Background:
Readmission to the hospital is a burden not only to patients, families, and health care workers but also to the health care system. The U.S. government spends an estimated $12 billion a year on potentially preventable readmissions for Medicare patients. As readmission rates have been accepted as a quality measure, hospitals with high rates face public disclosure and potentially serious pay cuts.
Objectives:
The objectives were (1) to detect the readmission rates of a hospitalist group at a community hospital, (2) to identify probable causes, and (3) to propose solutions to decrease the rate of readmissions.
Methods:
We conducted a retrospective chart review. Patients who were rehospitalized with the same diagnoses within 30 days were selected for the study among consecutive admissions over 1 year (January 1–December 31, 2008). Data collection included patient characteristics, admitting diagnoses, comorbidities, length of stay, discharge disposition, accuracy of medication list at discharge, if follow‐up appointments were scheduled, and mortality rates. Three hospitalists reviewing the charts commented on whether each readmission may have been preventable.
Results:
Among 5206 patients who were admitted to the hospitalist service over 1 year, 85 (1.6%) were re hospitalized with the same diagnoses within 30 days. Of the 85 readmitted patients, 47% were male, 82% were white, with a mean age of 58 ± 17 years. Eighty‐four percent of these patients had been hospitalized within the previous year. 76% of whom had at least 1 other hospitalization within the next 3 months. The top 3 diagnoses were sepsis, pneumonia, and COPD. Fifty‐two percent of patients had more than 6 comorbidities. Follow‐up appointments were made for only 28% of patients at first admission. Ninety percent of patients received an accurate medication list at discharge. Noncompliance was identified in 29% of the patients on readmission, Overall mortality rales were 2.4% and 15.3% during readmission and wilhin the next 3 months, respectively. Mortality within 3 months was higher in patients with sepsis (P = 0.005), more comorbidities (10 vs. 6, P = 0.004), longer length of stay at first hospitalization (5 vs. 3 days, P = 0.03), and among those discharged to nursing home at readmission (P = 0.03). Only 4.7% of readmissions were concluded to be preventable
Conclusions:
Our readmission rate (1.6%) is significantly lower than that of previous studies (23.2%). as this is a subgroup of readmissions only due to the same diagnoses. In this particular group, multiple unmodifiable factors such as multiple comorbidities, advanced‐stage disease, andlow functional status most likely could restrict the possibility of improvement in readmission rates. Patient education, family involvement in the discharge process, and having staff schedule follow‐up appointments before discharge potentially could avoid readmissions. We suspect that all cause‐related readmissions have more room for improvement, which should be the focus of the intervention.
Author Disclosure:
B. Cakir, none; G. Gammon, none; D. Austin, none.