Predictors of Re-admissions in Hospitalized Patients With Inflammatory Bowel Disease Treated at a Tertiary Care Hospital

Abstract

Introduction: To determine the readmission rate and identify factors associated with readmission in patients with inflammatory bowel disease (IBD). Methods: IBD patients who were hospitalized from July 1, 2004 to December 23, 2013 were identified using an IRB approved data repository. Demographics, comorbidities, disease type, disease location and behavior, tobacco use, indication for admission, surgery and type of surgery during admission, complications during admission, extraintestinal manifestations, and medications were collected. Patients with a repeat admission within 30 days of a prior admission were compared to patients without a re-admission. Results: Four hundred ninety index admissions were identified with 74 (15%) re-admissions. Re-admissions occurred after a mean of 16 days from the index admission. Surgery was performed in 43% of admissions of which 19% were emergent. Sixty percent of the patients were women and 29% were African American. Sixty-eight percent had Crohn’s disease (CD) of which 81% had stricturing or penetrating disease, 28% had perianal involvement, and 23% had upper tract involvement. Fifty-one percent had ileocolonic, 27% had colonic, and 22% had ileal disease location. Thirty-one percent of patients had ulcerative colitis, of which 78% had pancolitis. Sixty-nine percent had been exposed to at least 1 anti-TNF, 66% had been exposed to an immune suppressant, and 71% had at least one prior surgery. Twenty-seven percent had psychiatric comorbid conditions. Abdominal pain was the most common reason for index admission (54%) followed by diarrhea (27%), flare requiring intravenous medications (13%), gastrointestinal bleeding (12%), and bowel obstruction (8%). Complications that occurred 5% of the time or more during the index admission included wound infection (5%) and UTI (5%). The following factors were associated with re-admissions: endoscopic activity (p=0.06), IC disease location (p=0.02), absence of perianal disease (p=0.10), depression or dual disorders (p=0.09), dehydration (p=0.03), female gender (p=0.05), elective surgery (p=0.07), surgery other than reversal of a stoma (p=0.05), subtotal colectomy (p=0.01), and wound infection complicating surgery (p=0.07). Conclusion: Re-admission occurs after 15% of index admissions in a tertiary referral IBD population. Female gender, Ileocolonic and non-perianal CD, endoscopic activity, dehydration, psychiatric co-morbid conditions, and surgery are associated with readmission. Further studies are needed to confirm these risk factors and to identify approaches to decrease readmission.