- Today
- Holidays
- Birthdays
- Reminders
- Cities
- Atlanta
- Austin
- Baltimore
- Berwyn
- Beverly Hills
- Birmingham
- Boston
- Brooklyn
- Buffalo
- Charlotte
- Chicago
- Cincinnati
- Cleveland
- Columbus
- Dallas
- Denver
- Detroit
- Fort Worth
- Houston
- Indianapolis
- Knoxville
- Las Vegas
- Los Angeles
- Louisville
- Madison
- Memphis
- Miami
- Milwaukee
- Minneapolis
- Nashville
- New Orleans
- New York
- Omaha
- Orlando
- Philadelphia
- Phoenix
- Pittsburgh
- Portland
- Raleigh
- Richmond
- Rutherford
- Sacramento
- Salt Lake City
- San Antonio
- San Diego
- San Francisco
- San Jose
- Seattle
- Tampa
- Tucson
- Washington
New Model Predicts Recurrent C. difficile Risk in IBD Patients
Machine learning model achieves 80% accuracy in identifying high-risk patients
Published on Feb. 4, 2026
Got story updates? Submit your updates here. ›
Researchers have developed a new supervised machine learning model that can accurately predict which patients with inflammatory bowel disease (IBD) are at high risk of developing recurrent Clostridioides difficile infection (CDI). The model, called RecurCDI-IBD, was trained on data from over 2,400 IBD patients and achieved an 80.05% accuracy rate in identifying those at elevated risk of recurrent CDI.
Why it matters
Recurrent CDI is a major challenge for patients with IBD, but predicting which patients are at highest risk has been difficult due to the complex interplay between the two conditions. This new machine learning model provides clinicians with a valuable tool to proactively identify high-risk IBD patients and enable earlier, more tailored care to prevent recurrent CDI episodes.
The details
The researchers developed the RecurCDI-IBD model using routinely collected clinical data from electronic health records, including demographics, IBD subtype, comorbidities, medications, and lab parameters. They trained the model on 80% of the patient data and tested it on the remaining 20%, achieving an area under the receiver operating characteristic curve (AUC) of 0.88, sensitivity of 0.76, specificity of 0.84, and precision of 0.83. Factors associated with increased risk of recurrent CDI included recent hospitalization, male sex, initial metronidazole therapy, and comorbidities, while initial fidaxomicin or vancomycin treatment and prior 5-aminosalicylic acid exposure were linked to lower risk.
- The study included IBD patients who developed CDI between 2013 and 2021.
The players
Sahil Khanna, MBBS, MS
Lead author of the study and a researcher at Mayo Clinic in Rochester, Minnesota.
Mayo Clinic
The academic medical center where the study was conducted.
National Institute of Diabetes and Digestive and Kidney Diseases
The National Institutes of Health division that provided funding for the study.
What they’re saying
“In this study, we developed a machine learning model that accurately predicts the risk of CDI recurrence in IBD patients based on routinely collected clinical data. Our model offers a new tool to proactively identify high-risk patients and enable earlier, more tailored care.”
— Sahil Khanna, MBBS, MS, Lead author (The American Journal of Gastroenterology)
What’s next
The researchers noted that while the model's performance was consistent in internal cross-validation, its accuracy metrics should be interpreted cautiously given the retrospective nature of the study. Further validation in more diverse patient populations will be needed to assess the model's real-world applicability.
The takeaway
This new machine learning model provides clinicians with a valuable tool to better identify IBD patients at high risk of developing recurrent C. difficile infections, enabling more proactive and personalized care to prevent these challenging complications.
