Association of Serial Kansas City Cardiomyopathy Questionnaire Assessments With Death and Hospitalization in Patients With Heart Failure With Preserved and Reduced Ejection Fraction: A Secondary Analysis of 2 Randomized Clinical Trials

Key Points

Question  When assessing a patient with heart failure, which patient-reported outcome assessment (current score, prior score, or a change in score) is most prognostic of subsequent death and heart failure hospitalization?

Findings  In this study, while current, prior, or a change in the Kansas City Cardiomyopathy Questionnaire score were all significantly associated with mortality and heart failure hospitalization in isolation, when either prior and current Kansas City Cardiomyopathy Questionnaire score or change and current score were considered together, only the current Kansas City Cardiomyopathy Questionnaire score was significantly associated with the outcomes, irrespective of ejection fraction.

Meaning  While interpreting current, prior, or a change in Kansas City Cardiomyopathy Questionnaire score, the most recent assessment provides the most important information about the risks for subsequent clinical events.

Importance  While there is increasing emphasis on incorporating patient-reported outcome measures in routine care for patients with heart failure (HF), how best to interpret longitudinally collected patient-reported outcome measures is unknown.

Objective  To examine the strength of association between prior, current, or a change in Kansas City Cardiomyopathy Questionnaire (KCCQ) scores with death and hospitalization in patients with HF with preserved (HFpEF) and reduced (HFrEF) ejection fractions.

Design, Setting, and Participants  Secondary analyses of the Treatment of Preserved Cardiac Function Heart Failure With an Aldosterone Antagonist (TOPCAT) trial of 1372 patients with HFpEF, conducted between August 2006 and January 2012, and the HF-ACTION trial that included 1669 patients with HFrEF, conducted between April 2003 and February 2007.

Exposures  Prior, current, and change in KCCQ Overall Summary scores (KCCQ-os) in 5-point increments (higher scores indicate better health status).

Main Outcomes and Measures  Time to cardiovascular death/first HF hospitalization (primary outcome) and all-cause death (secondary outcome).

Results  Of 1767 eligible TOPCAT participants, 882 were women (49.9%), and the mean (SD) age was 71.5 (9.7) years. Of 2130 eligible HF-ACTION participants, 599 were women (28.1%), and the mean age was 58.6 (12.7) years. Each 5-point difference in prior or current KCCQ-os scores was associated with a 6% (95% CI, 4%-8%; P < .001) to 9% (95% CI, 7%-11%; P < .001) lower risk for subsequent cardiovascular death/first HF hospitalization in patients with HFpEF and 6% (95% CI, 4%-9%; P < .001) to 8% (95% CI, 5%-10%; P < .001) lower risk for subsequent cardiovascular death/first HF hospitalization in patients with HRpEF and HFrEF in unadjusted analyses. Results were similar for change in KCCQ-os. In models with the prior and current KCCQ-os, only the current KCCQ-os was significantly associated with 10% (95% CI, 7%-12%; P < .001) and 7% (95% CI, 3%-11%; P < .001) lower risk for subsequent cardiovascular death/first HF hospitalization in patients with HFpEF and HFrEF, respectively. Similar results were observed when the current and Δ KCCQ-os were considered together, when adjusted for important patient and treatment characteristics, when including 3 sequential KCCQ-os scores, and when examining all-cause death as the outcome.

Conclusions and Relevance  In serial health status evaluations of patients with HF, the most recent KCCQ score was most strongly associated with subsequent death and cardiovascular hospitalization in HFpEF and HFrEF. Measuring serial patient-reported outcome measures in the clinical care of patients with HF can provide an updated assessment of prognosis.

Trial Registration  clinicaltrials.gov Identifier: NCT00094302 (TOPCAT) and NCT00047437 (HF-ACTION)

Systematically quantifying patients’ health status (their symptoms, function, and quality of life) has the potential to transform clinical care by incorporating patients’ experiences into the assessment of novel treatments, into measures of health care quality, as a means to monitor the symptoms and prognosis of patients, and as a foundation for engaging them in shared decision making.¹ Not only does the routine use of patient-reported outcome measures (PROMs) directly address the Institute of Medicine’s goals for more patient-centered care,² it also responds to growing calls for PROMs to be an integral part of quality assessment and improvement.³ As new technologies, such as patient portals and smart devices, further improve the feasibility of routinely collecting PROMs, a key challenge is to develop a framework through which clinicians can clinically interpret these measures.⁴

Quiz Ref IDChronic heart failure (HF) is a prototypical example of why PROMs must be integrated into routine clinical care. First, patients feel strongly about their quality of life,⁵ which is a primary goal of treatment,⁶ and monitoring and sharing with patients the changes in their health status can be valuable.⁷ Second, PROMs, such as the Kansas City Cardiomyopathy Questionnaire (KCCQ), have been shown to be strongly and independently associated with subsequent death and hospitalization.^8-11 This suggests that PROMs could be useful in risk stratifying patients for more intensive therapy, a guideline-recommended process for clinical treatment.⁶ Further underscoring the importance of developing strategies to interpret serial PROMs in HF are the efforts of agencies, such as the International Consortium for Health Outcomes Measurement¹² and the Centers for Medicare and Medicaid Services,¹³ to create PROM-based measures for quantifying the quality of care delivered to patients with chronic HF. As these efforts mature, clinicians will be required to collect serial PROMs, and improving the clinical usefulness of these measures can transform their collection from an additional mandate to a clinically useful process of care. Accordingly, we sought to understand how best to leverage health status assessments (current, prior, or the change in scores over time) to stratify the risk of subsequent mortality and HF hospitalizations in stable outpatients with HF. In so doing, this study will help clinicians better interpret these scores and facilitate the transformation of emerging Medicare performance measures into clinically useful tools that can improve care.

We accessed deidentified data from 2 clinical trials through the National Heart, Lung, and Blood Institute data repository (https://biolincc.nhlbi.nih.gov/studies/hf_action/ and https://biolincc.nhlbi.nih.gov/studies/topcat/), with institutional review board approval from Saint Luke’s Hospital of Kansas City, Missouri, which granted a waiver of informed consent. The Treatment of Preserved Cardiac Function Heart Failure With an Aldosterone Antagonist (TOPCAT) trial randomized 3445 symptomatic patients with HF with preserved ejection fraction (HFpEF; left ventricular ejection fraction ≥45%) from multiple countries between August 2006 and January 2012 to examine whether spironolactone would reduce a composite end point of cardiovascular death, cardiac arrest, and HF-related hospitalization when added to standard medical therapy.¹⁴ There was no significant reduction in the primary outcome with spironolactone when added to standard medical therapy in the overall population, although there was a significant improvement in the KCCQ Overall Summary score (KCCQ-os).⁵ However, in TOPCAT, there were concerns about the validity of data from patients enrolled in Russia and Georgia (48.7% of trial population),^15,16 so we decided, a priori, to exclude patients from Russia and Georgia (n = 1678) in the current analyses, leaving a final analytic cohort from TOPCAT of 1372 patients (Figure 1).

The Heart Failure: A Controlled Trial Investigating Outcomes of Exercise Training (HF-ACTION) trial randomized 2331 outpatients with HF with reduced ejection fraction (HFrEF; left ventricular ejection fraction ≤35%) between April 2003 and February 2007 to supervised aerobic exercise training followed by home-based exercise or standard medical therapy alone.¹⁷ The trial did not show a significant reduction in all-cause mortality or hospitalization,¹⁷ but there was a significant improvement in KCCQ-os.¹⁸ After excluding 201 patients who declined to share data through the National Heart, Lung, and Blood Institute and 461 patients with missing KCCQ data, our analytic cohort from HF-ACTION included 1669 patients (Figure 1).

Quiz Ref IDThe KCCQ is a 23-item, validated questionnaire to assess HF-specific symptoms, function, and quality of life.¹⁹ It has been validated in patients with HFpEF and HFrEF^8,20 and proven to be both reproducible and sensitive to important changes in HF health status.^8,19,21,22 The questionnaire has several domains, including physical limitation, symptoms, quality of life, and social limitations, that can be summarized by the KCCQ-os. Scores are transformed into values between 0 and 100, with higher scores indicating better HF-specific health status, and a 5-point difference is clinically important.²¹

Details about patient characteristics, serial KCCQ assessments, and primary outcome results for both trials have been previously described.^14,17 The KCCQ was collected at several points, including at randomization, 4, 12, and 24 months in the TOPCAT trial⁵ and at randomization, 3, 6, and 9 months in the HF-ACTION trial.¹⁸ Because we anticipated substantial changes in health status after randomization and wanted to examine patients in a more chronic state of their HF, we excluded the KCCQ assessment at randomization for these analyses. To contextualize our analyses, we envisioned a patient being seen at the second follow-up visit and the clinician wanting to know which KCCQ assessment to pay most attention to in terms of prognosis: the current score (second visit score), the prior score (first visit score), or the change from the first to second visit (Figure 2). Patients without KCCQ-os assessments at both visits were excluded.

All analyses were performed separately for patients with HFpEF (TOPCAT) and HFrEF (HF-ACTION). Patient and treatment characteristics were compared across ranges of KCCQ-os using χ² or Fisher exact test for categorical variables and 1-way analysis of variance for continuous variables. We then constructed a series of models to examine the association of KCCQ-os with time to cardiovascular death or first HF hospitalization (primary outcome) and with time to all-cause death (secondary outcomes). Using our conceptual framework of a physician seeing the patient at their second follow-up visit, we examined the prognostic importance of the following assessments and combinations: (1) current KCCQ-os; (2) prior KCCQ-os; (3) change from prior to current KCCQ-os, all in isolation and as combinations of (4) prior and current KCCQ-os; and (5) change from prior to current KCCQ-os and current KCCQ-os (Figures 3 and 4) to assess the incremental contribution of other assessments to the current assessment of patients’ health status.

In the primary analyses, these associations were examined in unadjusted models to mirror their potential use in clinical care, where multivariable-adjusted analyses are unlikely to be routinely performed. However, to illuminate the independent prognostic value of KCCQ scores, we also conducted multivariable Cox proportional hazards models as secondary analyses, with adjustment for variables determined, a priori, that could potentially confound the association between health status and clinical outcomes. These variables included age, sex, race/ethnicity, systolic blood pressure, heart rate, body mass index, smoking status, hypertension, diabetes, atrial fibrillation, prior myocardial infarction, prior stroke, chronic obstructive lung disease, pacemaker implantation, left ventricular ejection fraction, use of angiotensin-converting enzyme inhibitor or angiotensin receptor blocker, use of β-blocker, use of aldosterone antagonist (HF-ACTION only), etiology of HF (ischemic vs nonischemic; HF-ACTION only), randomization to spironolactone (TOPCAT only), and randomization to exercise training (HF-ACTION only). All variables were assessed at the time of randomization. Hazard ratios (HRs) for KCCQ-os scores were scaled per 5 points because prior studies have shown this to be a clinically important magnitude of change from patients’ and clinicians’ perspectives.^21,23 We explored nonlinear relationships between various KCCQ scores and outcomes by testing restricted cubic splines, but none were significant, suggesting that the associations between KCCQ assessments with clinical outcomes were linear.

Because it is possible that even earlier scores might provide a trajectory beyond what would be captured with the current and prior scores, we also conducted a sensitivity analysis that examined the association between each of 3 sequential KCCQ scores with subsequent prognosis. For TOPCAT, this included the 4-month, 12-month, and 24-month scores to predict events beyond the 24-month visit, and for HF-ACTION, this included the 3-month, 6-month, and 9-month scores to predict events beyond 9 months (eFigures 1 and 2 in the Supplement). Patients without KCCQ-os assessments at all 3 visits were excluded. We labeled these 3 sequential KCCQ assessments as first, second, and third visits.

Among the patients in the analytic cohorts, the clinical data were quite complete, with only 0.7% of data missing in TOPCAT and 3.9% of data missing in HF-ACTION. Missing data were addressed using multiple imputation (IVEWare). We considered a 2-sided P value less than .05 as statistically significant and did not adjust for multiple comparisons. All analyses were performed using SAS Software, version 9.4 (SAS Institute).

Our analytic cohort included 1372 patients with HFpEF from TOPCAT (mean [SD] follow-up, 3.3 [1.4] years) and 1669 patients with HFrEF from HF-ACTION (mean [SD] follow-up, 2.7 [1.0] years; Figure 1). In both cohorts, patients with worse health status at baseline were more likely to be younger, have a higher body mass index, and have diabetes (Table and eTable in the Supplement). In patients with HFpEF, there were 321 patients who experienced a cardiovascular death or first HF hospitalization during follow-up and 229 patients who died, with an event rate of 7.1 and 5.0 per 100 person-years, respectively. In patients with HFrEF, there were 327 patients with cardiovascular death or first HF hospitalization and 237 patients who died, with an event rate of 7.2 and 5.2 per 100 person-years, respectively.

The results from our primary analyses of the associations of individual and combined KCCQ-os assessments with cardiovascular death or first HF hospitalization are shown in Figure 3. The results were similar in patients with HFpEF and HFrEF. Assuming the patient was being evaluated at the second follow-up appointment, higher scores on the current KCCQ-os assessment or the prior KCCQ-os assessment were both associated with a lower risk of cardiovascular death or first HF hospitalization (HR, 0.91; 95% CI, 0.89-0.93 and HR, 0.92; 95% CI, 0.90-0.95 in patients with HFpEF and HFrEF, respectively, per 5-point higher KCCQ-os in univariate analyses). Results were similar for the prior KCCQ-os assessment. Quiz Ref IDAn increase in KCCQ-os from prior to current visit was similarly associated with lower risk of cardiovascular death or first HF hospitalization (HR, 0.94; 95% CI, 0.91-0.97 and HR, 0.95; 95% CI, 0.91-0.99 in patients with HFpEF and HFrEF, respectively, per 5-point change in KCCQ-os). However, when the current visit KCCQ-os was included in the model with either the prior visit score or the change from prior to current score, only the current KCCQ-os was significantly associated with cardiovascular death or first HF hospitalization, indicating that the most recent KCCQ-os was the most prognostically important assessment. Results were similar in fully adjusted models (eFigure 3 in the Supplement).

Results for all-cause mortality were similar to that for cardiovascular mortality or first HF hospitalization (Figure 4). Higher scores on current and prior KCCQ-os, taken individually, were associated with lower risks of all-cause mortality (HR, 0.93; 95% CI, 0.90-0.96 and HR, 0.93; 95% CI, 0.90-0.95 in patients with HFpEF and HFrEF, respectively, per 5-point higher KCCQ-os in univariate analyses; Figure 4). Similarly, an increase in KCCQ-os from the prior to current visit was associated with a lower risk of all-cause mortality (HR, 0.95; 95% CI, 0.92-0.99 and HR, 0.91; 95% CI, 0.87-0.96 in patients with HFpEF and HFrEF, respectively, per 5-point change in KCCQ-os). However, when the current KCCQ-os was included along with either the prior or change in KCCQ-os, only the current KCCQ-os was significantly associated with lower risks for all-cause death (Figure 4). The results were similar in fully adjusted models (eFigure 4 in the Supplement).

When shifting the analyses to the second and third assessments, only the last KCCQ-os score was significantly associated with the risks for both primary and secondary outcomes (eFigures 1 and 2 in the Supplement), congruent with the main analyses. When 3 sequential KCCQ-os assessments were considered together, the last KCCQ-os was most significantly associated with the primary and secondary outcomes (eFigures 1 and 2 in the Supplement). In addition, for a few of the outcomes (primary outcome in HFpEF and secondary outcome in HFrEF), there was a direct association with the earliest KCCQ-os.

As health care strives to become more patient-centered and payers increasingly demand evidence of health care value, the pressure to include PROMs in routine care is increasing. However, understanding how best to use and interpret these measures is unknown and important to understand so that they can have intrinsic value to clinicians. This study provides important insights into the interpretation of longitudinally collected health status data in patients with HF. When serially monitoring health status in patients with HF, we found that patients’ current health status was most prognostic. Quiz Ref IDImportantly, the association of KCCQ-os with outcomes was very similar in patients with HFpEF and HFrEF. Moreover, there were few differences in the unadjusted and adjusted results, suggesting that the prognostic information from PROMs is unique and minimally confounded by other clinical data. Collectively, these findings suggest that serial monitoring with the KCCQ can provide updated prognostic information in all patients with HF, which can be an important reference for subsequent treatment decisions.

This study extends prior work describing the association between KCCQ-os and other clinical outcomes. Both the initial KCCQ-os and change in KCCQ-os have previously been shown to be independently prognostic of survival and HF admission in patients with HFpEF⁸ and HFrEF.^8-10,20 However, none of these studies examined the incremental prognostic significance of the later KCCQ-os. To our knowledge, the question of which KCCQ assessments are most important when more than 1 measure is available has not been previously established and is of critical importance when considering the use of the KCCQ in clinical practice. Given the growing demand for PROMs in HF, highlighted by the Centers for Medicare and Medicaid Services Merit-Based Incentive Payment System that rewards clinicians for collecting PROMs and their efforts to create a PROM-based quality measure,²⁴ it is important that clinicians understand how best to interpret serial PROMs.

There are multiple potential clinical benefits of longitudinal and routine health status assessment, which seems to be increasingly feasible to collect. The use of the internet, smart devices, or patient portals in electronic health records can facilitate serial collection of PROMs, which can then be incorporated into patients’ records.⁴ The instantaneous integration and availability of longitudinal HF health status data also have the potential to support better patient-physician communication and shared decision making and may even improve patients’ outcomes.⁴ Our study provides novel insights into how clinicians may use serial KCCQ data, beyond its use as a cross-sectional description of patients’ symptoms, function, and quality of life, to update patients’ prognoses and empower them to receive timely interventions using the most current risk estimates. While it has been suggested that PROMs be collected as an outcome for disease management programs,²⁵ to be able to deliver more patient-centered care, the KCCQ may also be considered in selecting appropriate patients for HF disease management programs because it continually updates risks for hospitalization and death.²⁶Quiz Ref ID Furthermore, as higher KCCQ scores have been shown to be associated with lower cost and resource use,²⁷ serially monitoring KCCQ scores in routine clinical care may support more efficient care by allowing timely identification of patients with worsening health status who may benefit from appropriate interventions. Such potential benefits of routinely monitoring the health status of patients with HF warrant future study²⁸ and may prove to be a less expensive approach than using serial biomarker (eg, N-terminal pro-B-type natriuretic peptide) or imaging.

These analyses should be considered in the context of the following potential limitations. First, although the follow-up times between KCCQ assessments in TOPCAT were longer than in HF-ACTION, this is not dissimilar from real-world practice where clinic follow-up is irregular. Despite differences in follow-up time, we found similar results for patients with HFrEF and HFpEF. Second, our findings are based on rigorously collected data from clinical trials on patients eligible for these trials. Our results should ideally be replicated in a real-world registry of allcomers with HF, once KCCQ scores are routinely monitored in clinical care. Third, the effect of longitudinally changing patient characteristics as well as multiple KCCQ assessments over time was not examined in the current analyses. It is possible that the integration of KCCQ results over multiple points could prove to be a better marker for death and hospitalization. When including 3 sequential KCCQ scores, we found that the latest was most prognostically important, but the earliest assessment was also associated with some outcomes. Future studies, with longer periods of follow-up, will be needed to see if more complex analyses of trajectories provide improved prognostic value, but the simplicity of focusing on the latest assessment may outweigh the incremental value of more complex algorithms. Finally, there is the potential for unmeasured confounding in this study, and future studies may identify other patient factors more strongly associated with prognosis than patients’ health status.

When routinely monitoring symptoms, function, and quality of life in patients with HF, serially collected health status data can provide updated risk estimates for cardiovascular death or HF hospitalization as well as death from any cause. These data support the longitudinal assessment of health status for accurate, contemporary risk estimation.

Corresponding Author: John A. Spertus, MD, MPH, Saint Luke’s Mid America Heart Institute, 4401 Wornall Rd, Cardiovascular Research, 9th Floor, Kansas City, MO 64111 (spertusj@umkc.edu).

Accepted for Publication: September 12, 2017.

Correction: This article was corrected on December 20, 2017, to correct an error in the Abstract. The first sentence of the Results section of the Abstract should be replaced with 2 sentences that read “Of 1767 eligible TOPCAT participants, 882 were women (49.9%), and the mean (SD) age was 71.5 (9.7) years. Of 2130 eligible HF-ACTION participants, 599 were women (28.1%), and the mean age was 58.6 (12.7) years.”

Published Online: November 1, 2017. doi:10.1001/jamacardio.2017.3983

Author Contributions: Drs Pokharel and Tang had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Pokharel, Khariton, Tang, Nassif, Spertus.

Acquisition, analysis, or interpretation of data: Pokharel, Tang, Nassif, Chan, Arnold, Jones, Spertus.

Drafting of the manuscript: Pokharel.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Tang, Jones.

Obtained funding: Spertus.

Administrative, technical, or material support: Pokharel, Khariton, Spertus.

Supervision: Pokharel, Nassif, Spertus.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Spertus owns the copyright to KCCQ and has provided consultative services to Novartis, Bayer, Cytokinetics, and United Healthcare. No other disclosures were reported.

Funding/Support: Both TOPCAT and HF-ACTION trials were supported by the National Heart, Lung, and Blood Institute. Drs Pokharel, Khariton, and Nassif are supported by the National Heart, Lung, and Blood Institute of the National Institutes of Health under award T32HL110837. Dr Chan is supported by grant R01HL123980 from the National Heart, Lung, and Blood Institute.

Role of the Funder/Sponsor: The sponsor had no role in the design and conduct of the study; collection, management, analysis and interpretation of the data; preparation; review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: All information and materials in the article are original. The views expressed in this article are those of the authors and do not necessarily represent the views of the National Heart, Lung, and Blood Institute.

Meeting Presentation: Presented as abstracts at the 2017 American College of Cardiology Annual Scientific Sessions; March 19, 2017; Washington, DC.