Risk factors for 30-day readmission after colorectal surgery: does transfer status matter?

Rates of readmission after colorectal surgery (CRS) range from 9% to 25% and cost the US $300 million annually. The aim of this study was to identify risk factors for 30-d readmission after CRS. Our hypothesis was that transfer from an outside hospital before CRS increases incidence of 30-d readmission.

Using the Healthcare Cost and Utilization Project Nationwide Readmissions Database, a retrospective analysis of surviving adult patients who underwent inpatient colon and/or rectal resection from 2010 to 2014 was performed. Using multivariable logistic regression, we assessed the direct effect of potential risk factors for readmission, including demographics, hospital characteristics, comorbidities, indication for CRS, and transfer status to the index hospital where the CRS was performed.

A total of 336,792 patients were included, and 13% (n = 43,546) were readmitted within 30 d. Overall, 1% of patients (n = 3652) were transferred from another hospital for surgery, and these patients were more likely to be older, have comorbidities, have public insurance, and have low income. After adjustment, age, insurance type, household income, comorbidities, and primary indication for CRS were all significant predictors of readmission. Transfer status did not meaningfully impact the incidence of readmission after accounting for patient and hospital characteristics (aOR 1.08, 95% CI 1.00, 1.18). Patients with primary indications of trauma (aOR 1.88, 95% CI 1.48, 2.38), inflammatory bowel disease (aOR 1.64, 95% CI 1.1.56, 1.71), and ischemia (aOR 1.77, 95% CI 1.1.59, 1.97) were most likely to be readmitted. Patients treated at a rural nonteaching hospital, compared with those at a urban teaching hospital, were significantly less likely to be readmitted (aOR 0.80, 95% CI 0.76, 0.83).

Preoperative factors, such as patient comorbidities and primary indications for CRS, are important risk factors for postoperative readmission. Although transfer status does not independently predict readmission, it serves as a proxy for a high-risk group of patients that could be targeted for future interventions.