- Email: [email protected]
Socioeconomic and partner status in chronic heart failure: Relationship to exercise capacity, quality of life, and clinical outcomes
Clinical Investigation
Socioeconomic and partner status in chronic heart failure: Relationship to exercise capacity, quality of life, and clinical outcomes Amanda K. Verma, MD, a Phillip J. Schulte, PhD, b Vera Bittner, MD, MSPH, c Steven J. Keteyian, PhD, d Jerome L. Fleg, MD, e Ileana L. Piña, MD, MPH, f Ann M. Swank, PhD, g Meredith Fitz-Gerald, RN, BSN, c Stephen J. Ellis, PhD, h William E. Kraus, MD, h David J. Whellan, MD, MHS, i Christopher M. O'Connor, MD, h and Robert J. Mentz, MD h St Louis, MO; Rochester, MN; Birmingham, AL; Detroit, MI; Bethesda, MD; New York, NY; Durham, NC; and Philadelphia, PA
Background Prognosis in patients with heart failure (HF) is commonly assessed based on clinical characteristics. The association between partner status and socioeconomic status (SES) and outcomes in chronic HF requires further study. Methods We performed a post hoc analysis of HF-ACTION, which randomized 2,331 HF patients with ejection fraction ≤35% to usual care ± aerobic exercise training. We examined baseline quality of life and functional capacity and outcomes (all-cause mortality/hospitalization) by partner status and SES using adjusted Cox models and explored an interaction with exercise training. Outcomes were examined based on partner status, education level, annual income, and employment. Results Having a partner, education beyond high school, an income N$25,000, and being employed were associated with better baseline functional capacity and quality of life. Over a median follow-up of 2.5 years, higher education, higher income, being employed, and having a partner were associated with lower all-cause mortality/hospitalization. After multivariable adjustment, lower mortality was seen associated with having a partner (hazard ratio 0.91, 95% CI 0.81-1.03, P = .15) and more than a high school education (hazard ratio 0.91, CI 0.80-1.02, P = .12), although these associations were not statistically significant. There was no interaction between any of these variables and exercise training on outcomes (all P N .5). Conclusions Having a partner and higher SES were associated with greater functional capacity and quality of life at baseline but were not independent predictors of long-term clinical outcomes in patients with chronic HF. These findings provide information that may be considered as potential variables impacting outcomes. (Am Heart J 2017;183:54-61.)
Heart failure (HF) is a complex medical condition requiring multiple medications and lifestyle modifications to manage. Socioeconomic status (SES) and social support are thought to affect medical management and outcomes in HF. 1,2 For instance, HF patients with partners have higher medication adherence and event-free survival. 3 Further-
From the aWashington University School of Medicine, St Louis, MO, bMayo Clinic, Rochester, MN, cUniversity of Alabama, Birmingham, AL, dHenry Ford Hospital, Detroit, MI, eNational Heart, Lung, and Blood Institute, Bethesda, MD, fMontefiore-Einstein Medical Center, New York, NY, gUniversity of Louisville, Louisville, KY, hDuke University Medical Center, Durham, NC, and iJefferson Medical College, Philadelphia, PA. Funding: This research was supported by Grants HL063747 and HL093374 from the National Heart, Lung, and Blood Institute. RJM receives research support from the National Institutes of Health (U10HL110312). RCT No. NCT00047437 Submitted May 3, 2016; accepted October 6, 2016. Reprint requests: Amanda K. Verma, MD, Washington University School of Medicine, 660 South Euclid Ave, St Louis, MO 63110. E-mail: [email protected] 0002-8703 © 2016 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ahj.2016.10.007
more, individuals with HF living in socioeconomically disadvantaged areas have higher rates of hospitalization than do their counterparts living in less disadvantaged areas. 4 With a growing body of evidence regarding the impact of SES on health, the American Heart Association recently released a statement emphasizing the need to focus on social determinants of health in addition to traditional modifiable and nonmodifiable lifestyle, physiologic, and genetic risk factors when treating patients with heart disease. 5 Prior studies, however, have not systematically investigated partner status and SES in a population of patients with chronic HF receiving optimal medical therapy. Moreover, the data regarding SES and social support in HF are derived from small studies that did not examine the multiple components of SES and partner status together. In the HF-ACTION study, data on partner status and SES were obtained at baseline by patient report. We investigated the associations between baseline partner status/SES and outcomes in patients with chronic systolic HF enrolled in HF-ACTION.
American Heart Journal Volume 183
Methods HF-ACTION (clinicaltrials.gov, NCT00047437) was a randomized controlled clinical trial of aerobic exercise training in patients with chronic HF; the design, rationale, and primary results have been published. 6, 7 The study enrolled 2,331 ambulatory patients between April 2003 and February 2007 who had left ventricular ejection fraction (EF) ≤35% and New York Heart Association (NYHA) class II to IV HF while on optimal HF therapy for at least 6 weeks. Patients were randomized to either usual care or exercise training in addition to usual care and followed up for at least 12 months. Exercise training consisted of 36 supervised exercise sessions over the initial 3 months, followed by home training on a treadmill or stationary cycle for another 2 years. The protocol was approved by the institutional review board for each site involved in the study, and written informed consent was obtained for all patients. Partner status and SES were defined as reported in prior analyses from the HF-ACTION trial data set. 8 Partner status was ascertained by self-report of either having a partner (married or living with a partner) or no partner (single, never married, divorced, separated, or widowed). Socioeconomic status was measured by self-reported income (most recent annual household income before taxes; lower income defined as b$25,000 and higher income defined as ≥$25,000), employment status (employed defined as student, self-employed, part-time employed, or full-time employed; unemployed defined as homemaker, volunteer, retired, disabled, or unemployed), and education level (lower education defined as high school graduate/equivalent or less and higher education defined as greater than high school education, including some college, associate degree, college graduate, and/or graduate school degree). Functional capacity was assessed by cardiopulmonary exercise testing to evaluate peak oxygen consumption (peak VO2) as well as by 6-minute walk test to measure distance walked. Adherence to therapy was evaluated in patients randomized to exercise training by measuring total time of exercise performed during supervised training sessions and patient self-report of home training sessions. Quality of life was measured by the Kansas City Cardiomyopathy Questionnaire (KCCQ). The primary end point was all-cause mortality and hospitalization, with secondary end points including all-cause death and composite cardiovascular death or HF hospitalization. These outcomes were adjudicated by a committee blinded to treatment assignment.
Statistical analysis Patients were grouped per independent variable of interest in dichotomous categories (eg, partner or no partner) and baseline characteristics were described. Continuous variables were reported as the median with 25th-75th percentiles and compared with the independent variable of interest using the Wilcoxon rank sum statistic.
Verma et al 55
Categorical variables were reported as percentages and compared with the independent variable of interest with Pearson χ 2 or exact test. The relationship of the independent variables of interest with outcomes was assessed using Cox proportional hazards ratios and was adjusted based on the variables consistently used in HF-ACTION post hoc analyses (see Table II footnote). 9 Regression models used complete case data. In addition, several additional variables were included that were thought to potentially confound outcomes associations: age, sex, race, history of depression, and Beck Depression Score. 10,11 Kaplan-Meier estimates were generated for the primary outcome. The relationship of independent variables was also investigated with regard to functional capacity and adherence to therapy. We also assessed for interaction of the independent variables and exercise treatment. A 2-tailed P value b.05 was considered statistically significant for all analyses. No adjustment was made for the assessment of multiple end points in the present article given the exploratory nature of the investigation. The SAS system, version 9.2, was used for analyses (SAS, Cary, NC), all of which were conducted by Duke Clinical Research Institute.
Results The baseline characteristics of patients based on partner status and SES are described in Table I. Of the patients included in the study, 61% reported having a partner, 60% had more than a high school education, 59% had an income ≥$25,000, and 24% were employed. Patients with a current partner had higher income and tended to be older, male, and white. Patients with greater than a high school education included a larger proportion of women and white. Higher-income patients tended to be older, male, and white. Employment was more common in younger individuals. There were no statistically significant differences in medical therapies based on partner status or SES, with the exception of β-blocker use being more prevalent in individuals who were employed. Internal cardioverter/ defibrillator use was significantly more prevalent among those who had a partner, had greater than a high school education, had higher income, and were unemployed. Comorbidities such as diabetes and hypertension were more common in individuals with less education, lower income, and unemployed status. The NYHA class was also worse in patients with less education, lower income, and unemployed status. Table II describes baseline functional capacity and quality of life of patients based on partner status and SES. There were statistically significant differences in all groups with regard to exercise tolerance (measured by peak VO2 and 6-minute walk test) and health-related quality of life (measured by the KCCQ), favoring having a partner, higher level of education, higher income, and being employed.
American Heart Journal January 2017
56 Verma et al
Table I. Baseline characteristics by partner status and SES Partner status No partner (n = 899)
Partner (n = 1411)
Education level High school or less (n = 920)
Patient characteristics Age (y) 58 (49-67)⁎ 61 (53-69)⁎ 60 (51-68) Female sex 40.9%⁎ 20.3%⁎ 20.3%⁎ Race Black 45.7%⁎ 24.0%⁎ 39.2%⁎ White 49.8%⁎ 70.2%⁎ 56.4%⁎ Other 4.5%⁎ 5.8%⁎ 4.4%⁎ BMI (kg/m2) 30 (26-36)⁎ 30 (26-34)⁎ 30 (26-35) Medical history Hypertension 61.3% 58.7% 63.7%⁎ Diabetes 33.4% 31.2% 33.9% Depression 22.2% 20.2% 19.3% Laboratory results Na (mmol/L) 139 (137-141) 139 (137-141) 139 (137-141) Creatinine (mg/dL) 1.2 (1.0-1.5) 1.2 (1.0-1.5) 1.2 (1.0-1.5) Pro-BNP (pg/mL) 735 857 789 (321-1766) (346-1878) (373-1750) Medications and devices β-Blocker 95.2% 94.2% 94.2% ACE inhibitor or 94.2% 94.3% 94.2% ARB MRA 45.7% 44.7% 43.8% ICD 33.9%⁎ 44.3%⁎ 36.6%⁎ HF history HF hospitalizations prior 6 mo None 69.2%⁎ 76.3%⁎ 73.7% 1 21.9%⁎ 19.0%⁎ 20.7% 2 5.6%⁎ 3.1%⁎ 3.6% ≥3 3.4%⁎ 1.6%⁎ 2.0% Left ventricular EF 25 (20-30) 25 (25-30) 25 (20-30) NYHA III-IV vs II 38.8% 35.1% 39.6%⁎
Income level
Employment
More than high school (n = 1359)
Lower income (n = 853)
Higher income (n = 1220)
Unemployed (n = 1731)
Employed (n = 551)
59 (51-68) 29.0%⁎
58 (50-68)⁎ 33.6%⁎
60 (52-68)⁎ 24.3%⁎
61 (53-70)⁎ 27.9%
55 (47-61)⁎ 29.8%
28.4%⁎ 65.7%⁎ 5.9%⁎
43.2%⁎ 51.1%⁎ 5.7%⁎
24.1%⁎ 71.0%⁎ 4.9%⁎
29.8% 65.0% 5.2% 30 (26-35) 53.0%⁎ 25.0%⁎ 19.1%
30 (26-35)
31 (26-36)⁎
30 (26-35)⁎
33.9% 60.8% 5.3% 30 (26-35)
57.7%⁎ 31.0% 22.4%
65.4%⁎ 37.4%⁎ 22.7%
57.3%⁎ 28.9%⁎ 20.2%
62.2%⁎ 34.5%⁎ 21.7%
139 (137-141) 1.2 (1.0-1.5) 831 (331-1812)
139 (137-141) 139 (137-141) 139 (137-141) 1.2 (1.0-1.5) 1.2 (1.0-1.5) 1.2 (1.0-1.5)⁎ 857 819 868 (375-1793) (323-1821) (376-1992)⁎
139 (138-141) 1.1 (0.9-1.3)⁎ 612 (255-1411)⁎
94.6% 94.6%
93.8% 93.9%
95.4% 95.0%
93.9%⁎ 94.0%
96.4%⁎ 95.5%
45.8% 41.9%⁎
44.1% 36.0%⁎
45.2% 43.4%⁎
45.2% 42.6%⁎
45.4% 32.1%⁎
73.5% 19.8% 4.2% 2.5% 25 (20-30) 34.7%⁎
72.1% 20.9% 4.0% 3.0% 25 (20-30) 43.11%⁎
74.6% 19.5% 4.1% 1.7% 25 (20-30) 32.8%⁎
74.8%⁎ 19.0%⁎ 3.6%⁎ 2.6%⁎
69.2%⁎ 23.8%⁎ 5.7%⁎ 1.3%⁎
25 (20-30) 40.8%⁎
25 (20-30) 21.8%⁎
Continuous variables are shown as median (25th-75th percentile) and categorical variables are shown as % (n). Abbreviations: BMI, body mass index; BNP, brain natriuretic peptide; ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blocker; MRA, mineralocorticoid receptor antagonist; ICD, implantable cardioverter defibrillator. ⁎ P b .05 for comparison.
Adherence to exercise was measured in patients randomized to the exercise training arm of the HF-ACTION study. Table III shows that patients with a partner and who had higher income had greater supervised exercise and total exercise times compared with their counterparts without a partner and with lower income, respectively. The associations between partner status and SES and primary outcomes are shown in Table IV. In the unadjusted model, having a partner, higher education, higher income, and being employed were all significantly associated with lower all-cause mortality/hospitalization and cardiovascular death/HF hospitalization. Figure presents the Kaplan-Meier curve for all-cause mortality/ hospitalization by partner status and education. After risk adjustment, all-cause mortality/hospitalization and cardiovascular death/HF hospitalization were non-
significantly lower in patients with a partner and higher education (all-cause mortality/hospitalization: partner-adjusted hazard ratio [HR] 0.91, 95% CI 0.81-1.03; higher education–adjusted HR 0.91, 95% CI 0.80-1.02; cardiovascular death/HF hospitalization: partner-adjusted HR 0.86, 95% CI 0.72-1.03; higher education–adjusted HR 0.86, 95% CI 0.73-1.02). There was no interaction between any of the partner status or SES variables and exercise training on outcomes (all P N .5).
Discussion Our data describe the associations of partner status and SES with exercise capacity, quality of life, and clinical outcomes in patients with chronic systolic HF. Those with a higher SES and a partner had better baseline exercise tolerance and health-related quality of life,
American Heart Journal Volume 183
Verma et al 57
Table II. Baseline exercise capacity and quality of life by partner status and SES Partner status
No partner (n = 899)
Partner (n = 1411)
Education level High school or less (n = 920)
Income level
More than high school (n = 1359)
Lower income (b$25,000) (n = 853)
Higher income (≥$25,000) (n = 1220)
Employment
Unemployed (n = 1731)
Employed (n = 551)
Peak VO2 13.8 (11.2-17.0)⁎ 14.9 (11.8-18.0)⁎ 13.7 (10.9-16.7)⁎ 15.1 (12.0-18.0)⁎ 13.3 (11.0-16.7)⁎ 15.2 (12.3-18.1)⁎ 13.7 (11.1-17.0)⁎ 16.5 (13.7-19.7)⁎ (mL kg−1 min−1) 6-min walk 357 (290-420)⁎ 381 (305-444)⁎ 355 (287-420)⁎ 382 (306-442)⁎ 350 (276-414)⁎ 388 (316-445)⁎ 358 (286-421)⁎ 415 (350-474)⁎ distance (m) KCCQ overall 66 (49-82)⁎ 69 (52-84)⁎ 66 (49-83)⁎ 69 (53-84)⁎ 63 (48-80)⁎ 71 (55-85)⁎ 65 (48-82)⁎ 76 (63-88)⁎ summary score Continuous variables are shown as median (25th-75th percentile) and categorical variables are shown as % (n). Partner reference group is no partner, higher education reference group is lower education (high school or less), higher income reference group is lower income (b$25,000), and employment reference group is unemployed. ⁎ P b .05.
Table III. Adherence to exercise by partner status and SES Partner status No partner (n = 899)
Partner (n = 1411)
Education level High school or less (n = 920)
Income level
More than high school (n = 1359)
Lower income (b$25,000) (n = 853)
Higher income (≥$25,000) (n = 1220)
Employment Unemployed (n = 1731)
Employed (n = 551)
Supervised exercise 560 (220–810)⁎ 675 (443-846)⁎ 629 (296-830) 640 (386-838) 546 (189-780)⁎ 670 (454-856)⁎ 640 (360-837) 629 (370-838) (min) Total exercise (min) 63 (24-103)⁎ 85 (49-126)⁎ 75 (32-117) 79 (44-118) 65 (19-102)⁎ 85 (51-123)⁎ 80 (38-118) 70 (40-118) Continuous variables are shown as median (25th-75th percentile). Partner reference group is no partner, higher education reference group is lower education (high school or less), higher income reference group is lower income (b$25,000), and employment reference group is unemployed. ⁎ P b .05.
whereas having a partner and higher income were associated with better adherence to exercise therapy. Partner status and SES were not significantly associated with the primary outcome, although there was a trend for better outcomes among individuals with a partner and with greater than a higher school education. Socioeconomic status encompasses education, income, and employment, which are resources that help support an individual's ability to function and manage the demands of life. Furthermore, partner status is a marker of significant social support. Previous analyses have demonstrated a correlation between SES and life expectancy. 12-15 In the present article, we extend these results by demonstrating in patients with chronic HF the associations of partner status and SES with mortality and hospitalization as well as with functional capacity, health-related quality of life, and adherence to prescribed therapy. Quality of life and functional capacity are the “visible” parameters that patients with HF experience on a day-to-day basis and how many providers, at least in part, gage the status of the underlying illness. Indeed, these variables are also incorporated into both the patient and physician decision making to help guide therapy. 16 Furthermore, the multifaceted nature of quality of life and functional capacity is difficult to fully characterize
because it differs among individuals and is not easily predicted. For instance, some may find a costly HF medication that improves dyspnea to improve quality of life, whereas others may find that despite improved symptom burden, the cost of the medication negatively impacts their definition of quality of life. 17 Prior studies have shown that lower SES is related to exercise intolerance in patients with congenital heart disease and to impaired exercise capacity in patients with coronary heart disease. 18-20 Indeed, in a recent HF-ACTION analysis, lower perceived social support was associated with decreased exercise time. 2 However, functional capacity in relation to individual SES and partner status has not previously been studied in patients with HF. Here, we found that having a partner, more than a high school education, higher income, and being employed were all associated with improved functional capacity and quality of life. Partner status has been shown to be associated with outcomes in various cardiovascular disease states. Studies have shown an association between being married and reduced all-cause cardiovascular mortality in patients with coronary artery disease, post–myocardial infarction, HF, and after heart transplantation. 21-28 These observations may be mediated by risk factor modification related
American Heart Journal January 2017
58 Verma et al
Table IV. Clinical outcomes by partner status and SES Partner HR (95% CI)
Higher education P
HR (95% CI)
Higher income P
HR (95% CI)
Employed P
HR (95% CI)
P
All-cause mortality or hospitalization Unadjusted 0.90 (0.81-1.00) .042 0.86 (0.78-0.95) .004 0.87 (0.78-0.97) .011 0.71 (0.63-0.80) b.001 Adjusted⁎ 0.91 (0.81-1.03) .15 0.91 (0.80-1.02) .12 0.97 (0.85-1.10) .64 0.97 (0.83-1.12) .67 All-cause mortality Unadjusted 0.89 (0.73-1.10) .28 0.73 (0.60-0.90) .002 0.83 (0.68-1.03) .10 0.53 (0.40-0.70) b.001 Adjusted† 1.07 (0.84-1.35) .58 0.84 (0.67-1.05) .13 1.12 (0.88-1.43) .35 1.05 (0.76-1.45) .77 Cardiovascular death or HF hospitalization Unadjusted 0.77 (0.66-0.89) b.001 0.75 (0.65-0.87) b.001 0.77 (0.66-0.90) b.001 0.50 (0.40-0.61) b.001 Adjusted‡ 0.86 (0.72-1.03) .11 0.86 (0.73-1.02) .08 0.94 (0.79-1.14) .56 0.88 (0.69-1.10) .256 Hazard ratios are reported with associated 95% CI. Partner reference group is no partner, higher education reference group is lower education (high school or less), higher income reference group is lower income (b$25,000), and employment reference group is unemployed. ⁎ Adjusted for Weber class, Kansas City Cardiomyopathy Questionnaire (KCCQ) symptom stability score, blood urea nitrogen, country, left ventricular EF, sex, age, race, history of depression, Beck Depression Score at baseline, β-blocker dosage, mitral regurgitation grade, and ventricular conduction before baseline cardiopulmonary exercise test. † Adjusted for exercise duration on baseline cardiopulmonary exercise test, serum creatinine level, body mass index (BMI), sex, age, race, history of depression, Beck Depression Score at baseline, loop diuretic dose, left ventricular EF, Canadian Cardiovascular Society angina classification, and ventricular conduction before baseline cardiopulmonary exercise test. ‡ Adjusted for loop diuretic dose, left ventricular EF, mitral regurgitation grade, ventricular conduction before baseline cardiopulmonary exercise test, KCCQ symptom stability score, blood urea nitrogen, race, sex, age, Weber class, history of depression, Beck Depression Score at baseline, and VE/VCO2 on baseline cardiopulmonary exercise test.
to “benefits from spousal support with regard to seeking treatment, adherence to treatment, and recommended lifestyle changes, as well as greater financial resources, which make medical treatment and healthy lifestyle choices affordable.” 22 Importantly, though, the implications of marital status likely also depend on quality of the relationship as has been demonstrated previously. 29,30 There are only a few, small studies that specifically examined the association of partner status in patients with HF, with most showing longer event-free survival and decreased rehospitalization in patients with a partner. 22-28, 31, 32 Our study further adds to these findings in a large HF cohort in addition to measuring other socioeconomic factors concurrently. There was an insignificant trend for individuals with a partner to have better outcomes. The partner relationship is complex and is thought to mediate its effects via improvement in medication adherence and decrease in depressive symptoms. 3, 28 Indeed, we found that patients with a partner had better adherence to exercise therapy compared with those without a partner. Interestingly, our study also found that partner status was associated with a statistically significant 12% reduction in the primary outcome of all-cause mortality or hospitalization only when excluding depression from the adjusted model, thus supporting the concept that having a partner is potentially mediated in part by affecting mood. We performed additional analyses to assess whether prior exposure to partnership (ie, patients who were widowed, divorced, or separated) was associated with outcomes to understand if the experience of a past or current relationship may in itself be important; however, there was no significant difference among patients with no partner, a prior partner, or a current partner. Although our study examined partners
who lived with the patient, future studies could also examine whether this relationship holds true when examining a patient's social network outside an immediate partner. We also observed an insignificant association between more than a high school education and reduced all-cause mortality/hospitalization and cardiovascular death or HF hospitalization. Prior studies of education have generally found similar associations of reduction in mortality and rehospitalization or equivocal associations with outcomes in patients with HF; however, these studies have had small sample sizes and some have used regional data rather than patient-reported data. 4, 33-35 Although employment status was not significantly associated with outcomes, it was associated with higher functional capacity and quality of life. One could posit that the working individuals are able to do so because they have a better functional capacity and quality of life, thus permitting them to do so. Indeed, employed individuals were younger, with less comorbidity, a larger proportion in NYHA classes II to III vs IV as compared with individuals who were not employed. Different types of employment such as manual labor vs sedentary work may also confer different associations with outcomes. Thus, future studies are needed to better define this relationship. Several studies have examined SES with regard to outcomes in patients with HF and have shown that lower income level is associated with an increase in hospitalization and mortality in patients with HF. 4, 35-39 Our study did not find a clear relationship between income level and clinical outcomes. Several key differences in these studies with regard to income analysis may account for the discrepancy of our findings. For instance, several of these studies assessed income indirectly by evaluating
American Heart Journal Volume 183
Figure
A
B
Verma et al 59
outcomes that many of these studies propose. As we have shown, perhaps by not measuring income by accessibility to medication and care, low income may not independently portend worse outcomes. Although our findings show an association of SES with outcomes in HF, the implication of these findings is not straightforward. It is unclear if interventions could impact outcomes as this study shows the association of socioeconomic and partner status on outcomes rather than causality of variables. Moreover, potential interventions to influence socioeconomic and partner status are complex and cannot be readily implemented. Socioeconomic status and partner status should be acknowledged as factors that could provide useful information in understanding and perhaps risk stratifying individuals with regard to outcomes. In addition, these findings also pose interesting avenues of exploration to try to understand possible causality of the complex relationship of SES and partner status to then allow realistic interventions to be tested.
Limitations This was a retrospective analysis of a clinical trial, and therefore, causal relationships cannot be determined. Furthermore, there were strict inclusion/exclusion criteria in the context of the trial such that the findings may not apply to the broader HF population. The definitions of partner status and SES may vary based on culture, geographic location, and over time. Thus, these findings may potentially be limited by the definitions imposed here. Information regarding partner status and SES was obtained by self-report and may be subject to biased reporting.
Conclusions
Unadjusted all-cause mortality or hospitalization by education (A) and partner status (B). Kaplan-Meier estimates of all-cause mortality of hospitalization by education and partner status through 4 years of follow-up.
mean or median income in the geographic location where a patient lived or asking the patient whether he/she was financially “comfortable.” 4, 36, 40 Geographic area may better correlate with access to care rather than the income itself, and subjective ideas of financial stability can be a proxy for other factors that do not just include finances. Furthermore, with the high percentage of patients on appropriate HF medications at baseline in HF-ACTION, it does not seem that income was an overt barrier to getting medical therapies, which is the theorized reason for worse
In patients with chronic HF, having a partner, being educated beyond high school, having an income N$25,000, and being employed were associated with improved baseline functional capacity and quality of life. Having a partner and more than a high school education may also be associated with lower rates of all-cause mortality and hospitalization; however, further studies will be needed to confirm these associations. These data suggest the potential importance of life partnership and SES with regard to long-term outcomes and in maintaining functional status and quality of life. Future studies of social support interventions and SES-tailored care would be useful to determine if these factors may be helpful in the management of HF.
Disclosures All authors have reported that they have no relationships relevant to the contents of this manuscript to report. All authors were responsible for the design, analyses, drafting, and editing of the presented research.
60 Verma et al
References 1. Fiscella K, Tancredi D. Socioeconomic status and coronary heart disease risk prediction. JAMA 2008;300(22):2666-8. 2. Cooper LB, Mentz RJ, Sun JL, et al. Psychosocial factors, exercise adherence, and outcomes in heart failure patients: insights from Heart Failure: A Controlled Trial Investigating Outcomes of Exercise Training (HF-ACTION). Circ Heart Fail 2015;8(6):1044-51. 3. Wu JR, Lennie TA, Chung ML, et al. Medication adherence mediates the relationship between marital status and cardiac event-free survival in patients with heart failure. Heart Lung 2012;41(2):107-14. 4. Eapen ZJ, McCoy LA, Fonarow GC, et al. Utility of socioeconomic status in predicting 30-day outcomes after heart failure hospitalization. Circ Heart Fail 2015. 5. Havranek EP, Mujahid MS, Barr DA, et al. Social determinants of risk and outcomes for cardiovascular disease: a scientific statement from the American Heart Association. Circulation 2015;132(9): 873-98. 6. O'Connor CM, Whellan DJ, Lee KL, et al. Efficacy and safety of exercise training in patients with chronic heart failure: HF-ACTION randomized controlled trial. JAMA 2009;301(14):1439-50. 7. Whellan DJ, O'Connor CM, Lee KL, et al. Heart Failure and A Controlled Trial Investigating Outcomes of Exercise Training (HF-ACTION): design and rationale. Am Heart J 2007;153(2):201-11. 8. Mentz RJ, Babyak MA, Bittner V, et al. Prognostic significance of depression in blacks with heart failure: insights from Heart Failure: A Controlled Trial Investigating Outcomes of Exercise Training. Circ Heart Fail 2015;8(3):497-503. 9. Mentz RJ, Bittner V, Schulte PJ, et al. Race, exercise training, and outcomes in chronic heart failure: findings from Heart Failure—A Controlled Trial Investigating Outcomes in Exercise TraiNing (HF-ACTION). Am Heart J 2013;166(3):488-95. 10. O'Connor CM, Whellan DJ, Wojdyla D, et al. Factors related to morbidity and mortality in patients with chronic heart failure with systolic dysfunction: the HF-ACTION predictive risk score model. Circ Heart Fail 2012;5(1):63-71. 11. Mentz RJ, Schulte PJ, Fleg JL, et al. Clinical characteristics, response to exercise training, and outcomes in patients with heart failure and chronic obstructive pulmonary disease: findings from Heart Failure and A Controlled Trial Investigating Outcomes of Exercise TraiNing (HF-ACTION). Am Heart J 2013;165(2):193-9. 12. Kelishadi R, Poursafa P. A review on the genetic, environmental, and lifestyle aspects of the early-life origins of cardiovascular disease. Curr Probl Pediatr Adolesc Health Care 2014;44(3):54-72. 13. de Andrade LO, Pellegrini Filho A, Solar O, et al. Social determinants of health, universal health coverage, and sustainable development: case studies from Latin American countries. Lancet 2015;385(9975): 1343-51. 14. Braveman P, Gottlieb L. The social determinants of health: it's time to consider the causes of the causes. Public Health Rep 2014;129(Suppl 2): 19-31. 15. Rendall MS, Weden MM, Favreault MM, et al. The protective effect of marriage for survival: a review and update. Demography 2011;48(2): 481-506. 16. Olano-Lizarraga M, Oroviogoicoechea C, Errasti-Ibarrondo B, et al. The personal experience of living with chronic heart failure: a qualitative meta-synthesis of the literature. J Clin Nurs 2016. 17. Karimi M, Clark AM. How do patients' values influence heart failure self-care decision-making? A mixed-methods systematic review. Int J Nurs Stud 2016;59:89-104. 18. Shishehbor MH, Litaker D, Pothier CE, et al. Association of socioeconomic status with functional capacity, heart rate recovery, and all-cause mortality. JAMA 2006;295(7):784-92.
American Heart Journal January 2017
19. Diller GP, Inuzuka R, Kempny A, et al. Detrimental impact of socioeconomic status on exercise capacity in adults with congenital heart disease. Int J Cardiol 2013;165(1):80-6. 20. Cohen B, Vittinghoff E, Whooley M. Association of socioeconomic status and exercise capacity in adults with coronary heart disease (from the Heart and Soul Study). Am J Cardiol 2008;101(4): 462-6. 21. Molloy GJ, Stamatakis E, Randall G, et al. Marital status, gender and cardiovascular mortality: behavioural, psychological distress and metabolic explanations. Soc Sci Med 2009;69(2):223-8. 22. Quinones PA, Kirchberger I, Heier M, et al. Marital status shows a strong protective effect on long-term mortality among first acute myocardial infarction-survivors with diagnosed hyperlipidemia—findings from the MONICA/KORA myocardial infarction registry. BMC Public Health 2014;14:98. 23. Tam V, Arnaoutakis GJ, George TJ, et al. Marital status improves survival after orthotopic heart transplantation. J Heart Lung Transplant 2011;30(12):1389-94. 24. Watkins T, Mansi M, Thompson J, et al. Effect of marital status on clinical outcome of heart failure. J Investig Med 2013;61(5): 835-41. 25. Howie-Esquivel J, Spicer JG. Association of partner status and disposition with rehospitalization in heart failure patients. Am J Crit Care 2012;21(3):e65-73. 26. Calvillo-King L, Arnold D, Eubank KJ, et al. Impact of social factors on risk of readmission or mortality in pneumonia and heart failure: systematic review. J Gen Intern Med 2013;28(2):269-82. 27. Chung ML, Lennie TA, Riegel B, et al. Marital status as an independent predictor of event-free survival of patients with heart failure. Am J Crit Care 2009;18(6):562-70. 28. Staniute M, Brozaitiene J, Burkauskas J, et al. Type D personality, mental distress, social support, and health-related quality of life in coronary artery disease patients with heart failure: a longitudinal observational study. Health Qual Life Outcomes 2015;13(1):1. 29. Donoho CJ, Seeman TE, Sloan RP, et al. Marital status, marital quality, and heart rate variability in the MIDUS cohort. J Fam Psychol 2015;29(2):290-5. 30. Joseph NT, Kamarck TW, Muldoon MF, et al. Daily marital interaction quality and carotid artery intima-medial thickness in healthy middle-aged adults. Psychosom Med 2014;76(5):347-54. 31. Tsuchihashi M, Tsutsui H, Kodama K, et al. Medical and socioenvironmental predictors of hospital readmission in patients with congestive heart failure. Am Heart J 2001;142(4):E7. 32. Luttik ML, Jaarsma T, Veeger N, et al. Marital status, quality of life, and clinical outcome in patients with heart failure. Heart Lung 2006;35(1):3-8. 33. Roe-Prior P. Sociodemographic variables predicting poor post-discharge outcomes for hospitalized elders with heart failure. Medsurg Nurs 2007;16(5):317-21. 34. Schwarz KA, Elman CS. Identification of factors predictive of hospital readmissions for patients with heart failure. Heart Lung 2003;32(2): 88-99. 35. Damiani G, Salvatori E, Silvestrini G, et al. Influence of socioeconomic factors on hospital readmissions for heart failure and acute myocardial infarction in patients 65 years and older: evidence from a systematic review. Clin Interv Aging 2015;10:237-45. 36. Heo S, Moser DK, Chung ML, et al. Social status, health-related quality of life, and event-free survival in patients with heart failure. Eur J Cardiovasc Nurs 2012;11(2):141-9.
American Heart Journal Volume 183
37. Rathore SS, Masoudi FA, Wang Y, et al. Socioeconomic status, treatment, and outcomes among elderly patients hospitalized with heart failure: findings from the National Heart Failure Project. Am Heart J 2006;152(2):371-8. 38. Philbin EF, Dec GW, Jenkins PL, et al. Socioeconomic status as an independent risk factor for hospital readmission for heart failure. Am J Cardiol 2001;87(12):1367-71.
Verma et al 61
39. Hawkins NM, Jhund PS, McMurray JJ, et al. Heart failure and socioeconomic status: accumulating evidence of inequality. Eur J Heart Fail 2012;14(2):138-46. 40. Bikdeli B, Wayda B, Bao H, et al. Place of residence and outcomes of patients with heart failure: analysis from the telemonitoring to improve heart failure outcomes trial. Circ Cardiovasc Qual Outcomes 2014;7(5):749-56.
Socioeconomic and partner status in chronic heart failure: Relationship to exercise capacity, quality of life, and clinical outcomes Amanda K. Verma, MD, a Phillip J. Schulte, PhD, b Vera Bittner, MD, MSPH, c Steven J. Keteyian, PhD, d Jerome L. Fleg, MD, e Ileana L. Piña, MD, MPH, f Ann M. Swank, PhD, g Meredith Fitz-Gerald, RN, BSN, c Stephen J. Ellis, PhD, h William E. Kraus, MD, h David J. Whellan, MD, MHS, i Christopher M. O'Connor, MD, h and Robert J. Mentz, MD h St Louis, MO; Rochester, MN; Birmingham, AL; Detroit, MI; Bethesda, MD; New York, NY; Durham, NC; and Philadelphia, PA
Background Prognosis in patients with heart failure (HF) is commonly assessed based on clinical characteristics. The association between partner status and socioeconomic status (SES) and outcomes in chronic HF requires further study. Methods We performed a post hoc analysis of HF-ACTION, which randomized 2,331 HF patients with ejection fraction ≤35% to usual care ± aerobic exercise training. We examined baseline quality of life and functional capacity and outcomes (all-cause mortality/hospitalization) by partner status and SES using adjusted Cox models and explored an interaction with exercise training. Outcomes were examined based on partner status, education level, annual income, and employment. Results Having a partner, education beyond high school, an income N$25,000, and being employed were associated with better baseline functional capacity and quality of life. Over a median follow-up of 2.5 years, higher education, higher income, being employed, and having a partner were associated with lower all-cause mortality/hospitalization. After multivariable adjustment, lower mortality was seen associated with having a partner (hazard ratio 0.91, 95% CI 0.81-1.03, P = .15) and more than a high school education (hazard ratio 0.91, CI 0.80-1.02, P = .12), although these associations were not statistically significant. There was no interaction between any of these variables and exercise training on outcomes (all P N .5). Conclusions Having a partner and higher SES were associated with greater functional capacity and quality of life at baseline but were not independent predictors of long-term clinical outcomes in patients with chronic HF. These findings provide information that may be considered as potential variables impacting outcomes. (Am Heart J 2017;183:54-61.)
Heart failure (HF) is a complex medical condition requiring multiple medications and lifestyle modifications to manage. Socioeconomic status (SES) and social support are thought to affect medical management and outcomes in HF. 1,2 For instance, HF patients with partners have higher medication adherence and event-free survival. 3 Further-
From the aWashington University School of Medicine, St Louis, MO, bMayo Clinic, Rochester, MN, cUniversity of Alabama, Birmingham, AL, dHenry Ford Hospital, Detroit, MI, eNational Heart, Lung, and Blood Institute, Bethesda, MD, fMontefiore-Einstein Medical Center, New York, NY, gUniversity of Louisville, Louisville, KY, hDuke University Medical Center, Durham, NC, and iJefferson Medical College, Philadelphia, PA. Funding: This research was supported by Grants HL063747 and HL093374 from the National Heart, Lung, and Blood Institute. RJM receives research support from the National Institutes of Health (U10HL110312). RCT No. NCT00047437 Submitted May 3, 2016; accepted October 6, 2016. Reprint requests: Amanda K. Verma, MD, Washington University School of Medicine, 660 South Euclid Ave, St Louis, MO 63110. E-mail: [email protected] 0002-8703 © 2016 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ahj.2016.10.007
more, individuals with HF living in socioeconomically disadvantaged areas have higher rates of hospitalization than do their counterparts living in less disadvantaged areas. 4 With a growing body of evidence regarding the impact of SES on health, the American Heart Association recently released a statement emphasizing the need to focus on social determinants of health in addition to traditional modifiable and nonmodifiable lifestyle, physiologic, and genetic risk factors when treating patients with heart disease. 5 Prior studies, however, have not systematically investigated partner status and SES in a population of patients with chronic HF receiving optimal medical therapy. Moreover, the data regarding SES and social support in HF are derived from small studies that did not examine the multiple components of SES and partner status together. In the HF-ACTION study, data on partner status and SES were obtained at baseline by patient report. We investigated the associations between baseline partner status/SES and outcomes in patients with chronic systolic HF enrolled in HF-ACTION.
American Heart Journal Volume 183
Methods HF-ACTION (clinicaltrials.gov, NCT00047437) was a randomized controlled clinical trial of aerobic exercise training in patients with chronic HF; the design, rationale, and primary results have been published. 6, 7 The study enrolled 2,331 ambulatory patients between April 2003 and February 2007 who had left ventricular ejection fraction (EF) ≤35% and New York Heart Association (NYHA) class II to IV HF while on optimal HF therapy for at least 6 weeks. Patients were randomized to either usual care or exercise training in addition to usual care and followed up for at least 12 months. Exercise training consisted of 36 supervised exercise sessions over the initial 3 months, followed by home training on a treadmill or stationary cycle for another 2 years. The protocol was approved by the institutional review board for each site involved in the study, and written informed consent was obtained for all patients. Partner status and SES were defined as reported in prior analyses from the HF-ACTION trial data set. 8 Partner status was ascertained by self-report of either having a partner (married or living with a partner) or no partner (single, never married, divorced, separated, or widowed). Socioeconomic status was measured by self-reported income (most recent annual household income before taxes; lower income defined as b$25,000 and higher income defined as ≥$25,000), employment status (employed defined as student, self-employed, part-time employed, or full-time employed; unemployed defined as homemaker, volunteer, retired, disabled, or unemployed), and education level (lower education defined as high school graduate/equivalent or less and higher education defined as greater than high school education, including some college, associate degree, college graduate, and/or graduate school degree). Functional capacity was assessed by cardiopulmonary exercise testing to evaluate peak oxygen consumption (peak VO2) as well as by 6-minute walk test to measure distance walked. Adherence to therapy was evaluated in patients randomized to exercise training by measuring total time of exercise performed during supervised training sessions and patient self-report of home training sessions. Quality of life was measured by the Kansas City Cardiomyopathy Questionnaire (KCCQ). The primary end point was all-cause mortality and hospitalization, with secondary end points including all-cause death and composite cardiovascular death or HF hospitalization. These outcomes were adjudicated by a committee blinded to treatment assignment.
Statistical analysis Patients were grouped per independent variable of interest in dichotomous categories (eg, partner or no partner) and baseline characteristics were described. Continuous variables were reported as the median with 25th-75th percentiles and compared with the independent variable of interest using the Wilcoxon rank sum statistic.
Verma et al 55
Categorical variables were reported as percentages and compared with the independent variable of interest with Pearson χ 2 or exact test. The relationship of the independent variables of interest with outcomes was assessed using Cox proportional hazards ratios and was adjusted based on the variables consistently used in HF-ACTION post hoc analyses (see Table II footnote). 9 Regression models used complete case data. In addition, several additional variables were included that were thought to potentially confound outcomes associations: age, sex, race, history of depression, and Beck Depression Score. 10,11 Kaplan-Meier estimates were generated for the primary outcome. The relationship of independent variables was also investigated with regard to functional capacity and adherence to therapy. We also assessed for interaction of the independent variables and exercise treatment. A 2-tailed P value b.05 was considered statistically significant for all analyses. No adjustment was made for the assessment of multiple end points in the present article given the exploratory nature of the investigation. The SAS system, version 9.2, was used for analyses (SAS, Cary, NC), all of which were conducted by Duke Clinical Research Institute.
Results The baseline characteristics of patients based on partner status and SES are described in Table I. Of the patients included in the study, 61% reported having a partner, 60% had more than a high school education, 59% had an income ≥$25,000, and 24% were employed. Patients with a current partner had higher income and tended to be older, male, and white. Patients with greater than a high school education included a larger proportion of women and white. Higher-income patients tended to be older, male, and white. Employment was more common in younger individuals. There were no statistically significant differences in medical therapies based on partner status or SES, with the exception of β-blocker use being more prevalent in individuals who were employed. Internal cardioverter/ defibrillator use was significantly more prevalent among those who had a partner, had greater than a high school education, had higher income, and were unemployed. Comorbidities such as diabetes and hypertension were more common in individuals with less education, lower income, and unemployed status. The NYHA class was also worse in patients with less education, lower income, and unemployed status. Table II describes baseline functional capacity and quality of life of patients based on partner status and SES. There were statistically significant differences in all groups with regard to exercise tolerance (measured by peak VO2 and 6-minute walk test) and health-related quality of life (measured by the KCCQ), favoring having a partner, higher level of education, higher income, and being employed.
American Heart Journal January 2017
56 Verma et al
Table I. Baseline characteristics by partner status and SES Partner status No partner (n = 899)
Partner (n = 1411)
Education level High school or less (n = 920)
Patient characteristics Age (y) 58 (49-67)⁎ 61 (53-69)⁎ 60 (51-68) Female sex 40.9%⁎ 20.3%⁎ 20.3%⁎ Race Black 45.7%⁎ 24.0%⁎ 39.2%⁎ White 49.8%⁎ 70.2%⁎ 56.4%⁎ Other 4.5%⁎ 5.8%⁎ 4.4%⁎ BMI (kg/m2) 30 (26-36)⁎ 30 (26-34)⁎ 30 (26-35) Medical history Hypertension 61.3% 58.7% 63.7%⁎ Diabetes 33.4% 31.2% 33.9% Depression 22.2% 20.2% 19.3% Laboratory results Na (mmol/L) 139 (137-141) 139 (137-141) 139 (137-141) Creatinine (mg/dL) 1.2 (1.0-1.5) 1.2 (1.0-1.5) 1.2 (1.0-1.5) Pro-BNP (pg/mL) 735 857 789 (321-1766) (346-1878) (373-1750) Medications and devices β-Blocker 95.2% 94.2% 94.2% ACE inhibitor or 94.2% 94.3% 94.2% ARB MRA 45.7% 44.7% 43.8% ICD 33.9%⁎ 44.3%⁎ 36.6%⁎ HF history HF hospitalizations prior 6 mo None 69.2%⁎ 76.3%⁎ 73.7% 1 21.9%⁎ 19.0%⁎ 20.7% 2 5.6%⁎ 3.1%⁎ 3.6% ≥3 3.4%⁎ 1.6%⁎ 2.0% Left ventricular EF 25 (20-30) 25 (25-30) 25 (20-30) NYHA III-IV vs II 38.8% 35.1% 39.6%⁎
Income level
Employment
More than high school (n = 1359)
Lower income (n = 853)
Higher income (n = 1220)
Unemployed (n = 1731)
Employed (n = 551)
59 (51-68) 29.0%⁎
58 (50-68)⁎ 33.6%⁎
60 (52-68)⁎ 24.3%⁎
61 (53-70)⁎ 27.9%
55 (47-61)⁎ 29.8%
28.4%⁎ 65.7%⁎ 5.9%⁎
43.2%⁎ 51.1%⁎ 5.7%⁎
24.1%⁎ 71.0%⁎ 4.9%⁎
29.8% 65.0% 5.2% 30 (26-35) 53.0%⁎ 25.0%⁎ 19.1%
30 (26-35)
31 (26-36)⁎
30 (26-35)⁎
33.9% 60.8% 5.3% 30 (26-35)
57.7%⁎ 31.0% 22.4%
65.4%⁎ 37.4%⁎ 22.7%
57.3%⁎ 28.9%⁎ 20.2%
62.2%⁎ 34.5%⁎ 21.7%
139 (137-141) 1.2 (1.0-1.5) 831 (331-1812)
139 (137-141) 139 (137-141) 139 (137-141) 1.2 (1.0-1.5) 1.2 (1.0-1.5) 1.2 (1.0-1.5)⁎ 857 819 868 (375-1793) (323-1821) (376-1992)⁎
139 (138-141) 1.1 (0.9-1.3)⁎ 612 (255-1411)⁎
94.6% 94.6%
93.8% 93.9%
95.4% 95.0%
93.9%⁎ 94.0%
96.4%⁎ 95.5%
45.8% 41.9%⁎
44.1% 36.0%⁎
45.2% 43.4%⁎
45.2% 42.6%⁎
45.4% 32.1%⁎
73.5% 19.8% 4.2% 2.5% 25 (20-30) 34.7%⁎
72.1% 20.9% 4.0% 3.0% 25 (20-30) 43.11%⁎
74.6% 19.5% 4.1% 1.7% 25 (20-30) 32.8%⁎
74.8%⁎ 19.0%⁎ 3.6%⁎ 2.6%⁎
69.2%⁎ 23.8%⁎ 5.7%⁎ 1.3%⁎
25 (20-30) 40.8%⁎
25 (20-30) 21.8%⁎
Continuous variables are shown as median (25th-75th percentile) and categorical variables are shown as % (n). Abbreviations: BMI, body mass index; BNP, brain natriuretic peptide; ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blocker; MRA, mineralocorticoid receptor antagonist; ICD, implantable cardioverter defibrillator. ⁎ P b .05 for comparison.
Adherence to exercise was measured in patients randomized to the exercise training arm of the HF-ACTION study. Table III shows that patients with a partner and who had higher income had greater supervised exercise and total exercise times compared with their counterparts without a partner and with lower income, respectively. The associations between partner status and SES and primary outcomes are shown in Table IV. In the unadjusted model, having a partner, higher education, higher income, and being employed were all significantly associated with lower all-cause mortality/hospitalization and cardiovascular death/HF hospitalization. Figure presents the Kaplan-Meier curve for all-cause mortality/ hospitalization by partner status and education. After risk adjustment, all-cause mortality/hospitalization and cardiovascular death/HF hospitalization were non-
significantly lower in patients with a partner and higher education (all-cause mortality/hospitalization: partner-adjusted hazard ratio [HR] 0.91, 95% CI 0.81-1.03; higher education–adjusted HR 0.91, 95% CI 0.80-1.02; cardiovascular death/HF hospitalization: partner-adjusted HR 0.86, 95% CI 0.72-1.03; higher education–adjusted HR 0.86, 95% CI 0.73-1.02). There was no interaction between any of the partner status or SES variables and exercise training on outcomes (all P N .5).
Discussion Our data describe the associations of partner status and SES with exercise capacity, quality of life, and clinical outcomes in patients with chronic systolic HF. Those with a higher SES and a partner had better baseline exercise tolerance and health-related quality of life,
American Heart Journal Volume 183
Verma et al 57
Table II. Baseline exercise capacity and quality of life by partner status and SES Partner status
No partner (n = 899)
Partner (n = 1411)
Education level High school or less (n = 920)
Income level
More than high school (n = 1359)
Lower income (b$25,000) (n = 853)
Higher income (≥$25,000) (n = 1220)
Employment
Unemployed (n = 1731)
Employed (n = 551)
Peak VO2 13.8 (11.2-17.0)⁎ 14.9 (11.8-18.0)⁎ 13.7 (10.9-16.7)⁎ 15.1 (12.0-18.0)⁎ 13.3 (11.0-16.7)⁎ 15.2 (12.3-18.1)⁎ 13.7 (11.1-17.0)⁎ 16.5 (13.7-19.7)⁎ (mL kg−1 min−1) 6-min walk 357 (290-420)⁎ 381 (305-444)⁎ 355 (287-420)⁎ 382 (306-442)⁎ 350 (276-414)⁎ 388 (316-445)⁎ 358 (286-421)⁎ 415 (350-474)⁎ distance (m) KCCQ overall 66 (49-82)⁎ 69 (52-84)⁎ 66 (49-83)⁎ 69 (53-84)⁎ 63 (48-80)⁎ 71 (55-85)⁎ 65 (48-82)⁎ 76 (63-88)⁎ summary score Continuous variables are shown as median (25th-75th percentile) and categorical variables are shown as % (n). Partner reference group is no partner, higher education reference group is lower education (high school or less), higher income reference group is lower income (b$25,000), and employment reference group is unemployed. ⁎ P b .05.
Table III. Adherence to exercise by partner status and SES Partner status No partner (n = 899)
Partner (n = 1411)
Education level High school or less (n = 920)
Income level
More than high school (n = 1359)
Lower income (b$25,000) (n = 853)
Higher income (≥$25,000) (n = 1220)
Employment Unemployed (n = 1731)
Employed (n = 551)
Supervised exercise 560 (220–810)⁎ 675 (443-846)⁎ 629 (296-830) 640 (386-838) 546 (189-780)⁎ 670 (454-856)⁎ 640 (360-837) 629 (370-838) (min) Total exercise (min) 63 (24-103)⁎ 85 (49-126)⁎ 75 (32-117) 79 (44-118) 65 (19-102)⁎ 85 (51-123)⁎ 80 (38-118) 70 (40-118) Continuous variables are shown as median (25th-75th percentile). Partner reference group is no partner, higher education reference group is lower education (high school or less), higher income reference group is lower income (b$25,000), and employment reference group is unemployed. ⁎ P b .05.
whereas having a partner and higher income were associated with better adherence to exercise therapy. Partner status and SES were not significantly associated with the primary outcome, although there was a trend for better outcomes among individuals with a partner and with greater than a higher school education. Socioeconomic status encompasses education, income, and employment, which are resources that help support an individual's ability to function and manage the demands of life. Furthermore, partner status is a marker of significant social support. Previous analyses have demonstrated a correlation between SES and life expectancy. 12-15 In the present article, we extend these results by demonstrating in patients with chronic HF the associations of partner status and SES with mortality and hospitalization as well as with functional capacity, health-related quality of life, and adherence to prescribed therapy. Quality of life and functional capacity are the “visible” parameters that patients with HF experience on a day-to-day basis and how many providers, at least in part, gage the status of the underlying illness. Indeed, these variables are also incorporated into both the patient and physician decision making to help guide therapy. 16 Furthermore, the multifaceted nature of quality of life and functional capacity is difficult to fully characterize
because it differs among individuals and is not easily predicted. For instance, some may find a costly HF medication that improves dyspnea to improve quality of life, whereas others may find that despite improved symptom burden, the cost of the medication negatively impacts their definition of quality of life. 17 Prior studies have shown that lower SES is related to exercise intolerance in patients with congenital heart disease and to impaired exercise capacity in patients with coronary heart disease. 18-20 Indeed, in a recent HF-ACTION analysis, lower perceived social support was associated with decreased exercise time. 2 However, functional capacity in relation to individual SES and partner status has not previously been studied in patients with HF. Here, we found that having a partner, more than a high school education, higher income, and being employed were all associated with improved functional capacity and quality of life. Partner status has been shown to be associated with outcomes in various cardiovascular disease states. Studies have shown an association between being married and reduced all-cause cardiovascular mortality in patients with coronary artery disease, post–myocardial infarction, HF, and after heart transplantation. 21-28 These observations may be mediated by risk factor modification related
American Heart Journal January 2017
58 Verma et al
Table IV. Clinical outcomes by partner status and SES Partner HR (95% CI)
Higher education P
HR (95% CI)
Higher income P
HR (95% CI)
Employed P
HR (95% CI)
P
All-cause mortality or hospitalization Unadjusted 0.90 (0.81-1.00) .042 0.86 (0.78-0.95) .004 0.87 (0.78-0.97) .011 0.71 (0.63-0.80) b.001 Adjusted⁎ 0.91 (0.81-1.03) .15 0.91 (0.80-1.02) .12 0.97 (0.85-1.10) .64 0.97 (0.83-1.12) .67 All-cause mortality Unadjusted 0.89 (0.73-1.10) .28 0.73 (0.60-0.90) .002 0.83 (0.68-1.03) .10 0.53 (0.40-0.70) b.001 Adjusted† 1.07 (0.84-1.35) .58 0.84 (0.67-1.05) .13 1.12 (0.88-1.43) .35 1.05 (0.76-1.45) .77 Cardiovascular death or HF hospitalization Unadjusted 0.77 (0.66-0.89) b.001 0.75 (0.65-0.87) b.001 0.77 (0.66-0.90) b.001 0.50 (0.40-0.61) b.001 Adjusted‡ 0.86 (0.72-1.03) .11 0.86 (0.73-1.02) .08 0.94 (0.79-1.14) .56 0.88 (0.69-1.10) .256 Hazard ratios are reported with associated 95% CI. Partner reference group is no partner, higher education reference group is lower education (high school or less), higher income reference group is lower income (b$25,000), and employment reference group is unemployed. ⁎ Adjusted for Weber class, Kansas City Cardiomyopathy Questionnaire (KCCQ) symptom stability score, blood urea nitrogen, country, left ventricular EF, sex, age, race, history of depression, Beck Depression Score at baseline, β-blocker dosage, mitral regurgitation grade, and ventricular conduction before baseline cardiopulmonary exercise test. † Adjusted for exercise duration on baseline cardiopulmonary exercise test, serum creatinine level, body mass index (BMI), sex, age, race, history of depression, Beck Depression Score at baseline, loop diuretic dose, left ventricular EF, Canadian Cardiovascular Society angina classification, and ventricular conduction before baseline cardiopulmonary exercise test. ‡ Adjusted for loop diuretic dose, left ventricular EF, mitral regurgitation grade, ventricular conduction before baseline cardiopulmonary exercise test, KCCQ symptom stability score, blood urea nitrogen, race, sex, age, Weber class, history of depression, Beck Depression Score at baseline, and VE/VCO2 on baseline cardiopulmonary exercise test.
to “benefits from spousal support with regard to seeking treatment, adherence to treatment, and recommended lifestyle changes, as well as greater financial resources, which make medical treatment and healthy lifestyle choices affordable.” 22 Importantly, though, the implications of marital status likely also depend on quality of the relationship as has been demonstrated previously. 29,30 There are only a few, small studies that specifically examined the association of partner status in patients with HF, with most showing longer event-free survival and decreased rehospitalization in patients with a partner. 22-28, 31, 32 Our study further adds to these findings in a large HF cohort in addition to measuring other socioeconomic factors concurrently. There was an insignificant trend for individuals with a partner to have better outcomes. The partner relationship is complex and is thought to mediate its effects via improvement in medication adherence and decrease in depressive symptoms. 3, 28 Indeed, we found that patients with a partner had better adherence to exercise therapy compared with those without a partner. Interestingly, our study also found that partner status was associated with a statistically significant 12% reduction in the primary outcome of all-cause mortality or hospitalization only when excluding depression from the adjusted model, thus supporting the concept that having a partner is potentially mediated in part by affecting mood. We performed additional analyses to assess whether prior exposure to partnership (ie, patients who were widowed, divorced, or separated) was associated with outcomes to understand if the experience of a past or current relationship may in itself be important; however, there was no significant difference among patients with no partner, a prior partner, or a current partner. Although our study examined partners
who lived with the patient, future studies could also examine whether this relationship holds true when examining a patient's social network outside an immediate partner. We also observed an insignificant association between more than a high school education and reduced all-cause mortality/hospitalization and cardiovascular death or HF hospitalization. Prior studies of education have generally found similar associations of reduction in mortality and rehospitalization or equivocal associations with outcomes in patients with HF; however, these studies have had small sample sizes and some have used regional data rather than patient-reported data. 4, 33-35 Although employment status was not significantly associated with outcomes, it was associated with higher functional capacity and quality of life. One could posit that the working individuals are able to do so because they have a better functional capacity and quality of life, thus permitting them to do so. Indeed, employed individuals were younger, with less comorbidity, a larger proportion in NYHA classes II to III vs IV as compared with individuals who were not employed. Different types of employment such as manual labor vs sedentary work may also confer different associations with outcomes. Thus, future studies are needed to better define this relationship. Several studies have examined SES with regard to outcomes in patients with HF and have shown that lower income level is associated with an increase in hospitalization and mortality in patients with HF. 4, 35-39 Our study did not find a clear relationship between income level and clinical outcomes. Several key differences in these studies with regard to income analysis may account for the discrepancy of our findings. For instance, several of these studies assessed income indirectly by evaluating
American Heart Journal Volume 183
Figure
A
B
Verma et al 59
outcomes that many of these studies propose. As we have shown, perhaps by not measuring income by accessibility to medication and care, low income may not independently portend worse outcomes. Although our findings show an association of SES with outcomes in HF, the implication of these findings is not straightforward. It is unclear if interventions could impact outcomes as this study shows the association of socioeconomic and partner status on outcomes rather than causality of variables. Moreover, potential interventions to influence socioeconomic and partner status are complex and cannot be readily implemented. Socioeconomic status and partner status should be acknowledged as factors that could provide useful information in understanding and perhaps risk stratifying individuals with regard to outcomes. In addition, these findings also pose interesting avenues of exploration to try to understand possible causality of the complex relationship of SES and partner status to then allow realistic interventions to be tested.
Limitations This was a retrospective analysis of a clinical trial, and therefore, causal relationships cannot be determined. Furthermore, there were strict inclusion/exclusion criteria in the context of the trial such that the findings may not apply to the broader HF population. The definitions of partner status and SES may vary based on culture, geographic location, and over time. Thus, these findings may potentially be limited by the definitions imposed here. Information regarding partner status and SES was obtained by self-report and may be subject to biased reporting.
Conclusions
Unadjusted all-cause mortality or hospitalization by education (A) and partner status (B). Kaplan-Meier estimates of all-cause mortality of hospitalization by education and partner status through 4 years of follow-up.
mean or median income in the geographic location where a patient lived or asking the patient whether he/she was financially “comfortable.” 4, 36, 40 Geographic area may better correlate with access to care rather than the income itself, and subjective ideas of financial stability can be a proxy for other factors that do not just include finances. Furthermore, with the high percentage of patients on appropriate HF medications at baseline in HF-ACTION, it does not seem that income was an overt barrier to getting medical therapies, which is the theorized reason for worse
In patients with chronic HF, having a partner, being educated beyond high school, having an income N$25,000, and being employed were associated with improved baseline functional capacity and quality of life. Having a partner and more than a high school education may also be associated with lower rates of all-cause mortality and hospitalization; however, further studies will be needed to confirm these associations. These data suggest the potential importance of life partnership and SES with regard to long-term outcomes and in maintaining functional status and quality of life. Future studies of social support interventions and SES-tailored care would be useful to determine if these factors may be helpful in the management of HF.
Disclosures All authors have reported that they have no relationships relevant to the contents of this manuscript to report. All authors were responsible for the design, analyses, drafting, and editing of the presented research.
60 Verma et al
References 1. Fiscella K, Tancredi D. Socioeconomic status and coronary heart disease risk prediction. JAMA 2008;300(22):2666-8. 2. Cooper LB, Mentz RJ, Sun JL, et al. Psychosocial factors, exercise adherence, and outcomes in heart failure patients: insights from Heart Failure: A Controlled Trial Investigating Outcomes of Exercise Training (HF-ACTION). Circ Heart Fail 2015;8(6):1044-51. 3. Wu JR, Lennie TA, Chung ML, et al. Medication adherence mediates the relationship between marital status and cardiac event-free survival in patients with heart failure. Heart Lung 2012;41(2):107-14. 4. Eapen ZJ, McCoy LA, Fonarow GC, et al. Utility of socioeconomic status in predicting 30-day outcomes after heart failure hospitalization. Circ Heart Fail 2015. 5. Havranek EP, Mujahid MS, Barr DA, et al. Social determinants of risk and outcomes for cardiovascular disease: a scientific statement from the American Heart Association. Circulation 2015;132(9): 873-98. 6. O'Connor CM, Whellan DJ, Lee KL, et al. Efficacy and safety of exercise training in patients with chronic heart failure: HF-ACTION randomized controlled trial. JAMA 2009;301(14):1439-50. 7. Whellan DJ, O'Connor CM, Lee KL, et al. Heart Failure and A Controlled Trial Investigating Outcomes of Exercise Training (HF-ACTION): design and rationale. Am Heart J 2007;153(2):201-11. 8. Mentz RJ, Babyak MA, Bittner V, et al. Prognostic significance of depression in blacks with heart failure: insights from Heart Failure: A Controlled Trial Investigating Outcomes of Exercise Training. Circ Heart Fail 2015;8(3):497-503. 9. Mentz RJ, Bittner V, Schulte PJ, et al. Race, exercise training, and outcomes in chronic heart failure: findings from Heart Failure—A Controlled Trial Investigating Outcomes in Exercise TraiNing (HF-ACTION). Am Heart J 2013;166(3):488-95. 10. O'Connor CM, Whellan DJ, Wojdyla D, et al. Factors related to morbidity and mortality in patients with chronic heart failure with systolic dysfunction: the HF-ACTION predictive risk score model. Circ Heart Fail 2012;5(1):63-71. 11. Mentz RJ, Schulte PJ, Fleg JL, et al. Clinical characteristics, response to exercise training, and outcomes in patients with heart failure and chronic obstructive pulmonary disease: findings from Heart Failure and A Controlled Trial Investigating Outcomes of Exercise TraiNing (HF-ACTION). Am Heart J 2013;165(2):193-9. 12. Kelishadi R, Poursafa P. A review on the genetic, environmental, and lifestyle aspects of the early-life origins of cardiovascular disease. Curr Probl Pediatr Adolesc Health Care 2014;44(3):54-72. 13. de Andrade LO, Pellegrini Filho A, Solar O, et al. Social determinants of health, universal health coverage, and sustainable development: case studies from Latin American countries. Lancet 2015;385(9975): 1343-51. 14. Braveman P, Gottlieb L. The social determinants of health: it's time to consider the causes of the causes. Public Health Rep 2014;129(Suppl 2): 19-31. 15. Rendall MS, Weden MM, Favreault MM, et al. The protective effect of marriage for survival: a review and update. Demography 2011;48(2): 481-506. 16. Olano-Lizarraga M, Oroviogoicoechea C, Errasti-Ibarrondo B, et al. The personal experience of living with chronic heart failure: a qualitative meta-synthesis of the literature. J Clin Nurs 2016. 17. Karimi M, Clark AM. How do patients' values influence heart failure self-care decision-making? A mixed-methods systematic review. Int J Nurs Stud 2016;59:89-104. 18. Shishehbor MH, Litaker D, Pothier CE, et al. Association of socioeconomic status with functional capacity, heart rate recovery, and all-cause mortality. JAMA 2006;295(7):784-92.
American Heart Journal January 2017
19. Diller GP, Inuzuka R, Kempny A, et al. Detrimental impact of socioeconomic status on exercise capacity in adults with congenital heart disease. Int J Cardiol 2013;165(1):80-6. 20. Cohen B, Vittinghoff E, Whooley M. Association of socioeconomic status and exercise capacity in adults with coronary heart disease (from the Heart and Soul Study). Am J Cardiol 2008;101(4): 462-6. 21. Molloy GJ, Stamatakis E, Randall G, et al. Marital status, gender and cardiovascular mortality: behavioural, psychological distress and metabolic explanations. Soc Sci Med 2009;69(2):223-8. 22. Quinones PA, Kirchberger I, Heier M, et al. Marital status shows a strong protective effect on long-term mortality among first acute myocardial infarction-survivors with diagnosed hyperlipidemia—findings from the MONICA/KORA myocardial infarction registry. BMC Public Health 2014;14:98. 23. Tam V, Arnaoutakis GJ, George TJ, et al. Marital status improves survival after orthotopic heart transplantation. J Heart Lung Transplant 2011;30(12):1389-94. 24. Watkins T, Mansi M, Thompson J, et al. Effect of marital status on clinical outcome of heart failure. J Investig Med 2013;61(5): 835-41. 25. Howie-Esquivel J, Spicer JG. Association of partner status and disposition with rehospitalization in heart failure patients. Am J Crit Care 2012;21(3):e65-73. 26. Calvillo-King L, Arnold D, Eubank KJ, et al. Impact of social factors on risk of readmission or mortality in pneumonia and heart failure: systematic review. J Gen Intern Med 2013;28(2):269-82. 27. Chung ML, Lennie TA, Riegel B, et al. Marital status as an independent predictor of event-free survival of patients with heart failure. Am J Crit Care 2009;18(6):562-70. 28. Staniute M, Brozaitiene J, Burkauskas J, et al. Type D personality, mental distress, social support, and health-related quality of life in coronary artery disease patients with heart failure: a longitudinal observational study. Health Qual Life Outcomes 2015;13(1):1. 29. Donoho CJ, Seeman TE, Sloan RP, et al. Marital status, marital quality, and heart rate variability in the MIDUS cohort. J Fam Psychol 2015;29(2):290-5. 30. Joseph NT, Kamarck TW, Muldoon MF, et al. Daily marital interaction quality and carotid artery intima-medial thickness in healthy middle-aged adults. Psychosom Med 2014;76(5):347-54. 31. Tsuchihashi M, Tsutsui H, Kodama K, et al. Medical and socioenvironmental predictors of hospital readmission in patients with congestive heart failure. Am Heart J 2001;142(4):E7. 32. Luttik ML, Jaarsma T, Veeger N, et al. Marital status, quality of life, and clinical outcome in patients with heart failure. Heart Lung 2006;35(1):3-8. 33. Roe-Prior P. Sociodemographic variables predicting poor post-discharge outcomes for hospitalized elders with heart failure. Medsurg Nurs 2007;16(5):317-21. 34. Schwarz KA, Elman CS. Identification of factors predictive of hospital readmissions for patients with heart failure. Heart Lung 2003;32(2): 88-99. 35. Damiani G, Salvatori E, Silvestrini G, et al. Influence of socioeconomic factors on hospital readmissions for heart failure and acute myocardial infarction in patients 65 years and older: evidence from a systematic review. Clin Interv Aging 2015;10:237-45. 36. Heo S, Moser DK, Chung ML, et al. Social status, health-related quality of life, and event-free survival in patients with heart failure. Eur J Cardiovasc Nurs 2012;11(2):141-9.
American Heart Journal Volume 183
37. Rathore SS, Masoudi FA, Wang Y, et al. Socioeconomic status, treatment, and outcomes among elderly patients hospitalized with heart failure: findings from the National Heart Failure Project. Am Heart J 2006;152(2):371-8. 38. Philbin EF, Dec GW, Jenkins PL, et al. Socioeconomic status as an independent risk factor for hospital readmission for heart failure. Am J Cardiol 2001;87(12):1367-71.
Verma et al 61
39. Hawkins NM, Jhund PS, McMurray JJ, et al. Heart failure and socioeconomic status: accumulating evidence of inequality. Eur J Heart Fail 2012;14(2):138-46. 40. Bikdeli B, Wayda B, Bao H, et al. Place of residence and outcomes of patients with heart failure: analysis from the telemonitoring to improve heart failure outcomes trial. Circ Cardiovasc Qual Outcomes 2014;7(5):749-56.