Data-Driven Approach Predicts Daily CDI Risk

They then developed a machine learning process to help predict a patient's risk of developing CDI by repeatedly analyzing the data. The machine learning process extracted features that could point to constellations of symptoms, circumstances, and details of medical history most likely to result in CDI at any point in the hospital stay. The algorithm identified a total of 2,964 and 4,739 features in the MGH and UM models, respectively.

The MGH and UM models identified different sets of features that could predict the relative importance of risk factors, which varied significantly across hospitals. In particular, in-hospital locations appeared in the set of top risk factors at one hospital, and in the set of protective factors at the other. On average, both models were able to predict CDI five days in advance of clinical diagnosis, using risk stratification models tailored to an institution’s EHR system and patient population. The study was presented at the annual IDWeek meeting, held during October 2017, in San Diego (CA, USA).

“The records contained over 4,000 distinct variables. We have data pertaining to everything from lab results to what bed they are in, to who is in the bed next to them and whether they are infected. We included all medications, labs and diagnoses. And we extracted this on a daily basis,” said senior author Jenna Wiens, PhD, of the University of Michigan. “You can imagine, as the patient moves around the hospital, risk evolves over time, and we wanted to capture that.”

CDI is a serious illness resulting from infection of the internal lining of the colon by C. difficile bacteria, and typically develops after the use of broad-spectrum antibiotics that disrupt normal bowel flora, allowing the bacteria to flourish. The risk of CDI is particularly high in patients aged 65 years and older, and disease recurrence occurs in up to 25% of patients within 30 days of initial treatment. It is the leading cause of nosocomial diarrhea in industrialized countries.

Related Links:
Massachusetts General Hospital
Massachusetts Institute of Technology
University of Michigan Health System