Objectives: We aimed to identify the frequency and costs of, and the disease predictors and inpatient process issues that may predispose to, 30-day readmission for an inflammatory bowel disease (IBD) patient.
Methods: IBD patients admitted to an inpatient gastroenterology service were followed for a time-to-readmission analysis assessing factors associated with readmission within 30 days.
Results: Index admissions were more costly among those readmitted than among those not readmitted. Patients admitted with evidence of increased inflammation, infection, or obstruction or for dehydration or pain control had a higher risk of readmission. Patients treated with opioid analgesia during index admission were no less likely to be readmitted, and there was a 2.2-fold increase in readmissions when patients were discharged with no opioid analgesia. Scheduling variability and outpatient follow-up compliance were associated with readmission.
Conclusions: Predicting readmission is complex. A predictive model developed to be used at discharge yielded an area under the curve of 0.757.