The principal aim of this study was to develop a simple computerized report generator linked to a database, to answer specific pre-determined questions regarding the use of colonoscopy. Our goal was to develop a clinical tool that could be used to collect data from clinical practice. In this study we used the database to test the hypothesis that patients undergoing colonoscopy for constipation or bloating would not have an increased risk for colon neoplasia. Methods: A platform-independent report generator and database was developed in Filemaker PRO to determine the indications and outcome of colonoscopy in a large population of patients. The software underwent preliminary testing and was customized to meet local requirements at individual sites. 1081 patients at 2 centers (Portland VA Hospital (n=861) and Sinai Samaritan Medical Center, Milwaukee (n=200) are included in the preliminary results of this ongoing study. Results: Report entry required 1-2 minutes and was easily learned even by those not familiar with computers. Of 1081 patients studied, 931 had a complete colonoscopy with good bowel preparation and are included in the analysis. Asymptomatic patients undergoing colonoscopy as part of other studies constitute the control group. Patients with constipation/bloating were also compared to patients with a positive fecal occult blood test Indication for colonoscopy Normal Divertic ulosis Polyps Cancer Asymptomatic n=199 24% 40% 42% 0.5% Occult bleeding 16% 40% 51% 11% (+FOBT) n=55 Constipation/Bloating 29% 40% 42% 0%*n=52 Total n=931 23% 37% 42% 3.5%*= p<0.0001 compared to FOBT+ group Conclusions: 1.Simple computerized report generators linked to a database are useful to answer specific questions in multicenter outcomes studies related to endoscopy. 2. Patients with constipation and bloating as the principal indication for colonoscopy do not have an increased risk for neoplasia compared to asymptomatic patients.
ASJC Scopus subject areas
- Radiology Nuclear Medicine and imaging