A Risk Score to Predict Clostridioides difficile Infection

Area of Science:

• Infectious Diseases
• Epidemiology
• Clinical Risk Prediction

Background:

• Clostridioides difficile infection (CDI) is a significant cause of severe diarrhea.
• Identifying risk factors is crucial for prevention and management strategies.

Purpose of the Study:

• To identify demographic and clinical risk factors for incident Clostridioides difficile infection (CDI).
• To develop and validate risk scores for predicting future CDI development.

Main Methods:

• Retrospective analysis of electronic health records for adult members with and without laboratory-confirmed CDI.
• Logistic regression models were used to identify risk factors and develop predictive risk scores.
• Two risk score models were evaluated using different Index Dates for Risk Score Assessment (IDRSA).

Main Results:

• A large cohort study (9986 CDI cases, 2,230,354 controls) identified older age, female sex, white race, and increased healthcare utilization as CDI risk factors.
• Risk factors included hospitalizations, emergency visits, skilled nursing facility stays, and use of antibiotics and proton pump inhibitors.
• A risk score model using hospital discharge as IDRSA demonstrated excellent predictive performance (C-statistic 0.848) for CDI within 31-365 days.

Conclusions:

• The developed risk scores can effectively identify individuals at high risk for CDI.
• These findings support targeted enrollment for Clostridioides difficile vaccine trials.
• The risk scores can improve study feasibility by aiding in sample size and completion time estimations.