Patients report more severe daily limitations than recognized by their physicians

Abstract Background Patient limitations guide selection of heart failure therapies, for which indications often specify New York Heart Association Class. Objectives To determine the extent of patient‐reported limitations during daily activities and compare to New York Heart Association class assigned by providers during the same visit, and to left ventricular ejection fraction (LVEF) group. Methods and Results While waiting for their appointment, 948 patients on return visits to an ambulatory HF clinic completed a written questionnaire assessing specific activity limitations, which were compared to physician‐assigned NYHA class during the same visit. Patient‐reported limitation to perform daily activity ranged from 25% for bathing to 61% for yardwork or housework and 71% for jogging or hurrying. Most patients who did not report limitations to perform daily life activities were correctly classified as NYHA I by the physicians (76%), but 12% of the 376 patients classified as NYHA I reported limitations to showering or bathing and 73% reported limitations while doing yardwork or house work. Limitation to walking was reported by 172 patients (50%) classified as class II. Limitations to walking one block were most common in patients with LVEF ≥40% compared to patients with LVEF <40%, and least commonly, in HF with better EF (improved from 31 ± 13 to 52 ± 7). Conclusions Activity limitations are commonly reported by ambulatory HF patients, but underestimated by physicians. It is not clear how this should guide therapy validated for NYHA class but focused activity questions may merit wider use to track limitations and improvement in ambulatory HF.


| INTRODUCTION
Increasing penetration of recommended therapies for heart failure (HF) has improved outcomes, 1 but morbidity and mortality remain substantial. 2,3 Heart failure guidelines mandate consideration not only for prolonging survival but also on improving quality of life (QOL), 4,5 upon which patients may place similar or greater value. 6 A major contributor to QOL in HF is functional capacity, for which impairment can lead to loss of independence and reduced selfesteem. 7,8 The New York Heart Association (NYHA) functional classification system 9 is widely used in clinical practice, in clinical trials, 10 in translation of trials into guideline recommendations for therapies, 11 and in decisions about advanced HF therapies. 12,13 Understanding functional status and its improvement may be further enhanced by consideration of patient reported limitations to specific activities during daily life, as assessed by some of the questions in the Kansas City Cardiomyopathy Questionnaire. 14,15 Although NYHA classification has been revised several times since 1928, 9 it still relies on physicians' subjective estimate of a patient's ability to carry out "ordinary" activity. Consequently, the decision of a physician to classify a patient per NYHA functional class will depend on both patient and physician interpretation of "ordinary physical activity" and how to grade limitations as slight or marked. 16,17 Although NYHA class remains a robust discriminator of clinical outcome, reproducibility is low when the same patients are evaluated by different physicians. 14,17,18 In order to focus on patient-centered outcomes, 4 it would be potentially useful both in heart failure clinics and in clinical trials to consider patient's individual perceptions about limitations, adding information to frequently used NYHA classification. Although the nature of heart failure symptoms may be similar across ejection fraction (EF) groups, there are differences in demographics and comorbidities, such that the nature and degree of limitations may vary. 19 We hypothesized that patients may report limitations to specific daily activity that is not well-reflected in physician-assigned NYHA classification. The major aim of the present study was to determine the extent of patient-reported limitations during daily activities and to compare with the physician assignment of NYHA class on the same day. A further aim was to compare the frequency of perceived limitations across the HF different left ventricular ejection fraction (LVEF) types.

| Patient population and protocol
The present study is an observational cross-sectional analysis of patients seen in the ambulatory HF clinic at Brigham and Women's Hospital. As part of a quality improvement initiative to improve communication about quality of life, patients were asked to complete a questionnaire, while waiting for routine clinical visit. The questionnaire was handed to the clinic physician beneath the standard registration/vital sign form as the visit began. The clinic visit and the associated clinical note were completed directly by attending physicians, who had finished formal training in heart failure. There were no specific instructions regarding how to review or incorporate the information in the clinical assessment or documentation.
Patients who did not provide information about limitations to perform daily life activities, those with unknown LVEF and those whose physicians did not assign a NYHA class in the chart were excluded from the present analysis.

| Patient questionnaire
The two-sided, single page, self-administered questionnaire included questions about limitations to perform daily life activities, quality of life, symptoms, and patients' perception about clinical stability. To assess self-perceived limitation to perform daily life activities, the questionnaire included the following question: "Please indicate how much you are limited by your heart failure (shortness of breath or fatigue) in your ability to do the following over the last 4 weeks?". The activities listed included showering/bathing; walking one city block on level ground; yardwork, housework, or carrying groceries; climbing a flight of stairs (10 steps) without stopping; and hurrying or jogging (such as to catch a bus). 15 The degree of each limitation was classified as not at all limited; slightly limited; moderately limited; quite a bit limited; or extremely limited. Additional questions related to recent hospitalizations, ICD shocks, orthopnea, and overall quality of life. The Institutional Review Board of Brigham and Women's Hospital approved this retrospective observational study. No consents were obtained for the questionnaires, which were administered to patients as part of a clinical care and quality improvement initiative.

| Statistical analysis
Initially, normality of distribution was tested, validating the use of parametric statistics. Continuous variables are presented as mean ± SD and categorical variables are presented as frequencies and percentages. Multiple-group comparisons were performed using analysis of variance or chi-square tests as appropriate. Post-hoc correction for multiple comparisons between groups was performed using the Bonferroni method.
To compare patient-perceived daily life activity limitations across LVEF-HF types (HFrEF, HFpEF, and HFbEF), we performed multivariate logistic regression adjusted by age, body mass index, and comorbidities.
All statistical tests were two-sided, and P values <.05 were considered to indicate statistical significance. Statistical analysis was performed using Stata version 14 (Stata Corp, College Station, TX). Patients classified as NYHA I by the HF physicians were younger, had lower body mass index, and presented less comorbidities as diabetes and chronic kidney disease than the ones classified as NYHA II or III (Table 1). The majority of patients (75%) reported limitations to daily life activities, most commonly to hurrying or catching a bus (71%). Limitation to walking a block was reported by 45%, and 25% perceived some limitation even to bathing.

| RESULTS
While the frequency and degree of limitations were reported as higher by patients with higher NYHA designation (Figure 2), underclassification of NYHA by physicians was common.
Almost 12% of patients classified as NYHA I by HF physicians reported limitations to showering and bathing, and 20% of them presented limitations to walk one block on level ground. Among patients designated as class II, 52% described limitation, while walking a block and 73% perceived limitations, while doing housework or yardwork.
When patients were distributed accordingly to LVEF group (Table 2), most of them were classified as HFrEF (n = 400, LVEF: 26 ± 7%), followed by HFbetterEF EF (n = 382, LVEF: 52 ± 7%) and HFpEF (n = 166, LVEF: 60 ± 7%). Patients with HFbetterEF reported the fewest limitations to daily life activities after adjusting for age, gender, and comorbidities ( Figure 3). Although patients with HFpEF presented higher frequency of limitations when compared to patients with HFrEF in the univariate analysis, the significance of this comparison disappeared in the regression model adjusted for demographic characteristics and comorbidities.

| DISCUSSION
Ambulatory patients seen for routine visits in HF clinic reported substantial limitations to daily activities that were often not reflected in the NYHA class as assessed by experienced heart failure clinicians.
Patients with HFpEF and HFrEF were more limited in their activities than patients whose previously low-ejection fraction had improved (HFbetterEF).

| Symptom burden in patients with heart failure
The penetrance of guideline-recommended therapies was high in the present cohort, although the timing of the study preceded the introduction of newer therapies that may further improve functional classification. 11,[22][23][24] The burden of activity limitation was higher than expected for an ambulatory middle-aged HF population, with more than half of patients reporting limitations to climbing one flight of stairs, 30% of patients reporting limitations to dressing without stopping, and 25% presenting limitations to bathing. Skalska et al 25 reported a high dependence on assistance for bathing (28%) in a HF cohort, but their patients were much older (mean age 80 years) than our population with mean age of 57 years.

| Disparity between patient-reported limitations and assigned NYHA class
NYHA classification has been widely used in clinical trials as both an enrollment criterion and as an outcome measurement. It has consistently provided reliable discrimination for risk of hospitalizations and death. 14 Our study confirms that higher NYHA classification discriminates between patients with greater vs fewer limitations to routine activities. However, NYHA classification was not well calibrated to patient-reported activity limitations, as 10% patients classified as NYHA I described limitations performing low-intensity activities as showering and dressing, while most class II patients reported limitations when climbing one flight of stairs and during housework or yardwork. Alternatively, 5% of patients classified as NYHA III perceived no limitation to hurrying and jogging, which are moderate to high-intensity activities. As patients tend to avoid activities that are Discrepancies in classifying patients as NYHA II or III has been previously described, not only for differing perceptions between physicians regarding the same patient, 14 but also by comparisons of NYHA classification with a more objective measurement as "how far a patient can walk". 16

| Perceived limitations and LVEF subtype
In our study, patients with HFpEF and HFrEF perceived similar limitations to perform daily life activities after adjusting for age, gender, and comorbidities. Depending on the details of the populations studied, QOL impairment has been shown to be similar or slightly different between HFrEF and HFpEF groups. 19,27 This may reflect the characteristic pathophysiology of each of these LVEF HF subtypes, [28][29][30][31][32][33][34][35] and the fact that HFpEF in general encompasses older patients in whom age and accumulating comorbidities play an increasing role in limitation. 36 In our study, the difference between limitations reported by patients with HFrEF and HFpEF disappeared when the regression model was adjusted by age, body mass index, and comorbidities.
study. 21,37 The associated perception of better quality of life and less exercise limitation may reflect better circulatory reserve and also relative framing where they compare their current state to previously more severe limitations. Although patients with HFbetterEF have better survival 38   F I G U R E 3 Percentage of patients reporting limitations to perform daily life activities accordingly to left ventricular ejection fraction subtype. *P < .01 vs HFpEF & HFrEF. P values from multivariate logistic regression adjusted by age, body mass index, and comorbidities. HFrEF, heart failure with reduced ejection fraction (n = 400); HFbetterEF, heart failure with better ejection fraction (n = 382); HFpEF, heart failure with preserved ejection fraction (n = 166)