Validation of the risk questionnaire for colorectal cancer detection

Abstract

AIM: to develop and validate of the colorectal cancer screening questionnaire.PATIENTS AND METHODS: six hundred ninety seven patients without any significant complaints were prospectively included in study. All patients underwent colonoscopy and specific questionnaire. Regression model was performed to stratificate significance level of definite questions.RESULTS: colorectal cancer (CRC) was diagnosed in 39 (5.6%) patients, benign colon neoplasms — in 283 (40.6%) patients. In the ROC analysis, a cut-off of 23 points divides the group of patients into high and low risk of colorectal neoplasms with a sensitivity of 92.8% and a specificity of 39.2%. The area under the curve was 0.724, which corresponds to the good predictive value.CONCLUSION: the developed questionnaire show good predictive power for high CRC risk patients identification, and can be included in the screening program model.