Patient-centered risk stratification of disposition outcomes following radical cystectomy

•Comorbidity status is a strong predictor of morbidity following radical cystectomy.•Length of stay, discharge to other than home and readmissions are affected by age and comorbidity.•Quantifying rates of these events for patients may lead to better informed decision-making.

PurposePatient-centered care involves providing understandable information to facilitate individualized health decisions among patients. We sought to determine the effect of age and comorbidity status on clinically meaningful outcomes following radical cystectomy (RC), in an effort the help optimize patient selection and enhance discussions among those considering surgery.Materials and methodsIn a retrospective review, 6,460 patients were treated with RC for bladder cancer from the U.S. Premier Perspectives Database between 2007 and 2013. The influence of age and comorbidity count on the rates of inpatient mortality, prolonged length of stay (LOS), disposition to other than home and hospital readmission within the month of surgery or month after, were assessed. Comorbidity was calculated using the Elixhauser method. Prolonged LOS was defined as >10 days. Multivariable logistic regression models were used.ResultsFollowing RC, 16% of patients were discharged to somewhere other than home, 37% had a prolonged LOS and 2% died during the index admission. Among those discharged home after surgery, 27% of patients were readmitted. Prolonged LOS was associated with increasing comorbidities and age >70 years (P < 0.001). The adjusted likelihood of readmission increased with increasing burden of comorbid conditions (P < 0.001), however, not with age. The likelihood of being discharged to other than home increased with age and comorbidity count (P < 0.001). Mortality was associated with ≥3 comorbidities and age >70 years.ConclusionsIncreasing age and comorbidity are associated with poorer outcomes following RC, with comorbidity being the predominant factor. Our findings may improve surgical selection and better align patient expectations following surgery by providing estimated rates of postoperative events for patients considering RC.