Methods: We used the ICD-9 code 00845 to identify the initial occurrence of all hospitalizations involving CDI for the month of January 2013; designated as CDI-associated index admissions. We queried the database for all CDI-associated readmissions that occurred after an index admission from January-December 2013. Index admissions were then grouped into two categories; those without a readmission and those with one or more subsequent readmissions. Study variables included diagnoses, costs, LOS and ICU stay.
Results: In January 2013, there were a total of 29,551 individual CDI-related hospitalizations. In index cohort, 9.4% patients died during hospitalization and 68% did not undergo a subsequent CDI-associated hospitalization for the remaining period of the study. The remaining 22.5% experienced at least one hospital readmission related to CDI. The readmission group of patients underwent a total of 10,158 CDI-associated readmissions from January-December 2013; 32.3% of these involved a principal diagnosis of CDI. The number of median (IQR) readmissions per patient was 2 (1) and the time to readmission was 29 (59) days. The cost associated with readmission related to CDI was $8,000 ($8,000) and the LOS was 5 days. The readmission patients were significantly more likely to have a principal diagnosis of CDI (38.6% vs 32.3%, p<0.001), be older (age ≥65 years; 65.3% vs 63.5%, p<0.001) and be insured by Medicare (74.2% vs 68.7%, p<0.001) compared to the non-readmission group. Independent risk factors for CDI-associated readmission were IBD (aOR 1.32; 95%CI 1.15-1.52) and coverage by Medicare (aOR 1.37, 95%CI 1.26-1.49).
Conclusion: Our results indicate a significant healthcare burden involving readmissions, manifested as a high rate of CDI-associated re-hospitalizations and substantial associated hospital stays and costs. Insurance coverage by Medicare and the presence of IBD were independently associated with a higher risk of CDI-associated readmission.
Clorox: Grant Investigator , Research grant and Research support
STERIS: Grant Investigator , Research grant and Research support
M. Olyaee, None
C. J. Donskey, None
See more of: Poster Abstract Session