Knowledge is Power, Can it be Leveraged to Improve Heart Failure Care?

This article refers to ‘Knowledge about self-efficacy and outcomes in patients with heart failure and reduced ejection fraction’ by M. Yang et al., published in this issue on pages 1831–1839. Improving patients' symptoms, function and quality of life, along with reducing disease progression and mortality, are the primary goals in treating heart failure (HF). While US and European guidelines focus on promoting effective treatments, they also strongly emphasize the importance of patient education, ideally with multidisciplinary care teams.1, 2 Despite the emphasis on improving patients' self-care, there are no recommendations for clinicians on how best to screen for patients needing additional education. This is an important gap, because improving patients' knowledge could be a simple, low-cost intervention to improve care and outcomes, but efficiently targeting it to those who most need it would increase its use. A potential explanation for failing to outline strategies to improve patient education may be the absence of data highlighting its association with clinical outcomes. In this issue of the Journal, the paper by Yang and colleagues addresses this gap in knowledge.3 This article is particularly timely because there is increasing interest in the systematic use of patient-reported health status measures, such as the Kansas City Cardiomyopathy Questionnaire (KCCQ), in clinical practice, particularly considering its designation as a quality metric by the American College of Cardiology/American Heart Association Task Force on Performance Measures.4 A unique but often ignored component of the KCCQ-23 is its self-efficacy domain. This domain arose from qualitative research with patients and providers highlighting the complexity of HF management and the need to screen patients for those who might benefit from additional education.5 The self-efficacy domain is not incorporated into the KCCQ summary scores as it measures a very different concept than patients' health status, their symptoms, function and quality of life. There are two items in this scale that examine two separate but related concepts. One asks patients about their confidence in what to do if their condition worsens (Response) and the other about how to prevent their HF from worsening (Prevention). In this issue of the Journal, Yang et al.3 use data from three seminal HF trials (ATMOSPHERE, NCT0085658; PARADIGM-HF, NCT01035255; and DAPA-HF, NCT03036124) to provide a novel insight into the prognostic importance of the KCCQ self-efficacy domain. The authors categorized patients by their response to each question as having a poor (do not understand/not sure at all to somewhat understand/sure), fair (mostly understand/sure), or good (complete) in the KCCQ self-efficacy domain's Response and Prevention items and examined their independent association with hospitalization and death. This extends prior studies that have shown a strong independent association between other KCCQ domains and clinical events6 and has the potential to highlight both the prevalence of poor patient knowledge and the potential importance of addressing it. Remarkably, of the 20 159 subjects, only about a third (37.1%) reported a good understanding of how to prevent their HF from getting worse, and less than half (46.1%) reported a good understanding of how to respond to worsening symptoms. Further, slightly more than one-in-five patients understood their condition poorly, despite almost 60% having been previously hospitalized for HF. These findings are particularly alarming as these data arise from patients enrolled in clinical trials, who might be expected to represent a ‘best-case’ scenario for patient knowledge given the importance of informed consent and the selection of patients likely to adhere to treatment and follow-up. Beyond their descriptive data identifying the large gap in patients' understanding of HF self-care, they discovered a significant association between patients' self-efficacy responses with clinical outcomes. In an unadjusted model, participants with ‘good’, as compared with ‘poor’, response had a 15% and 18% lower hazard for all-cause mortality for Prevention and Response, respectively. Even after multivariable adjustment, the effect remained, suggesting that these items are not surrogates of other patient or disease severity characteristics (e.g. age, glomerular filtration rate, ejection fraction, N-terminal pro-B-type natriuretic peptide, etc.) traditionally associated with death. Interestingly, no significant associations were found between either item and HF hospitalization or time to first HF hospitalization. While this may seem paradoxical, it may be that those with greater knowledge of what to do or whom to call present earlier in the course of their disease progression and be hospitalized with less severe decompensation than those who delay contacting the medical system. This hypothesis is supported by the baseline differences in KCCQ clinical summary scores, which were ∼15 points higher in those with good, as compared with poor, knowledge on the Response and Prevention items – a very large clinical difference. It would have been valuable to further evaluate this possibility by examining disease severity at hospital presentation (e.g. need for intensive care or inotropic support, length of stay). Importantly, and as noted by the authors, it is also possible that the absence of a difference in hospitalization between patients with poor and good Prevention and Response skills could have contributed to the observed differences in survival and that better self-efficacy skills may improve survival. Collectively, these findings of an association with these Prevention and Response skills with survival underscore the importance of expanding future guidelines to highlight the importance of systems to screen patients with HF so that those needing additional education can be identified and educated. One potential strategy for early adopters to consider is to begin routine implementation of patient-reported outcomes into clinical care. Early experiences with the clinical use of the KCCQ at the University of Utah and Stanford reveal benefits from both patients' and providers' perspectives.7-9 Importantly, these efforts used a shortened version of the KCCQ which was created to simplify its use but omits the self-efficacy scale.10 For practices choosing to implement the KCCQ-12, adding the self-efficacy questions may be useful for identifying patients needing additional education. As we seek to better define new strategies to improve HF outcomes, there is a critical need to build on the existing literature defining the impact of education, including the use of evolving technology, to improve outcomes. While the coarse quantification of patients' knowledge from the KCCQ self-efficacy score would not make it an ideal outcome in a clinical trial, its use as an inclusion criterion for future trials of educational interventions or as a screening tool in practice appears promising. In summary, we applaud Yang and colleagues for their creative and important contributions to the field. While there may undoubtedly be unmeasured confounding (e.g. limited assessments of socio-economic status, minimal racial diversity, the possibility that more advanced disease may be out of patients' perceptible control), the finding that a simple assessment of patients' knowledge of how to Respond and Prevent HF exacerbations is associated with mortality underscores the importance of patient education. While our profession has made extraordinary advances in identifying devices and drugs to improve prognosis, we have failed to devote similar energy to ensuring that patients understand their disease and its management. We believe these data should motivate providers to develop better strategies for HF education, particularly given the low risk and costs of such interventions. The descriptive insights of how few patients enrolled in modern clinical trials were confident in managing their HF underscores the urgency with which we need to improve our current care, particularly if it offers the potential to improve our patients' survival. Conflict of interest: C.F.S. is supported by the National Heart, Lung, and Blood Institutes of Health Under Award Number T32HL110837. N.I. received an unrestricted research grant for the Department of Cardiology, Keio University School of Medicine from Bristol Myers Squibb. J.A.S. discloses providing consultative services on patient-reported outcomes and evidence evaluation to Alnylam, AstraZeneca, Bayer, Merck, Janssen, Bristol Myers Squibb, Edwards, Kineksia, 4DT Medical, Terumo, Cytokinetics, Imbria, and United Healthcare. He holds research grants from Bristol Myers Squibb, Abbott Vascular and Janssen. He owns the copyright to the Seattle Angina Questionnaire, Kansas City Cardiomyopathy Questionnaire, and Peripheral Artery Questionnaire and serves on the Board of Directors for Blue Cross Blue Shield of Kansas City.