Abstract
OBJECTIVE: To examine predictors of early readmissions after radical cystectomy (RC). Factors associated with preventable readmissions may be most evident in readmissions that occur within 3 days of discharge, commonly termed 'bounce-back' readmissions, and identifying such factors may inform efforts to reduce surgical readmissions. PATIENTS AND METHODS: We utilised the Healthcare Cost and Utilization Project's State Inpatient Databases to examine 1867 patients undergoing RC in 2009 and 2010, and identified all patients readmitted within 30 days of discharge. We assessed differences between patients experiencing bounce-back readmission compared to those readmitted 8-30 days after discharge using logistic regression models and also calculated abbreviated LACE scores to assess the utility of common readmissions risk stratification algorithms. RESULTS: The 30-day and bounce-back readmission rates were 28.4% and 5.6%, respectively. Although no patient or index hospitalisation characteristics were significantly associated with bounce-back readmissions in adjusted analyses, bounce-back patients did have higher rates of gastrointestinal (14.3% vs 6.7%, P = 0.02) and wound (9.5% vs 3.0%, P < 0.01) diagnoses, as well as increased index and readmission length of stay (5 vs 4 days, P = 0.01). Overall, the median abbreviated LACE score was 7, which fell into the moderate readmission risk category, and no difference was observed between readmitted and non-readmitted patients. CONCLUSION: One in five readmissions after RC occurs within 3 days of initial discharge, probably due to factors present at discharge. However, sociodemographic and clinical factors, as well as traditional readmission risk tools were not predictive of this bounce-back. Effective strategies to reduce bounce-back readmission must identify actionable clinical factors prior to discharge.