Aims

The incidence of colorectal cancer (CRC) is reduced for at least 10 years after a negative colonoscopy, compared with the general population. However, CRCs do occur in individuals after a negative colonoscopy. We investigated whether the colonoscopy volume and specialty of the endoscopists who perform the exam are associated with CRC after a negative complete colonoscopy.

Methods

A cohort of Ontario residents, 50–80 years old, who had a negative complete colonoscopy between January 1, 1992, and December 31, 1997, was identified by using linked administrative databases. Cohort members had no history of CRC or inflammatory bowel disease or a recent colonic resection. Each individual was followed through December 31, 2006, and those with a new diagnosis of CRC were identified. Multivariable analysis was used to evaluate the association of patient, endoscopist, and procedure setting characteristics with incident CRC.

Results

A cohort of 110,402 individuals with a negative complete colonoscopy was identified; the majority (86%) had their procedures performed in hospitals. During the 15-year follow-up period, 1596 (14.5%) developed CRC. There was no association between endoscopist colonoscopy volume and incident CRC. Among persons who had their colonoscopies at a hospital, those who had their procedures performed by a non-gastroenterologist were at significantly increased risk for developing subsequent CRC.

Conclusions

Endoscopist specialty is an important determinant of the effectiveness of colonoscopy in usual clinical practice. After a negative colonoscopy, those who have had their procedures performed by a gastroenterologist are less likely to develop CRC.

Section snippets

Data Sources

Data for this population-based cohort study were obtained from 4 data sources: the Canadian Institute for Health Information (CIHI), discharge abstract and same-day surgery database, the Ontario Health Insurance Plan (OHIP) database, the Registered Persons Database (RPDB), and the Ontario Cancer Registry (OCR).

The CIHI databases contain information for all acute care facilities about all discharges and procedures done during admission for residents of Ontario since April 1, 1988. All diagnostic

Identifying Incident Colorectal Cancers

We identified all CRCs (Table 2) through December 31, 2006, for the study cohort. Follow-up began on the date of the index colonoscopy. During follow-up we removed individuals from the study cohort if they developed CRC, moved out of the province, or died.

Explaining Incident Colorectal Cancers

We examined patient, endoscopist, and procedure setting characteristics associated with incident CRC. The patient characteristics examined were age, sex, and comorbidity. Comorbidity was assessed with the Deyo score calculated from ICD-9-CM diagnoses other than CRC recorded in CIHI for 5 years before the index colonoscopy.11

The endoscopist characteristics examined were endoscopist colonoscopy volume and endoscopist specialty. Endoscopist colonoscopy volume was defined by using the total number

Data Analysis

Data analysis was conducted by using SAS Version 9.13 (SAS Institute Inc, Cary, NC). Univariate tests of association were done with t tests for continuous variables and χ2 tests for categorical variables. We evaluated interactions between endoscopist colonoscopy volume, endoscopist specialty, and procedure setting. There was evidence of an interaction between endoscopist volume and procedure setting. To address this, we used separate Cox proportional hazards models for hospital and private

Results

From January 1, 1992, to December 31, 1997, we identified 110,402 individuals who had a negative complete colonoscopy. Table 3 shows the characteristics of the study cohort. The majority of individuals (86%) had their colonoscopy in a hospital. Compared with those who had their procedures done in a hospital, individuals who had their colonoscopies in private office/clinics were younger, more likely to be men, have less comorbidity, more likely to have their procedure done by an endoscopist in

Discussion

We report here the results of a population-based cohort study of 110,402 people age 50–80 years who had a negative complete colonoscopy during 1992 to 1997 and were followed for up to 15 years. For those individuals who underwent their colonoscopies in a hospital, which was the majority (86%), having the procedure performed by a non-gastroenterologist was independently associated with subsequent incident CRC. For those who underwent their colonoscopies in a private office/clinic, endoscopist

Cited by (141)

View all citing articles on Scopus

Conflicts of interest The authors disclose no conflicts.

View full text