Reducing Readmissions: Findings from an Urban Academic Medical Center

Background: Hospital readmission rates are increasingly being utilized as a measure of quality and to modify hospital reimbursement. Many have argued that a significant proportion of readmissions can be prevented if modifiable factors had been intervened upon, but recent studies suggest that the majority of readmissions are most likely not preventable. In order to reduce overall hospital readmissions, clinicians and administrators should target factors which specifically address these preventablereadmissions.Objectives: This study seeks to identify trends in readmissions at the George Washington University Hospital and to identify factors which are associated with preventable readmissions. Results of this study will be compared with previously published national data. Results suggesting any unique characteristics of this institution's systemic, clinician, or patient-based factors contributing to early hospital readmissions can then be used to identify and prioritize targets for future improvement projects with the aim of reducing hospital readmissions.Design: A cross sectional survey was used to determine the proportion of readmissions which are preventable and identify associated risk factors. The survey used was modified from previously validated tools.Methods: All attendings in the division of hospital medicine attended a brief orientation session. Patients presenting as 30-day readmissions were identified using weekly readmissions reports. For each of those patients, the most recent attending physician was asked to indicate whether the readmission was more likely than not a preventable readmission. If yes, they were asked to select from several possible factors which may have contributed to that readmission.Results: To date, the survey response rate is 98%. Of the 163 responses available in the preliminary analysis of this ongoing study, 57% of readmissions were thought to be potentially preventable. Of those preventable readmissions, the most frequently identified associated factors were patient or caregiver's inability to adequately manage condition (49%), patients' non-medical social issues (29%), patients' failure to keep follow up appointments (13%), patients' uncertainty of whom to contact for outpatient care or when to return to ED (11%).Conclusions: Initial analysis indicates that GWUH may have higher levels of preventable readmissions when compared to other similarly conducted studies, in which rates of preventable readmissions ranged from 26.9% to 38%. Potential reasons for this finding range from the subjective nature of defining preventability to institutional or patient-population specific causes. The factors driving readmissions at GWUH are largely patient-related rather than hospital process-related, adding to the growing debate over whether readmissions are a reliable measure of hospital quality.