Background: The Centers for Medicare and Medicaid Services Hospital Readmissions Reduction Program (HRRP) has created financial penalties for hospitals with higher risk-standardized readmission rates after hospitalization for specific conditions, including congestive heart failure (CHF). Identification and risk-standardization is performed with administrative data. Both to evaluate the utility of this metric for improving quality and to inform efforts by providers to reduce readmissions, more granular clinical information about patients included in the penalty is needed.
Methods: All patients who contributed to the CHF component of the HRRP penalty at our hospital between June, 2012 and December, 2013 were identified and medical records were reviewed by a physician (A.J.C.). Information extracted included index inpatient service, cause of readmission, medications, scheduled follow-up, whether echocardiogram was performed, NT pro-BNP as measured at admission and discharge, weight documented during index hospitalization, last known ejection fraction, and discharge disposition.
Results: During this time period, 212 readmitted CHF patients contributed to the HRRP penalty. Of those, 31 (14.6%) were excluded due to readmission to an outside hospital. Of the remaining 181, 6 (2.8%) were excluded as the cause of index admission was not confirmed to be CHF. These patients were excluded from all analyses. Of the remaining 175 patients, 79 (45.1%) were readmitted for recurrent CHF exacerbation while 96 (54.8%) were readmitted for reasons other than CHF. Seventy (40%) patients were discharged home with visiting nurse services, 44 (25.1%) discharged home without services, and 61 (34.9%) discharged to a skilled nursing facility. Of the 114 patients discharged home, 44 (38.6%) had follow-up scheduled at the time of discharge. The median length of time between the index admission and readmission was 13 days. Among the 79 patients readmitted for CHF-related causes, 39 (49.4%) initially had been hospitalized on the cardiology service while 40 (50.6%) were hospitalized on a general medical service. Of those 79 patients, 32 (40.5%) had left ventricular systolic dysfunction (LVSD). At time of index discharge, 28 (87.5%) of patients with LVSD had been prescribed a beta blocker and 12 (37.5%) had been prescribed an ACE inhibitor or ARB.
Conclusions: About half of patients who contributed to the CHF component of the HRRP penalty at our hospital were readmitted for recurrent CHF, the other half for different diagnoses. Of those who were readmitted with recurrent CHF, more than half did not have systolic dysfunction. Many of the patients with recurrent CHF due to systolic dysfunction were not initially discharged on ACE or ARB therapy, largely due to hypotension or renal dysfunction. These findings underscore the challenges of reducing preventable hospital readmission in this HRRP penalty population.