Title

A risk model for prediction of 30-day readmission rates after surgical treatment for colon cancer

Date of this Version

8-1-2020

Document Type

Article

Abstract

Purpose: The purpose of this study was to develop a risk model for the prediction of 30-day unplanned readmission rate after surgery for colon cancer. Method: This study was a cross-sectional analysis of data from Nationwide Readmissions Database, collected during 2010–2014. Patients ≥ 18 years of age who underwent surgery for colon cancer were included in the study. The primary outcome of the study was 30-day unplanned readmission rate. Results: There were 141,231 index hospitalizations for surgical treatment of colon cancers and 16,551 had unplanned readmissions. Age, sex, primary payer, Elixhauser comorbidity index, node positive or metastatic disease, length of stay, hospital bedsize, teaching status, hospital ownership, presence of stoma, surgery types, surgery procedures, infectious complications, surgical complications, mechanical wounds, pulmonary complications, and gastrointestinal complications were selected for the risk analysis during backward regression model. Based on the estimated coefficients of selected variables, risk scores were developed and stratified as low risk (≤ 1.08), moderate risk (> 1.08 to ≤ 1.5), and high risk (> 1.5) for unplanned readmission. Validation analysis (n = 42,269) showed that 7.1% of low-risk individuals, 11.1% of moderate-risk individuals, and 17.1% of high-risk individuals experienced unplanned readmissions (P < 0.001). Pairwise comparisons also showed statistically significant differences between low-risk and moderate-risk participants (P < 0.001), between moderate-risk and high-risk participants (P < 0.001), and between low-risk and high-risk participants (P < 0.001). The area under the ROC curve was 0.622. Conclusions: Our risk model could be helpful for risk-stratifying patients for readmission after surgical treatment for colon cancer. This model needs further validation by incorporating all possible clinical variables.

Link to Full Text

Share

COinS