The Heart Failure Readmission Quagmire: Taking a Deep Dive to Find Solutions

This article refers to ‘Aetiology, timing and clinical predictors of early vs. late readmission following index hospitalization for acute heart failure: insights from ASCEND-HF’, by M. Fudim et al., published in this issue on pages 304–314. Patients hospitalized with acute decompensated heart failure (HF) frequently experience the ordeal of another hospitalization shortly after discharge. Such hospitalizations are often unplanned and place substantial burden on the health care system.1 In several Western health care systems, 30-day HF readmission has received significant attention as an important metric of hospital-level quality of care.2, 3 Readmission within 30 days for HF and other common conditions, such as myocardial infarction and pneumonia, has been targeted by interventions that aim to lower health care spending on hospitalizations.2, 4 In the USA, for example, all hospitals receiving reimbursement for inpatient care through Medicare (a large public health insurance programme predominantly for adults aged ≥65 years) publicly report their 30-day readmission rates for these conditions, which allows for the comparison of readmission performance across hospitals. The hospitals that substantially exceed average readmission rates over a calendar year are identified and incur financial penalties.5, 6 The enactment of these health care policies has been followed by improvements in rates of readmission.1 However, several aspects of readmission after HF hospitalization remain poorly understood. Although recent work has demonstrated that the overall quality of hospital care may be an important contributor to 30-day readmission rates independently of patient-level factors,4 objective measures of care practices that adequately quantify care quality factors relevant to readmission have remained elusive.5, 6 Further, recent studies have demonstrated no consistent association between hospital-level readmission rates and 30-day or long-term risk-standardized mortality rates.6, 7 Finally, despite early gains with legislative action, readmission rates have remained nearly unchanged over several years.8 Therefore, the continued improvement of long-term outcomes associated with HF hospitalization requires a better characterization of the heterogeneous nature of HF readmissions. In this issue of the Journal, Fudim and colleagues9 leverage the detailed phenotypic characterization of HF in the randomized clinical trial Acute Study of Clinical Effectiveness of Nesiritide in Decompensated Heart Failure (ASCEND-HF)10 in order to better understand the timing of HF readmissions, their aetiologies and implications for survival at 6 months. Consistent with the findings from previous studies, Fudim et al. observed that the majority of 30-day readmissions among HF patients occur within the first 2 weeks and are not related to HF exacerbation.9 The authors also identified prior HF hospitalization in the past year, the presence of chronic respiratory disease, and clinical symptoms of decompensation during the index hospitalization as important predictors of early readmission.9 In addition to confirming some key observations noted in previous studies, this study9 advances our understanding of HF readmission in several important ways. Firstly, it extends the study of readmission to a more diverse patient population, not limited by either age or geography. This contrasts with several prior investigations of HF readmission that represent practices in a few selected hospitals or patients of a particular age group.2, 3, 11 By including younger patients with HF, the results of this study9 may highlight practices for a wider spectrum of HF patients. This may also be reflected in the 30-day readmission rate in the present study cohort,9 which is substantially lower than those in prior studies.1, 3, 12 Secondly, the study9 provides contemporary evidence of differences in post-hospitalization readmission rates across different geographic regions and shows that readmission rates at North American centres may be at least two-fold higher, even for patients with similar clinical characteristics. This may suggest that post-hospitalization care at North American sites, particularly those in the USA, is more reliant on inpatient resource use than it is in European and Asian countries. Finally, Fudim et al.9 observed that 30-day readmission was associated with a higher risk for 6-month mortality, thereby highlighting the important role of HF readmission in identifying higher-risk HF patients. However, it is unclear how these observations translate to a hospital level, particularly as prior studies have shown these two outcomes to be independent.5, 6 Although the current study9 adds to our understanding of readmission for HF at the patient level, translating this understanding to the health care system level to improve quality of care and outcomes would require additional work. Specifically, although understanding the clinical predictors of readmission is important for identifying patients at higher risk, the capture of non-traditional risk factors that may influence risk for readmission requires the modification of systems of care. At a patient level, such social drivers of readmission may include the nature of societal and family support, as well as emotional and financial well-being, which vary substantially. Therefore, collecting this information regularly and uniformly may help us to develop and evaluate interventions that are most suited to the needs of the patients served by a particular health system. At a hospital level, there is a need to use experiences from hospitals that have been successful in lowering HF readmission rates to guide efforts at hospitals that have been less successful.11 This may, in turn, require research studies that leverage a larger portfolio of research methodology, specifically through the use of mixed-methods research to better understand care dynamics at highly and poorly performing hospitals, analogous to their successful implementation in care for myocardial infarction.13 The hospital interventions that have been effective in lowering readmission rates include evidence-based transitions of care models and strategies to mitigate readmission risk through community care interventions. These include a wider use of transitional care centres, early post-discharge follow-up, specialized HF clinic engagement, and nurse-led community health interventions.14–16 Finally, further studies are needed to better elucidate the effects of readmission on patients, and the HF readmissions that have the most impact on patient well-being. Moreover, it is critical to define the types of readmission that are truly preventable in order to better direct efforts towards mitigating these events. In conclusion, better characterization of HF readmissions is critical to improving results on this important patient outcome. Future studies should evaluate whether or not a patient-centred approach to the design and implementation of interventions might influence the scourge of HF readmission. Conflict of interest: none declared.