Back to Program | 2010 Program and Abstracts Overview | 2010 Posters
Readmission Due to Infection After Colectomy for Cancer Is Strongly Associated with One-Year Mortality
David Y. Greenblatt*1, Maureen a. Smith2, Erin S. O’Connor1,2, Noelle K. Loconte3, Jinn-Ing Liou2, Heather B. Neuman1, Sharon M. Weber1
1Department of Surgery, University of Wisconsin, Madison, WI; 2Department of Population Health Sciences, University of Wisconsin, Madison, WI; 3Department of Medicine, University of Wisconsin, Madison, WI
Background & Objectives: Early hospital readmission is a common and costly problem in the Medicare population. Previous research has established a correlation between readmission after colectomy for cancer and one-year mortality. The objective of this study was to identify the most common causes of early readmission and their association with mortality. Methods: Medicare beneficiaries who underwent colectomy for stage I-III colon cancer from 1992-2002 were identified from the Surveillance, Epidemiology & End Results-Medicare database. The 30-day rehospitalization rate was determined, and readmission diagnoses were grouped using Agency for Healthcare Research & Quality Clinical Classifications Software. Multivariate logistic regression identified predictors of one-year mortality. Odds ratios (ORs) were adjusted for patient sociodemographics, comorbidity, hospital volume, operative factors, in-hospital complications, length of stay, discharge destination, and receipt of chemotherapy.Results: Of 42,348 discharged patients, 4,662 (11%) were readmitted within 30 days. The most common causes of rehospitalization were ileus/obstruction, pneumonia, bleeding, surgical site infection, and sepsis. The one-year mortality rate in readmitted patients was 26.2%, compared to 10.5% in nonreadmitted patients (p<0.001). Of the 1,027 patients readmitted for an infectious complication, 341 (33%) died within one year. After adjusting for potential confounders, all common causes of rehospitalization were significantly associated with one-year mortality. The strongest predictors of mortality were pneumonia and sepsis (adjusted ORs 4.03 and 3.97, respectively). Conclusions: There is a remarkable association between early readmission and one-year mortality in patients undergoing colectomy for cancer. In particular, readmission due to any infectious etiology is strongly correlated with mortality, with one third of such patients dying within a year. Interventions designed to prevent infectious complications may decrease readmissions and possibly one-year mortality, thereby reducing healthcare expenditures and improving patient outcomes.
Table. Frequency of the five most common readmission diagnoses in 42,348 Medicare patients who underwent colectomy for cancer, and adjusted odds ratios (AORs) for one-year mortality.
| Readmission Diagnosis | Frequency (% of Readmitted) | AOR (95% CI) for Mortality |
| Not Readmitted | 37,686 | Reference |
| Ileus/Obstruction & GI Complications | 1,320 (28.3%) | 1.95 (1.66-2.30) |
| Pneumonia | 329 (7.1%) | 4.03 (3.12-5.19) |
| Bleeding | 322 (6.9%) | 2.96 (2.27-3.85) |
| Surgical Site Infection | 316 (6.8%) | 1.76 (1.28-2.42) |
| Sepsis | 275 (5.9%) | 3.97 (2.98-5.29) |
Back to Program | 2010 Program and Abstracts Overview | 2010 Posters
