Development and Validation of a Nomogram Predictive Model and Scoring Tool for Assessing the Risk of Inadequate Bowel Preparation in Colonoscopy Patients Over 40 Years Old: A Retrospective Observational Study.

Abstract

Aims: To explore the potential risk factors contributing to inadequate bowel preparation in middle-aged and elderly patients (aged 40 and above) undergoing colonoscopy, and to subsequently devise and validate a comprehensive risk assessment tool and nomogram model for accurately predicting such preparation.

Design: A retrospective observational study was conducted at three campuses from January 2023 to December 2023.

Methods: Twenty-three clinical indicators derived from colonoscopy records were leveraged to inform the predictive models. By using multivariate and stepwise logistic regression analyses, a risk-scoring model and a nomogram prediction model were devised. Calibration curves were used to evaluate the model's accuracy, while decision curve analysis (DCA) and receiver operating characteristic (ROC) curves were used to evaluate the model's clinical applicability and discriminatory power, encompassing the Hosmer-Lemeshow χ² test for goodness-of-fit. TRIPOD was used to guide this study.

Results: A total of 6860 outpatients who met the criteria were selected and divided into a training set (n = 4116) and a validation set (n = 2744) according to the bowel preparation quality. BMI, comorbidity, history of constipation, frailty degree, prescribing doctor's specialty, nonpatient's own prescribing, anxiety level and abdominal surgery history were independent risk factors for inadequate bowel preparation. The corresponding risk scores were 2, 0, 2, 4 3, 3, 2 and 1 respectively; with a total score of ≥ 8.5 classifying patients into a high-risk group. The area under the curve for the training and validation sets were 0.740 and 0.730, respectively, and an optimal critical value threshold of 34%. DCA findings demonstrated that the nomogram model was clinically helpful throughout a broad threshold probability range.

Conclusions: The risk prediction nomogram model and assessment tool constructed in this study can help clinicians identify individuals at high risk for inadequate bowel preparation at an early stage, which is a guideline for personalised prevention and treatment.

Trial and protocol registration: The name of the trial register was outpatient discharge management after general intravenous anaesthesia. The clinical trial registration was 2024-0116, and the link to the trial at the registration website was https://ctms.z2hospital.com:8443/.

Reporting method: Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD) checklist.

Patient or public contribution: Data for this study were provided by 7724 outpatients over 40 years old who underwent their first colonoscopy between January 2023 and December 2023 at three campuses of a medical centre.