Discharge destination and readmission rates in older trauma patients.

BACKGROUND: In older trauma patients, the impact of discharge destination on readmission rates is not known. The objective of this study was to evaluate the association between the discharge destination and the 30-day readmission rate in older trauma patients.

MATERIALS AND METHODS: A previously validated database of all patients aged 45 years or older undergoing trauma evaluation at our level 1 trauma center between January 1, 2008 and December 31, 2008 was analyzed to retrospectively compare the incidences of 30-day readmission between patients discharged to home, to inpatient rehabilitation facilities, and to other extended care facilities (ECFs). Demographic information including age and gender and potentially confounding factors including injury severity, trauma activation level, comorbidities, medications, and preinjury functional status were included. Univariate analysis was undertaken using chi-square testing. Multiple logistic regression was performed with potential confounding variables to evaluate for independent contribution to readmission risk.

RESULTS: A total of 960 patients were evaluated; 81 patients (8.4%) were excluded, leaving 879 patients included in the analysis. Seventy-six patients (8.6%) were readmitted within 30 d of discharge. Overall, 6% of those discharged to home, 13% of those discharged to ECF, and 16% of those discharged to rehabilitation were readmitted (P < 0.01 on univariate analysis). Overall, 866 (98.5%) patients had data recorded for all variables analyzed using multiple logistic regression; among these, only discharge destination was independently associated with the rate of readmission (P < 0.01).

CONCLUSIONS: Discharge to ECFs and inpatient rehabilitation facilities appear to be an independent risk factor for hospital readmissions in this population despite controlling for injury severity and comorbidities. Recognition of this risk factor may aid in the disposition planning of these patients and suggests the need for further evaluation of this correlation at other US medical centers.