A Simple Prediction Model for Clostridioides difficile Infection: A Hospital-Based Administrative Database Study

Area of Science:

• Infectious Diseases
• Clinical Prediction Modeling
• Hospital Epidemiology

Background:

• Clostridioides difficile infection (CDI) is a significant cause of nosocomial diarrhea.
• Existing prediction scores for CDI often lack both accuracy and simplicity for clinical use.
• A need exists for a straightforward tool to predict CDI risk in hospitalized patients.

Purpose of the Study:

• To derive and validate a simple, accurate prediction score for primary hospital-onset Clostridioides difficile infection (CDI).
• To facilitate early medical intervention and treatment plan adjustments for CDI prevention.

Main Methods:

• Retrospective cohort study of adult inpatients in a Japanese secondary care hospital.
• Data derived from electronic medical records and administrative databases (January 2016 - September 2022).
• Logistic regression analysis used to develop and validate the CDI prediction score.

Main Results:

• The derived prediction score incorporated antibiotic use, acid suppressant use, Charlson comorbidity index, and Barthel index.
• The model demonstrated strong predictive accuracy with c-statistics of 0.89 (derivation) and 0.82 (validation).
• The prediction score was well-calibrated, indicating reliable risk assessment.

Conclusions:

• A simple prediction score for hospital-onset CDI has been successfully developed and validated.
• This tool supports early medical intervention and proactive management of CDI risk.
• The score can help reduce the incidence of primary hospital-onset CDI.