Related Experiment Video
Updated: Aug 9, 2025
Cefoperazone-treated Mouse Model of Clinically-relevant Clostridium difficile Strain R20291
Published on: December 10, 2016
A Predictive Model to Identify Complicated Clostridiodes difficile Infection
Jeffrey A Berinstein1, Calen A Steiner2,3, Samara Rifkin1
1Division of Gastroenterology and Hepatology, University of Michigan, Ann Arbor, Michigan, USA.
Insights
This study developed accurate machine learning models to predict severe Clostridioides difficile infection complications. The models showed good performance but performed worse in non-White patients, highlighting a need for further research to reduce disparities.
Area of Science:
- Infectious Diseases
- Medical Informatics
- Health Services Research
Background:
- Clostridioides difficile infection (CDI) is a significant cause of healthcare-associated infections, leading to severe outcomes like organ dysfunction, colectomy, and death.
- Existing risk scores for predicting severe CDI complications lack external validation and show poor performance.
- A hypothesis was formed that models built and validated on distinct patient cohorts would improve prediction accuracy for CDI complications.
Purpose of the Study:
- To develop and validate machine learning models for predicting severe complications of Clostridioides difficile infection.
- To compare the performance of lasso regression, random forest, and stacked ensemble algorithms in predicting CDI-related complications.
- To identify key variables associated with severe CDI outcomes.
Main Methods:
- A multicenter retrospective cohort study included 3646 adult patients diagnosed with CDI.
- Data were randomly split into training and validation sets, with 10-fold cross-validation used for model development.
- Three machine learning algorithms were employed to predict intensive care unit admission, colectomy, or death within 30 days of CDI diagnosis.
Main Results:
- All three developed models demonstrated strong predictive performance, with an area under the receiver operating curve (AUC) ranging from 0.88 to 0.89.
- Key predictors of severe CDI complications included albumin, bicarbonate, creatinine change, non-CDI ICU admission, and concurrent non-CDI antibiotics.
- Model performance remained robust across sensitivity analyses, but accuracy was notably lower for non-White patients compared to White patients.
Conclusions:
- A validated prediction model using a large, diverse patient population accurately estimates the risk of severe complications from Clostridioides difficile infection.
- Future research should focus on addressing the observed disparities in model accuracy between different racial groups.
- Further efforts are needed to enhance the overall performance of CDI complication prediction models.
Background:
Clostridioides difficile infection (CDI) is a leading cause of health care-associated infection and may result in organ dysfunction, colectomy, and death. Published risk scores to predict severe complications from CDI demonstrate poor performance upon external validation. We hypothesized that building and validating a model using geographically and temporally distinct cohorts would more accurately predict risk for complications from CDI.
Methods:
We conducted a multicenter retrospective cohort study of adults diagnosed with CDI. After randomly partitioning the data into training and validation sets, we developed and compared 3 machine learning algorithms (lasso regression, random forest, stacked ensemble) with 10-fold cross-validation to predict disease-related complications (intensive care unit admission, colectomy, or death attributable to CDI) within 30 days of diagnosis. Model performance was assessed using the area under the receiver operating curve (AUC).
Results:
A total of 3646 patients with CDI were included, of whom 217 (6%) had complications. All 3 models performed well (AUC, 0.88-0.89). Variables of importance were similar across models, including albumin, bicarbonate, change in creatinine, non-CDI-related intensive care unit admission, and concomitant non-CDI antibiotics. Sensitivity analyses indicated that model performance was robust even when varying derivation cohort inclusion and CDI testing approach. However, race was an important modifier, with models showing worse performance in non-White patients.
Conclusions:
Using a large heterogeneous population of patients, we developed and validated a prediction model that estimates risk for complications from CDI with good accuracy. Future studies should aim to reduce the disparity in model accuracy between White and non-White patients and to improve performance overall.
More Related Videos
12:58A Protocol to Characterize the Morphological Changes of Clostridium difficile in Response to Antibiotic Treatment
Published on: May 25, 2017
09:12A Protein Microarray Assay for Serological Determination of Antigen-specific Antibody Responses Following Clostridium difficile Infection
Published on: June 15, 2018