Factors Associated with Readmission of Cardiac Patients
Carole A. South-Winter, Ali Dag, SriniRagothamoan

The aim of this research project was to develop a risk score for cardiac patient readmissions that would be specific and sensitive to a particular population. Additionally, a score that includes more variables that can be manipulated which is also inexpensive and easy to administer. Prior research revealed several significant risk factors for hospital readmissions and this research has added to that research stream. Cardiac patient data were obtained from a large integrated rural health system in the Upper Midwestern states in the U.S. Statistically significant mean differences (t- tests) between two groups, readmitted cardiac patients (2,687) vs. non-readmitted cardiac patients (16,575), were found for ten variables: A multivariate test (namely, logistic regression) was also performed; the logistic regression results mostly confirm T-test results. Based on these statistical results, a Risk Score for Cardiac Patient Readmissions was developed and we found significant utility for such a score. Two key risk factors used in the presented Risk Score for Cardiac Patient Readmissions were not part of other widely used risk scores. The two new risk factors included are: No Shows and number of prescriptions. No shows could be a proxy for living alone, poverty, distance to the hospital, lack of interest, and other social/motivational factors. Number of prescriptions could be a proxy for comorbid illnesses, drug interactions, and age. The study determined that these variables, previously undefined in risk of admission categorization, have significant impact on risk for readmission.

Full Text: PDF     DOI: 10.15640/ijhs.v6n4a3