Perceived Risks of Endoscopy At the Denver Veterans Affairs Medical Center

Abstract

Background: Colonoscopy is generally safe but carries with it important finite risks including bleeding, perforation, and adverse reactions to sedation. Open access endoscopy allows providers to order tests without consultation or intimate procedural knowledge. At the Denver Veterans Affairs Medical Center (VAMC) patients (pts) are mailed a brochure in advance of their procedure that outlines indications and risks. The effectiveness of this passive educational approach is unknown. We describe the results of a self-completed survey from the Denver VAMC. Methods: From January 1 to March 31, 2004, 375 consecutive pts underwent colonoscopy in the outpatient endoscopy suite at the Denver VAMC. After IRB approval, all pts were offered participation in a survey study designed to address pre-procedural knowledge pertaining to indications, risks, and compliance with the provided brochure. Subjects were asked to provide minimal demographic data (highest level of education and income), and were asked whether the following were risks of colonoscopy: bleeding, diarrhea, infection, pain, adverse reaction to medicines, perforation, or no significant risks. Data were analyzed in a binary fashion (yes/no) for each risk. Results: Two-hundred ninety five pts (79%) completed the survey. Two-hundred fifty eight reported receiving the informational brochure and 257 read the brochure. Seventy-eight percent perceived some risks and 22% no risks. Risk perception was not statistically different between subjects who did vs. did not read the brochure. Twenty-one percent who read the brochure perceived no risks vs. 32% who did not (p = 0.14). Educational level and income appeared to predict subjects' ability to recognize procedure-related risks overall, however this was statistically significant only for selected risks (adverse reaction to sedation and perforation; p = 0.006 and p = 0.003, respectively for educational level beyond high school and p = 0.28 and p = 0.04, respectively for annual income above $20,000). In multivariate analysis, only educational level remained statistically significant. Conclusion: In a survey study of 295 veterans undergoing colonoscopy, more than 1 in 5 of subjects had the perception that this invasive procedure entailed no significant risks. Passive educational approaches, such as mailed brochures, may not be maximally effective in educating patients prior to colonoscopy. Data from this subject population suggest that future refinements in procedure-related risk education should take socioeconomic status into account with respect to information processing.