- Development and internal validation of a logistic model predicting 90-day clozapine-induced gastrointestinal hypomotility in psychiatric inpatients (n=1,426).
- Six predictors retained: age ≥60, BMI <24 kg/m2, clozapine dose ≥300 mg, treatment ≥5 years, trihexyphenidyl use, predominantly bedridden status.
- Model discrimination moderate (AUC 0.731; bootstrap-corrected 0.718), calibration acceptable, decision curve net benefit at 15% to 70% thresholds; external validation required.
Risk Manag Healthc Policy. 2026 Oct 29;19:656510. doi: 10.2147/RMHP.S656510. eCollection 2026.
ABSTRACT
OBJECTIVE: To develop and internally validate a prediction model for a first episode of clozapine-induced gastrointestinal hypomotility (CIGH) in clozapine-treated psychiatric inpatients.
METHODS: In this single-center retrospective cohort study, 1,426 adults with schizophrenia, schizoaffective disorder or bipolar disorder who had received clozapine for at least 4 weeks and were admitted to the Department of Psychiatry of Beijing United Family Rehabilitation Hospital between January 2019 and December 2025 were included. The index date was defined as the date of the current psychiatric admission, and all predictors were measured at or before that date. The outcome was a first CIGH event within 90 days, defined as paralytic ileus or intestinal obstruction, severe constipation requiring pharmacological intervention, or fecal impaction. Predictors were selected using LASSO regression, and the model was constructed with multivariable logistic regression. Performance was assessed using the area under the receiver operating characteristic curve (AUC), the calibration curve and decision curve analysis, and internal validation used 1,000 bootstrap resamples. Static and interactive (Shiny) nomograms were developed.
RESULTS: CIGH occurred in 356 of 1,426 patients (24.96%). Six predictors were retained: age ≥60 years (OR 1.469, 95% CI 1.031-2.094), BMI <24 kg/m2 (OR 1.699, 95% CI 1.099-2.628), daily clozapine dose ≥300 mg (OR 2.276, 95% CI 1.531-3.384), clozapine treatment duration ≥5 years (OR 2.309, 95% CI 1.431-3.725), concomitant use of trihexyphenidyl (OR 3.322, 95% CI 2.291-4.818) and predominantly bedridden status (OR 2.608, 95% CI 1.817-3.744). Discrimination was moderate (AUC 0.731, 95% CI 0.694-0.768; bootstrap-corrected AUC 0.718), calibration was acceptable, and decision curve analysis indicated net benefit across threshold probabilities of 15%-70%.
CONCLUSION: This internally validated model provides a non-invasive aid for stratifying 90-day CIGH risk in clozapine-treated psychiatric inpatients and for structuring gastrointestinal monitoring. External validation and prospective evaluation of clinical utility are required before routine implementation.
PMID:42974511 | PMC:PMC13726712 | DOI:10.2147/RMHP.S656510
Share Evidence Blueprint
Save to Google Notes
Search Google Scholar
Save as PDF
⭐ My Revision List
