Reducing hospital readmissions has been the focus of much attention recently. Patients who readmitted may have been ill‐prepared for discharge. One measure of preparedness has been endorsed by the National Quality Forum, but other measures of preparedness exist. We sought to determine which measure of preparedness better predicts readmissions.
The Vanderbilt Inpatient Cohort Study (VICS) is an ongoing prospective study that seeks to better define the relationship of social determinants of health with patient outcomes after hospitalization. The study population includes adults hospitalized with acute coronary syndromes (ACS) or acute decompensated heart failure (ADHF). Following hospital discharge, patients completed a phone interview that included two measures of preparedness for discharge: Care Transition Measure‐3 (CTM‐3, range 0‐100) and Brief PREPARED (B‐PREPARED, range 0‐22). Readmission during the 90 days following hospital discharge was analyzed as a binary outcome, as well as days to first readmission. We examined the correlation between the preparedness measures, preparedness and rate of readmission, odds of readmission, and time to readmission, as well as the discriminative ability of the preparedness measures in predicting readmission.
Among 805 patients, the mean age was 60.3 (+12.6) years. The median preparedness scores were CTM‐3 78 (interquartile range (IQR) 67‐100) and B‐PREPARED 20 (IQR 18‐22). The CTM‐3 and B‐PREPARED were moderately correlated (Spearman’s rho 0.4). In the 90 period following hospital discharge, readmission rates were significantly different by category of preparedness (Tables 1 and 2). In logistic regression, higher CTM‐3 (odds ratio (OR)=0.989) and B‐PREPARED scores (OR=0.931) were associated with lower odds of readmission within 90 days. Days to readmission were significantly different across categories of CTM‐3 (p=0.000) and B‐PREPARED (p=0.016) scores. Finally, using rank correlation the ability of B‐PREPARED to discriminate between patients readmitted and those who were not was slightly better than CTM‐3 (Somers’ d 0.075 vs. 0.067.
Readmission rates, odds of readmission, and days to readmission increased as preparedness for discharge decreased. B‐PREPARED measure performed slightly better in discriminating patients who are and are not readmitted within 90 days of hospital discharge.