Predictive Feasibility Of Colonoscopy Without Sedation Using A Nomogram Model

Colorectal cancer poses a global burden, and colonoscopy is an effective screening tool, with unsedated colonoscopy offering distinct advantages but facing completion challenges. A retrospective analysis of 745 patients who underwent unsedated colonoscopy between July 2021 and August 2022 was conducted, with participants divided into completion (n = 670) and incompletion (n = 75) groups. Univariate analysis revealed significant differences in gender, age, constipation history, anxiety grade, postoperative Visual Analogue Scale scores and examiner experience between groups. Multivariate logistic regression identified constipation (odds ratio [OR] = 4.981, 95% confidence interval [CI]: 2.755–5.633), high anxiety (OR = 8.499, 95% CI: 4.987–10.212) and examination by a less-experienced physician (OR = 3.319, 95% CI: 2.673–4.732) as independent risk factors. The constructed nomogram model demonstrated good predictive accuracy (Hosmer–Lemeshow test: χ2 = 4.561, p = 0.683) and discriminative ability (area under the receiver operating characteristic curve = 0.886, 95% CI: 0.749–0.970). This model enables clinicians to identify high-risk patients preoperatively, optimise examination plan selection (e.g. recommending sedation) or implement intensive interventions, thereby improving the completion rate of unsedated colonoscopy and enhancing patient experience.