Surgical intensive care unit admission variables predict subsequent readmission

Area of Science:

• Critical Care Medicine
• Health Services Research
• Surgical Outcomes

Background:

• Intensive care unit (ICU) readmissions increase healthcare resource utilization.
• Predicting and mitigating ICU readmissions is essential for optimizing patient care and resource allocation.
• Surgical ICU patient characteristics may differ between those readmitted and those not.

Purpose of the Study:

• To identify predictors of readmission in surgical ICU patients.
• To compare characteristics of readmitted versus non-readmitted ICU patients at the time of initial admission.
• To inform quality improvement initiatives for ICU discharge criteria.

Main Methods:

• Retrospective cohort study using a prospectively maintained database (APACHE IV quality database).
• Comparison of patients admitted between January 1 and December 31, 2011.
• Categorical variables analyzed using Fisher's exact test; continuous variables using t-tests.
• Multivariate logistic regression to identify independent predictors of readmission.

Main Results:

• A readmission rate of 12.8% (77 of 765 admissions) was observed.
• Readmitted patients (REA) had significantly higher admission severity of illness (APACHE III score) compared to non-readmitted patients (NREA).
• Multivariate analysis indicated emergency surgery, history of immunosuppression, and higher Acute Physiology Score as independent predictors of readmission.

Conclusions:

• Surgical ICU patients requiring readmission exhibit distinct admission profiles compared to those not readmitted.
• Understanding these differences is key to developing targeted quality improvement projects.
• Tailoring discharge criteria based on patient-specific risk factors may reduce ICU readmissions.