An Attempt to Estimate the Proportion of Preventable Hospital Readmissions and to Characterize Responsible Factors

• Main Author: Adhaduk, Mehul
• Format: Dissertation
• Language: English
• Published: ProQuest Dissertations & Theses 01-01-2018
• Subjects: Biostatistics; Epidemiology
• ISBN: 1392006872; 9781392006870

Objective: To estimate the proportion of preventable hospital readmissions and to characterize their potential causes. Methods: This is a descriptive epidemiological study. One hundred patients were randomly selected out of 169 patients from a database constructed under a Centers for Medicare & Medicaid Services grant. There was a total of 127 readmissions among 100 patients. Eighteen patients (25 readmissions) were excluded as these patients were admitted for psychiatry care. A readmission classification instrument was derived from a literature review, the CMS guide on preventable readmission, and the opinions of the study investigators. Preventable readmissions were identified with the readmission classification instrument by manual review. Results: Study population had a mean age of 67 years, was predominantly non-Hispanic white (97%) and 57% male and all were Medicare beneficiaries. Patient demographics were similar in preventable and non-preventable readmissions. We classified 17 (16.7%, CI 10%-25.3%) readmissions as preventable; the most common reasons for preventable readmissions were premature discharge (4 [23.5%]) and acute complication after a medical procedure (3 [17.6%)]. Median (interquartile range (IQR)) days after discharge was 9 (9) for preventable readmission compared to 15 (12) for non-preventable readmission, p-value 0.045. Conclusion: Preventable readmission represents a small portion of all readmissions. Leading causes of preventable readmissions were premature discharge and medical readmission after a medical procedure.