Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Sep 8.
Published in final edited form as: Psychol Health. 2019 May 21;34(10):1250–1266. doi: 10.1080/08870446.2019.1609676

Associations of perceived stress and state anger with symptom burden and functional status in patients with heart failure

Romano Endrighi a,b, Andrew J Dimond a, Andrew J Waters a, Christopher C Dimond c, Kristie M Harris d, Stephen S Gottlieb e, David S Krantz a
PMCID: PMC8425712  NIHMSID: NIHMS1737200  PMID: 31111738

Abstract

Background:

Psychosocial stress and anger trigger cardiovascular events, but their relationship to heart failure (HF) exacerbations is unclear. We investigated perceived stress and anger associations with HF functional status and symptoms.

Methods and Results:

In a prospective cohort study (BETRHEART), 144 patients with HF (77% male; 57.5 ± 11.5 years) were evaluated for perceived stress (Perceived Stress Scale; PSS) and state anger (STAXI) at baseline and every 2 weeks for 3 months. Objective functional status (6-min walk test; 6MWT) and health status (Kansas City Cardiomyopathy Questionnaire; KCCQ) were also measured biweekly. Linear mixed model analyses indicated that average PSS and greater than usual increases in PSS were associated with worsened KCCQ scores. Greater than usual increases in PSS were associated with worsened 6MWT. Average anger levels were associated with worsened KCCQ, and increases in anger were associated with worsened 6MWT. Adjusting for PSS, anger associations were no longer statistically significant. Adjusting for anger, PSS associations with KCCQ and 6MWT remained significant.

Conclusion:

In patients with HF, both perceived stress and anger are associated with poorer functional and health status, but perceived stress is a stronger predictor. Negative effects of anger on HF functional status and health status may partly operate through psychological stress.

Keywords: Anger, functional status, heart failure, perceived stress, symptoms

Introduction

Heart failure (HF) is a chronic disease in which the heart is unable to pump sufficiently to provide adequate blood flow to meet the body’s needs. It is characterised by frequent exacerbations, a high rate of rehospitalisations and a poor five-year survival (Coles et al., 2014; Dharmarajan et al., 2013; Joffe et al., 2013). HF is not only characterised by changes in cardiac function: it is a syndrome defined in part by deteriorations in a patient’s symptoms and functional status (Feenstra, Grobbee, Jonkman, Hoes, & Stricker, 1998; Zaya, Phan, & Schwarz, 2012). Symptoms experienced by HF patients include shortness of breath, swelling of the legs, nausea, dizziness and fatigue. Poor health status and symptoms and impaired functional status in patients with HF are themselves consistently associated with each other and predictive of adverse HF outcomes such as rehospitalisation and mortality (Alahdab, Mansour, Napan, & Stamos, 2009; Flynn et al., 2012; Heidenreich et al., 2006; Pocock et al., 2006; Wegrzynowska-Teodorczyk et al., 2013; Zaya et al., 2012)

Psychosocial factors such as depression, stress and anger are also prospectively associated with adverse HF outcomes including hospitalisations and/or death (Albert et al., 2009; Endrighi et al., 2016; Keith et al., 2017; Pelle, Gidron, Szabo, & Denollet, 2008; Perlman, Ferguson, Bergum, Isenberg, & Hammarsten, 1971; Rod, Andersen, & Prescott, 2011; Rumsfeld et al., 2003). The vast majority of research on psychosocial factors and HF outcomes has focused on depression. Among patients with HF, the presence of depression as well as changes in depressive symptoms over time, confers a greater than twofold risk for hospitalisations and death for cardiovascular causes independent of established risk factors (Kop, Synowski, & Gottlieb, 2011; Sherwood et al., 2007, 2011). A high level of perceived stress appears to confer about a 25% increased risk for new incidence of coronary heart disease (Richardson et al., 2012). However, less is known about the impact of increased perceived stress in patients with HF.

Patients with HF experience frequent periods where their symptoms and functional status worsen. Changes in symptoms and functional status are two important consequences of HF that are related to each other, and also independently associated with increased risk of future HF-related events (Ingle et al., 2007; Zaya et al., 2012). Cross-sectional studies indicate that depression and distress are associated with HF symptom burden (Bekelman et al., 2007; Smith, Pedersen, Van Domburg, & Denollet, 2008). With respect to objectively assessed functional status in patients with HF, one cross-sectional study found that depressive symptoms were associated with small but significant impairments in functional status assessed via the six-minute walk test (6MWT) (Gottlieb et al., 2009). Repeated assessments of depression have been utilised in cardiac patients to study clinical outcomes. With respect to HF specifically, at least one study provides longitudinal evidence that changes in depressive symptoms over time predict symptom worsening and adverse outcomes in patients with HF (Sherwood et al., 2011).

Traits such as hostility and an angry emotional state have also been implicated as risk factors for poor outcomes in cardiac patients (Chida & Steptoe, 2009; Suls & Bunde, 2005). For example, there is evidence that following anger outbursts, the risk of myocardial infarction increases between 2 and 4.7 times (Mittleman et al., 1995; Mostofsky, Maclure, Tofler, Muller, & Mittleman, 2013), and risk of daily life myocardial ischemia is doubled (Gabbay et al., 1996). Anger is also related to cardiac symptoms in women with suspected coronary disease (Krantz et al., 2006), and momentary anger precedes more severe symptoms in other chronic disease patients (Russell, Smith, & Smyth, 2016).

However, few studies have examined the possible effects of anger levels over time in patients with HF. One prospective study demonstrated a modest association between trait anger and longer length of hospital stay (Jenner, Strodl, & Schweitzer, 2009). In another study, trait anger and anger expression predicted a greater number of all-cause hospitalisations over a 3-year follow-up (Keith et al., 2017). Since psychological factors such as depression are related to functional status as measured by the 6MWT, it is therefore possible that acutely increased anger in patients with HF would also be associated with decrements in functional status as well as increases in HF symptom burden.

A better understanding of how fluctuations in psychosocial factors precipitate HF exacerbations may help improve risk assessments by identifying ‘which patient is at risk’ and ‘when they are at risk’. Therefore, the primary aim of this study is to examine between-person (average level) and within-person (short-term change) associations of perceived stress and anger with self-reported health status and symptoms, and with objectively-assessed functional status in patients with HF. A prior report in the present sample of patients (Endrighi et al., 2016) demonstrated that higher average levels of perceived stress measured repeatedly over a three-month period were prospectively related to increased risk of cardiovascular-related hospitalisations or death. Based on this finding and on associations between acute anger episodes and cardiac events in prior research (Mostofsky et al., 2013), we hypothesised that patients with higher (vs. lower) average levels of perceived stress and anger measured repeatedly would exhibit poorer health and functional status. We also expected that short-term increases (compared to patient’s average levels) of perceived stress and anger would be associated with corresponding declines in HF-related health status and symptoms and objectively-assessed functional status. We also expect that using a strategy of repeated assessments would provide a stronger and more consistent association with outcomes compared to a single baseline-only assessment. A secondary study aim is to assess the independent effects, if any, of perceived stress and anger on HF-related health status and symptoms and functional status. Anger is often a response to situations perceived as stressful (Kubzansky & Winning, 2016), and individuals exhibiting greater anger experience more frequent and severe daily stressors (Smith & Frohm, 1985). However, it is currently unknown whether perceived stress and anger affect health and functional status conjointly or operate independently. In the absence of such evidence, we hypothesised that anger and perceived stress would demonstrate independent relationships with HF-related health status and symptoms and functional status in patients with HF.

Method

Design and participants

Patients were participants in the BETRHEART Study (Behavioral Triggers of Heart Failure; Endrighi et al., 2016; Keith et al., 2017), a prospective cohort study of patients (N=144) with a primary diagnosis of systolic HF receiving treatment at the University of Maryland Medical Center (UMMC). Inclusion criteria were: Left ventricular ejection fraction (LVEF) ≤40% and symptomatic HF (NYHA class II–IV) for ≥3 months. Exclusion criteria were: significant mitral valve disease, myocarditis in past six-months, past six-month alcohol abuse, thyroid dysfunction as primary HF etiology, LV assist device, active cancer treatment or significant cognitive impairment indicated via prior diagnosis or inability to provide informed consent and/or understand instructions. The study was approved by the IRBs at UMMC and the Uniformed Services University.

As described previously (Endrighi et al., 2016), the study design consisted of two clinical assessment visits for each patient – a baseline assessment when the patient entered the study and a follow-up assessment after three months. In addition, five phone interviews were scheduled every two-weeks between the baseline and three-month assessments. A subset of patients was also scheduled for up to five additional abbreviated clinic assessment visits (interim clinic assessments) during which patients performed the 6MWT, completed a brief clinical questionnaire and provided a blood sample (not described here).

Procedure

Psychological, health and functional status assessment (baseline and three months)

Patients received clinically indicated HF treatment and medication management at the UMMC. In addition, patients were administered several validated psychosocial and clinical questionnaires including the Perceived Stress Scale (PSS; Cohen, Kamarck, & Mermelstein, 1983), the State Trait Anger Expression Inventory-2 (STAXI-2; Spielberger et al., 1985), the Kansas City Cardiomyopathy Questionnaire (KCCQ; Green, Porter, Bresnahan, & Spertus, 2000) and an objective measure of functional status, the six-minute walk test (6MWT; Balke, 1963).

Telephone interviews

Phone interviews occurred at 2, 4, 6, 8 and 10 weeks after the baseline assessment, and all patients were administered the PSS and the STAXI-2. In addition, we gathered data on hospitalisation or death events verified by medical records that occurred after the baseline visit (data not presented here; see Endrighi et al., 2016).

Interim assessments in clinic

After each telephone interview, patients who scored in the top or bottom 30% of the PSS distribution at that interview (approximately reflecting ≥15 and ≤9 stress scores) were scheduled for an interim assessment visit. At these visits, patients were required to complete the KCCQ and perform the 6MWT to accurately assess health and functional status. These assessments were restricted to individuals scoring in the top/bottom 30% of the PSS distribution to limit patient burden and because the larger BETRHEART study aims involved comparing patients with high versus low stress. The cut-off was chosen a priori to ensure an adequate number of patients at each assessment. Thus, all participants completed the KCCQ and the 6MWT at the baseline and 3-month assessment, and patients only differed in the number of KCCQ and 6MWT assessments over the course of the study based on PSS scores obtained at each biweekly telephone interview (see Results section).

Measures

Perceived Stress Scale

The PSS is a 10-item measure of generalised perceptions of stress in the preceding two weeks (Cohen et al., 1983). The scale is based on a model of stress as the extent to which individuals perceive events or situations as stressful, unpredictable or beyond their control (Folkman & Lazarus, 1988). The PSS demonstrates good reliability and validity criteria in a variety of settings and clinical and nonclinical populations (Arnold, Smolderen, Buchanan, Li, & Spertus, 2012; Cohen & Janicki-Deverts, 2012; Edmondson, Green, Ye, Halazun, & Davidson, 2014; Prior et al., 2016). Scores range from 0 to 40 with higher scores indicating greater stress perception. The PSS was administered to all patients at the Baseline assessment, in the bi-weekly telephone interviews, and at the 3-month assessment.

State Trait Anger Expression Inventory-2

The STAXI-2 (Spielberger et al., 1985) is a 57-item questionnaire that measures several anger dimensions: State Anger (15 items), Trait Anger (10 items), Anger Expression (16 items) and Anger Control (16 items). The State Anger items ask how the respondent feels ‘right now’ (e.g. ‘I feel furious’; ‘I feel like hitting someone’). The Trait Anger items ask how the respondent ‘generally’ feels or reacts to events. Respondents endorse each statement on a 4-point Likert scale that ranges from one (‘almost never’) to four (‘almost always’). Scores range from 15 to 60 with higher scores indicating greater anger. The STAXI-2’s validity and reliability have been established in healthy and in medical patient populations (Azevedo, Wang, Paulo, & Isabela, 2010; Spielberger et al., 1985).

In the present study analyses, the State Anger scale on the STAXI-2 was used since the present study was interested in examining changes and variability due to patients’ recent anger levels. The STAXI-2 was administered to all patients at the baseline assessment, in the bi-weekly telephone interviews and at the 3-month assessment.

HF health status and symptoms

The KCCQ is a validated 23-item scale that measures health status and symptoms burden in HF populations (Green et al., 2000). Patients are asked to indicate to what degree they were limited by symptoms of HF in the preceding two weeks. The KCCQ is sensitive to clinical changes in patients with HF and is a valid and reliable predictor of worsening HF clinical status and mortality and increased 30-day hospital readmissions (Dai et al., 2016; Flynn et al., 2012; Heidenreich et al., 2006; Spertus et al., 2005). The KCCQ Summary Score (range 0–100) was used; higher values indicate better health status and symptom profile. The KCCQ was administered to all patients at the baseline and 3-month assessment, and to the subset of patients who were scheduled for each bi-weekly interim clinical assessment.

Objectively assessed functional status

The six-minute Walk Test (6MWT) measures physical functional limitations among patients with HF (Balke, 1963; Guyatt et al., 1985). Scores consist of the distance the patient is able to walk on a level course for 6 min; longer distances reflect better functional status. A low score on the 6MWT is associated with poor prognosis in patients with HF, including rehospitalisations and mortality (Alahdab et al., 2009; Jehn et al., 2009). The 6MWT assessments were carried out on all patients at the baseline and 3-month month assessment, and to the subset of patients scheduled for each biweekly interim clinical assessment.

Covariates

The covariates age, sex, income, NYHA classification, ejection fraction, creatinine levels and history of hypertension were selected a priori based on established associations with HF exacerbations in the literature and for consistency with a previous published report from this cohort. Time was included as a continuous variable in all models. Missing values for creatinine (n=4) and income (n=1) were calculated using mean imputation of missing values; similar results were obtained if the five cases were treated as missing.

Statistical analyses

Baseline sample characteristics and descriptive data for predictor and outcome variables are reported as means and standard deviations (SD), or N and percentages. Data analysis used linear mixed models (LMMs) in SAS (PROC MIXED). In all models, we employed a random (subject-specific) intercept, and an autoregressive model of order 1 for the residuals within subjects. LMMs are well-suited to model multilevel repeated assessment data because they take into account clustering of data by patients (level 2) and are robust in modelling unequal number of observations at different time points (Littell, Milliken, Stroup, Wolfinger, & Schabenberger, 2006; Myers et al., 2012).

Separate models were fit for the perceived stress and state anger predictors. Model 1 tests associations between the predictor and outcomes adjusted for study covariates but not accounting for the effect of the other predictor. In Model 2, we examine whether associations between predictor and outcome variables are independent of each other (i.e. is perceived stress associated with study outcomes when controlling for the effect of anger, and vice versa?) by refitting the perceived stress models adjusted for the effect of anger, and the anger models adjusted for the effect of stress.

For both models a ‘Mean score’ and a ‘Deviation score’ for the predictor variables are used. The ‘Mean score’ (level 2 variable) is computed by aggregating over all available data for each patient. The ‘Deviation score’ (level 1) is computed as the difference between scores at each assessment and each patient’s ‘Mean score’ (i.e. the deviation from patient’s average score). A ‘Mean score’ effect reflects a person’s characteristic or between-subject effect. A ‘Deviation score’ effect denotes a within-subject association, i.e. whether patients report higher values on the dependent variable when reporting higher values than usual on the predictor variable. Associations between the predictor variable measured at baseline only (‘Baseline PSS score’ and ‘Baseline STAXI score’) and outcome variables measured at all time-points are also presented.

Parameter estimates of associations between predictor and outcome variables adjusted for study covariates are reported as B with standard error (SE). Parameter estimates of associations between the predictor variables, adjusted for the effect of one another, with the outcome variables are reported as B with (SE). Baseline only analyses are also adjusted for the baseline value of the respective outcomes.

Finally, we conducted a sensitivity analysis to rule out possible bias because unequal number of patients were given the 6MWT and KCCQ at each biweekly (i.e. every two week) interim in clinic assessment between baseline and 3 months. This analysis included only the baseline and 3-month assessments with complete data (thus excluding the interim clinic assessments).

Results

Demographic and clinical characteristics of the sample are presented in Table 1, and descriptive statistics for predictors and outcome variables over time are summarised in Table 2. The sample was majority African American, of low income and hypertensive. Only a small minority was in full or part time employment. Slightly more than half of the sample was classified as disabled for employment purposes. At baseline, PSS anger scores were correlated (r=0.51, p<0.001).

Table 1.

Baseline characteristics of study sample (N=144).

Variable Mean (SD) or %
Age (years) 57.51 (11.52)
% male sex 77.00%
BMI (kg/m2) 30.87 (7.50)
% smoking 27.20%
Ejection fraction (%) 23.10 (7.48)
Creatinine (mg/dL) 1.38 (0.71)
Systolic BP (mm/Hg) 121.05 (19.32)
% NYHA class
 II		 54.90%
 III 43.10%
 IV 2.10%
% Hypertension 79.20%
% Employment status
 Full or part time		 22.20%
 Disabled 51.40%
 Unemployed 4.90%
 Retired 21.50%
Income in US$
 <15k		 34.30%
 15–30k 26.60%
 >30–70k 39.20%
Race
 Caucasian		 29.20%
 African American 70.10%
 American Indian 0.70%
Open in a new tab

Abbreviations: BMI=body mass index; BP=blood pressure; NYHA=New York Heart Association.

Table 2.

Summary statistics of study measures.

Measure Baseline Wk-2 Wk-4 Wk-6 Wk-8 Wk-10 3-Month
PSS (0–40)
M (SD)		 13.16
(8.36)		 12.00
(8.22)		 12.29
(8.13)		 10.94
(7.35)		 11.62
(7.91)		 10.08
(7.71)		 10.04
(7.91)
STAXI-2 (15–60)
M (SD)		 16.78
(5.39)		 18.02
(6.59)		 18.02
(6.28)		 16.47
(4.32)		 17.16
(6.37)		 17.18
(6.59)		 16.70
(5.40)
KCCQa (0–100)
M (SD)		 71.64
(21.07)		 71.90
(20.98)		 76.14
(20.82)		 80.84
(18.49)		 79.76
(16.76)		 79.98
(16.83)		 79.61
(18.97)
6MWTa (feet)
M (SD)		 1062.35
(253.27)		 1098.26
(267.18)		 1182.26
(277.28)		 1167.32
(246.57)		 1185.53
(257.77)		 1143.33
(290.67)		 1160.04
(273.16)
Open in a new tab

Abbreviations: PSS=Perceived Stress Scale; STAXI-2=State Anger Score; KCCQ=Kansas City Cardiomyopathy Questionnaire-summary score; 6MWT=6 min Walk Test.

a

Between week-2 and week-10, PSS and STAXI-2 data were obtained from all patients via phone call interviews. Data on the KCCQ and 6MWT instead were obtained via clinic assessments on subsamples of patients scoring high and low on the PSS following each phone call interview (see text for details).

The number of patients that attended the baseline and the 3-month clinical assessment sessions were 144 and 126, respectively. The number of patients seen at the 2– 4-, 6-, 8- and 10-week interim clinic assessment was 81, 79, 73, 72 and 83, respectively. The vast majority of patients (89.5%) attended at least one interim clinic assessment in addition to baseline and 3-month follow-up, and therefore were represented at three or more assessments of KCCQ and 6MWT in the analysis. Average time between telephone interview and interim clinic assessment was 4 days (SD = 3).

Perceived stress and HF health status (KCCQ)

Patients with higher, compared to lower, ‘Mean PSS score’ reported poorer HF health status and symptoms based on KCCQ Summary Scores during the study (B = −1.61, SE=0.17, p<0.001; see Table 3, Model 1). ‘Deviation PSS score’ also had a robust association with poorer HF symptoms (B = −0.26, SE=0.08, p<0.001), indicating that when a patient reported greater stress than his/her average level (‘More than Usual’ vs. ‘Less than Usual’; Figure 1A) they experienced worsening HF health status. ‘Baseline PSS score’ was also associated with poorer symptoms and health status at later time points (B = −0.45, SE=0.15, p=0.004).

Table 3.

Associations between perceived stress, health and functional status.

KCCQ 6MWT
B (SE) B (SE)
Model 1
 Mean PSS score (between-subjects)
 Deviation PSS score (within-subjects)
 Baseline PSS score		 −1.61* (0.17)
−0.26* (0.08)
−0.45‡ (0.15)		 −5.20† (2.72)
−2.94* (0.77)
 0.04 (1.20)
Model 2
 Mean PSS score (between-subjects)
 Deviation PSS score (within-subjects)
 Baseline PSS score		 −1.57* (0.20)
−0.24‡ (0.08)
−0.30§ (0.15)		 −3.79 (3.82)
−2.93* (1.02)
 1.91 (1.54)
Open in a new tab

Note: Values are parameter estimate (B) and standard error (SE) of associations between perceived stress and: the Kansas City Cardiomyopathy Questionnaire (KCCQ-heart failure health status and symptoms) and the Six-minute Walk Test (6MWT-functional status). Parameter estimates in Model 1 are adjusted for the effect of age, sex, BMI, income, NYHA classification, ejection fraction, creatinine level and hypertension status. The ‘Baseline PSS score’ analysis is additionally adjusted for the baseline value of the respective outcome. Parameter estimates in Model 2 are adjusted for the effect of state anger (STAXI). In all analyses presented in Model 1 and 2, time is included as a continuous variable.

*

P≤0.001

†

p=0.07

‡

p≤0.01

§

p≤0.05.

Figure 1.

Figure 1.

Open in a new tab

Within-patient associations of perceived stress with HF symptoms (A) and functional status (B). Note. Data shown in (A) are KCCQ Summary Scores (Mean ±SEM). Data shown in (B) are distances walked on the 6MWT in feet (Mean±SEM). Data are aggregated over all assessments. ‘Less than Usual’ are assessments (ns = 338 and 297 assessments for KCCQ and 6MWT, respectively) where a participant’s PSS score was lower than the mean of their own (participant-specific) PSS scores. ‘More than Usual’ are assessments (ns = 296 and 256 assessments for KCCQ and 6MWT, respectively) where a participant’s PSS score was higher than the mean of their own (participant-specific) PSS score. KCCQ=Kansas City Cardiomyopathy Questionnaire-Summary Score; 6MWT=six-minute walk test.

Perceived stress and functional status (6MWT)

‘Mean PSS score’ was marginally associated with the 6MWT (B = −5.20, SE=2.72, p=0.058), but ‘Deviation PSS score’ was significantly associated with poorer functional status (B = −2.94, SE=0.77, p<0.001). This indicates that when a patient experienced greater perceived stress than his or her average (usual) level, that patient walked a shorter distance on the 6MWT (‘More than Usual’ vs. ‘Less than Usual’; Figure 1B). However, ‘Baseline PSS score’ was not associated with performance on the 6MWT assessed at later time points (B=0.04, SE=1.20, n.s.).

State anger and HF health status (KCCQ)

Patients with higher, compared to lower ‘Mean STAXI score’ had poorer HF symptom and health status during the study (B = −1.30, SE=0.29, p<0.001); see Table 4). However, ‘Deviation STAXI score’ was not associated with HF symptoms (p=0.30), indicating that at any time point during the study, patients with higher anger levels compared to their average did not experience worse HF symptoms. ‘Baseline STAXI score’ was also significantly associated with impairment in KCCQ scores at later time points (B = −0.45, SE=0.18, p=0.01).

Table 4.

Associations between State Anger, health and functional status

KCCQ 6MWT
B (SE) B (SE)
Model 1
 Mean STAXI score (between-subjects)
 Deviation STAXI score (within-subjects)		 −1.30* (0.29)
−0.10 (0.10)		 −2.41 (4.68)
−2.41† (1.21)
 Baseline STAXI score −0.45‡ (0.18) 2.07 (2.06)
Model 2
 Mean STAXI score (between-subjects)		 −0.07 (0.29) 3.29 (6.57)
 Deviation STAXI score (within-subjects) −0.05 (0.10) −1.83 (1.22)
 Baseline STAXI score −0.31§ (0.18) 0.42 (2.21)
Open in a new tab

Note: Values are parameter estimate (B) and standard error (SE) of associations between state anger (STAXI) and the Kansas City Cardiomyopathy Questionnaire (KCCQ-heart failure health status and symptoms) and the six-minute Walk Test (6MWT-functional status). Parameter estimates in Model 1 are adjusted for the effect of age, sex, BMI, income, NYHA classification, ejection fraction, creatinine level and hypertension status. The ‘Baseline STAXI score’ analysis is additionally adjusted for the baseline value of the respective outcome. Parameter estimates in Model 2 are adjusted for the effect of perceived stress. In all analyses presented in Model 1 and 2, time is included as a continuous variable.

*

p≤0.001

†

p≤0.05

‡

p≤0.01

§

p=0.09

State anger and functional status (6MWT)

‘Mean STAXI score’ was not significantly associated with patient’s functional status (p = 0.61). There was, however, a significant association between ‘Deviation STAXI score’ and patient’s functional status (B = −2.41, SE=1.21, p=0.046) indicating that when a patient experienced greater anger than his/her average level, the patient walked a shorter distance on the 6MWT. ‘Baseline STAXI score’ was not significantly associated with 6MWT distances assessed at later time points (p=0.32) (see Table 4).

Independent effects of perceived stress and state anger

Adjusted parameter estimates for associations between study predictors and outcome variables are summarised in Table 3 (Model 2; PSS adjusted for STAXI) and Table 4 (Model 2; STAXI adjusted for PSS). After adjusting for State Anger, ‘Mean PSS score’ was still significantly associated with KCCQ scores (B = −1.57, SE=0.20, p<0.001), but the ‘Mean PSS score’ association with 6MWT was reduced to not significant. However, adjustment for Anger scores did not eliminate the significant relationships between ‘Deviation PSS score’ and both KCCQ (B = −0.24, SE=0.08, p=0.003) and 6MWT (B = −2.93, SE=1.02, p=0.004). ‘Baseline PSS score’ also retained its association with KCCQ scores at later time points (B = −0.30, SE=0.15, p=0.04).

When adjusted for Perceived Stress, the relationship between ‘Mean STAXI score’ and the KCCQ became nonsignificant (p=0.81). Likewise, the association between ‘Deviation STAXI score’ and patient’s functional status (6MWT) became nonsignificant after adjustment for Perceived Stress (p=0.14). The relationship between ‘Baseline STAXI score’ and the KCCQ scores measured at later time points was also weakened (p=0.09).

Sensitivity analyses

To rule out possible bias attributable to unequal numbers of patients in the interim clinic assessments, sensitivity analyses were conducted. These analyses were limited to baseline and 3-month assessments data only, thus excluding the interim clinic assessments data, which was based on PSS scores (see Methods section for details). These analyses indicated that all the findings reported in Tables 3 and 4 remained significant.

Discussion

Results of this study demonstrate consistent associations of perceived stress with health and functional status (KCCQ and 6MWT performance), independent of traditional HF risk factors. In patients with HF, average perceived stress level was associated with poorer HF health and functional status, as were short-term elevations in stress. The magnitude of differences between high and low perceived stress categories appears to be clinically significant based on prior determinations in the literature (Bohannon & Crouch, 2017; Kelkar et al., 2016).

Anger relationships were somewhat weaker. Anger models not adjusted for perceived stress showed that patients with greater mean anger scores measured repeatedly over 3 months had poorer health status compared to patients with lower mean anger scores. Short-term elevations in anger were only associated with impaired functional status, indicating that compared to patient’s usual anger, increases in anger levels were associated with shorter distance walked on the 6MWT. These findings corroborate and extend prior findings indicating that patient distress can adversely impact HF symptoms and functional and health status (Baldasseroni et al., 2014; Bekelman et al., 2007; Jenner et al., 2009; Kop et al., 2011).

There are several mechanisms that might explain the present findings. First, it is possible that patients who have more functional impairments and more severe symptoms become more distressed, and that perceived stress and anger reflect more severe disease and functional impairment. However, this is unlikely to account for the present results since the study is prospective and involves multiple assessments as well as changes over time. It is also possible that perceived stress and anger reports reflect general negative affect and depression because of the overlap among stress, depressed affect and anger (Suls & Bunde, 2005). Prior cross-sectional evidence indicates that the 6MWT performance may be affected by symptoms and mood states (Kop et al., 2011), particularly depression (Gottlieb et al., 2009). Relationships between stress and HF-related health and symptoms and functional status might also be attributed to the shared features of perceived stress with HF symptoms (e.g. shortness of breath, fatigue, etc; Kop et al., 2011). The observed associations of perceived stress with symptoms and functional status are also likely to involve pathophysiological mechanisms (e.g. sympathetic nervous system activity, increased blood pressure, etc. that may directly affect the development and worsening of disease processes; Kop et al., 2011; Mostofsky, Penner, & Mittleman, 2014). Stress is also associated with risk factors such as reduced ability to follow complex medication and lifestyle regimens (Feenstra et al., 1998; Zaya et al., 2012), increased alcohol and cigarette consumption, and reduced physical activity (Rod, Gronbaek, Schnohr, Prescott, & Kristensen, 2009).

A prior report based on the present BETRHEART study cohort demonstrated that patients with higher versus lower average levels of perceived stress experienced more cardiovascular hospitalisations or death, but short-term elevations in stress did not increase the risk of a subsequent adverse event (Endrighi et al., 2016). In the present analyses, average levels, as well as short-term elevations in perceived stress, are both associated with worsened HF symptoms and functional status, with these associations still evident after adjustment for anger. In contrast, when significant anger findings were adjusted for Perceived Stress, associations with KCCQ and 6MWT were weakened or reduced to nonsignificant. Thus, perceived stress was more strongly associated with health and functional status than anger, and associations involving anger may largely be due to its overlap with perceived stress.

Given the role of stress and anger as potentially modifiable cardiovascular disease risk factors, few studies have examined the nature of the relationships between State Anger, Perceived Stress and their conjoint or independent effects on cardiovascular health in the same study (Kubzansky & Winning, 2016; Smith & Frohm, 1985; Suls & Bunde, 2005). Anger outbursts are a significant predictor of adverse cardiovascular events including myocardial infarction and acute coronary syndromes (Mostofsky et al., 2014). In the present study, anger was associated with HF functional status and symptoms, but these associations were not independent of Perceived Stress. Thus, the relationships between anger and indices of cardiovascular function and health status appear to depend on the particular clinical outcome measure being assessed. Prior studies also did not attempt to differentiate between the emotional state of anger and perceived stress, and it is possible that the anger measures used in prior studies also measure elements of perceived stress.

Study limitations

Assessments of 6MWT and KCCQ outcomes were taken for all patients at baseline and at 3 months and, additionally, at biweekly clinic visits in a subset of patients based on preceding PSS scores. This resulted in an unequal number of patients during the interim assessments. However, sensitivity analyses indicated that comparable findings were obtained for these outcomes using only baseline and 3-month assessments which were not subject to this limitation. Thus, the present findings were robust and not affected by potential bias due to unequal numbers of observations (Myers et al., 2012). Our sample was majority African American, male and had impaired LV function (systolic HF), which may limit generalisability to other populations. Finally, the present study findings may be limited to the particular stress and anger measures utilised in the study.

The design of this study using repeated assessments permitted examination of both between- and within-subject associations between stress/anger and HF symptoms and functional status. Although perceived stress was queried for the two weeks prior to each clinic/telephone assessment and functional status and HF symptoms were measured a few days after perceived stress assessments, we cannot rule out the possibility that associations can be accounted for by the reverse effects of functional status and/or HF symptoms on psychological state. Future research using lagged predictor variables may be useful in further addressing this issue. This point notwithstanding, elevations in stress over a short period of time may potentially impair patient’s health and functional status which may in turn adversely affect HF outcomes, including rehospitalisation.

Clinical implications

HF symptoms and functional status fluctuate over time, and these changes are predictive of HF worsening and/or improvement (Alahdab et al., 2009; Flynn et al., 2012; Heidenreich et al., 2006; Ingle et al., 2007; Pocock et al., 2006; Wegrzynowska-Teodorczyk et al., 2013; Zaya et al., 2012). Based on prior research, the differences in the present study between high and low within-patient deviations from average levels (see Figure 1) for the KCCQ scores (Flynn et al., 2012) and changes in 6MWT appear to be of clinically significant magnitude (Bohannon & Crouch, 2017; Kelkar et al., 2016). Therefore, an assessment of acute and sustained psychological factors that predict symptoms and functional status is clinically relevant since it can help in risk stratification, i.e. identifying ‘which patients are at risk’ and ‘when they are at risk’. Symptoms and functional status are also elements of patients’ perceived quality of life (Bekelman et al., 2007; Green et al., 2000), and enhancing quality of life is an important outcome for patients with chronic disease and for the healthcare system (Kaplan, 2003). The present findings also suggest several stress-related targets for intervention that might be aimed at reducing the burden of HF symptoms and their impact on functional status and reported quality of life (see Samartzis, Dimopoulos, Tziongourou, & Nanas, 2013). Behavioural interventions to reduce anger in combination with enhancing stress coping techniques may be targeted to patients who report experiencing frequent high levels of anger.

Future research

The present findings suggest several directions for future research. Although psychosocial interventions may be effective in improving quality of life in patients with HF (Samartzis et al., 2013), and there is some evidence that they are effective in stable cardiac patients (Blumenthal et al., 2016), it is not currently known whether such interventions delivered at times of increased distress will be effective in improving objective functional status or HF-related health status. In addition, future research is needed to determine whether such interventions would be effective in preventing and/or reducing HF exacerbations such as hospitalisations or other clinical HF outcomes. Depending on the outcome being examined (i.e. reported health, functional status or adverse events), improving either chronic stress levels and/or preventing short-term increases in distress over time may be most effective in reducing adverse HF outcomes.

Acknowledgments

Funding

This work was supported by National Heart Lung and Blood Institute grant RO1-HL085730.

Footnotes

Disclosure statement

No potential conflict of interest was reported by the authors.

References

  1. Alahdab MT, Mansour IN, Napan S, & Stamos TD (2009). Six minute walk test predicts long-term all-cause mortality and heart failure rehospitalization in African-American patients hospitalized with acute decompensated heart failure. Journal of Cardiac Failure, 15(2), 130–135. doi: 10.1016/j.cardfail.2008.10.006 [DOI] [PubMed] [Google Scholar]
  2. Albert NM, Fonarow GC, Abraham WT, Gheorghiade M, Greenberg BH, Nunez E, … Young JB (2009). Depression and clinical outcomes in heart failure: An OPTIMIZE-HF analysis. American Journal of Medicine, 122(4), 366–373. doi:S0002–9343(08)01176–5 [pii]; 10.1016/j.amjmed.2008.09.046 [DOI] [PubMed] [Google Scholar]
  3. Arnold SV, Smolderen KG, Buchanan DM, Li Y, & Spertus JA (2012). Perceived stress in myocardial infarction: Long-term mortality and health status outcomes. Journal of American College of Cardiology, 60(18), 1756–1763. doi: 10.1016/j.jacc.2012.06.044 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Azevedo FBD, Wang Y-P, A. ,CG, Paulo LA, & Isabela BM (2010). Application of the Spielberger’s State-Trait Anger Expression Inventory in clinical patients. Arquivos de Neuro-Psiquiatria, 68(2), 231–234. doi: 10.1590/S0004-282X2010000200015 [DOI] [PubMed] [Google Scholar]
  5. Baldasseroni S, Pratesi A, Orso F, Foschini A, Marella AG, Bartoli N, … Tarantini F (2014). Age-related impact of depressive symptoms on functional capacity measured with 6-minute walking test in coronary artery disease. European Journal of Preventive Cardiology, 21(5), 647–654. doi: 10.1177/2047487313494581 [DOI] [PubMed] [Google Scholar]
  6. Balke B (1963). A simple field test for the assessment of physical fitness. Report of Civil Aeromedical Research Institute (US), 63, 1–8. [PubMed] [Google Scholar]
  7. Bekelman DB, Havranek EP, Becker DM, Kutner JS, Peterson PN, Wittstein IS, … Dy SM (2007). Symptoms, depression, and quality of life in patients with heart failure. Journal of Cardiac Failure, 13(8), 643–648. doi: 10.1016/j.cardfail.2007.05.005 [DOI] [PubMed] [Google Scholar]
  8. Blumenthal JA, Sherwood A, Smith PJ, Watkins L, Mabe S, Kraus WE, … Hinderliter A (2016). Enhancing cardiac rehabilitation with stress management training: A randomized, clinical efficacy trial. Circulation, 133(14), 1341–1350. doi: 10.1161/CIRCULATIONAHA.115.018926 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Bohannon RW, & Crouch R (2017). Minimal clinically important difference for change in 6-minute walk test distance of adults with pathology: A systematic review. Journal of Evaluation in Clinical Practice, 23(2), 377–381. doi: 10.1111/jep.12629 [DOI] [PubMed] [Google Scholar]
  10. Chida Y, & Steptoe A (2009). The association of anger and hostility with future coronary heart disease: A meta-analytic review of prospective evidence. Journal of American College of Cardiology, 53(11), 936–946. doi: 10.1016/j.jacc.2008.11.044 [DOI] [PubMed] [Google Scholar]
  11. Cohen S, & Janicki-Deverts D (2012). Who’s stressed? Distributions of psychological stress in the United States in probability samples from 1983, 2006, and 20091. Journal of Applied Social Psychology, 42(6), 1320–1334. doi: 10.1111/j.1559-1816.2012.00900.x [DOI] [Google Scholar]
  12. Cohen S, Kamarck T, & Mermelstein R (1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24(4), 385–396. [PubMed] [Google Scholar]
  13. Coles AH, Fisher K, Darling C, Yarzebski J, McManus DD, Gore JM, … Goldberg RJ (2014). Long-term survival for patients with acute decompensated heart failure according to ejection fraction findings. American Journal of Cardiology, 114(6), 862–868. doi: 10.1016/j.amjcard.2014.06.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Dai S, Manoucheri M, Gui J, Zhu X, Malhotra D, Li S, … Jiang H (2016). Kansas City cardiomyopathy questionnaire utility in prediction of 30-day readmission rate in patients with chronic heart failure. Cardiology Research and Practice, 2016, 4571201. doi: 10.1155/2016/4571201 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Dharmarajan K, Hsieh AF, Lin Z, Bueno H, Ross JS, Horwitz LI, … Krumholz HM (2013). Diagnoses and timing of 30-day readmissions after hospitalization for heart failure, acute myocardial infarction, or pneumonia. JAMA, 309(4), 355–363. [pii]; 10.1001/jama.2012.216476 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Edmondson D, Green P, Ye S, Halazun HJ, & Davidson KW (2014). Psychological stress and 30-day all-cause hospital readmission in acute coronary syndrome patients: An observational cohort study. PLoS One, 9(3), e91477. doi: 10.1371/journal.pone.0091477 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Endrighi R, Waters AJ, Gottlieb SS, Harris KM, Wawrzyniak AJ, Bekkouche NS, … Krantz DS (2016). Psychological stress and short-term hospitalisations or death in patients with heart failure. Heart. 102, 1820–1825. doi: 10.1136/heartjnl-2015-309154 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Feenstra J, Grobbee DE, Jonkman FA, Hoes AW, & Stricker BH (1998). Prevention of relapse in patients with congestive heart failure: The role of precipitating factors. Heart, 80(5), 432–436. doi: 10.1136/hrt.80.5.432 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Flynn KE, Lin L, Moe GW, Howlett JG, Fine LJ, Spertus JA, … Weinfurt KP (2012). Relationships between changes in patient-reported health status and functional capacity in outpatients with heart failure. American Heart Journal, 163(1), 88–94. doi:S0002–8703(11)00713–7 [pii]; 10.1016/j.ahj.2011.09.027 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Folkman S, & Lazarus RS (1988). The relationship between coping and emotion: Implications for theory and research. Social Science and Medicine, 26(3), 309–317. [DOI] [PubMed] [Google Scholar]
  21. Gabbay FH, Krantz DS, Kop WJ, Hedges SM, Klein J, Gottdiener JS, & Rozanski A (1996). Triggers of myocardial ischemia during daily life in patients with coronary artery disease: Physical and mental activities, anger and smoking. Journal of American College of Cardiology, 27(3), 585–592. doi: 10.1016/0735-1097(95)00510-2 [DOI] [PubMed] [Google Scholar]
  22. Gottlieb SS, Kop WJ, Ellis SJ, Binkley P, Howlett J, O’Connor C, … Cooper L (2009). Relation of depression to severity of illness in heart failure (from Heart Failure And a Controlled Trial Investigating Outcomes of Exercise Training [HF-ACTION]). American Journal of Cardiology, 103(9), 1285–1289. doi: 10.1016/j.amjcard.2009.01.025 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Green CP, Porter CB, Bresnahan DR, & Spertus JA (2000). Development and evaluation of the Kansas City Cardiomyopathy Questionnaire: A new health status measure for heart failure. Journal of American College of Cardiology, 35(5), 1245–1255. doi:S0735–1097(00)00531–3 [pii] doi: 10.1016/S0735-1097(00)00531-3 [DOI] [PubMed] [Google Scholar]
  24. Guyatt GH, Sullivan MJ, Thompson PJ, Fallen EL, Pugsley SO, Taylor DW, & Berman LB (1985). The 6-minute walk: A new measure of exercise capacity in patients with chronic heart failure. Canadian Medical Association Journal, 132(8), 919–923. [PMC free article] [PubMed] [Google Scholar]
  25. Heidenreich PA, Spertus JA, Jones PG, Weintraub WS, Rumsfeld JS, Rathore SS, … Consortium COR (2006). Health status identifies heart failure outpatients at risk for hospitalization or death. Journal of American College of Cardiology, 47(4), 752–756. doi: 10.1016/j.jacc.2005.11.021 [DOI] [PubMed] [Google Scholar]
  26. Ingle L, Rigby AS, Carroll S, Butterly R, King RF, Cooke CB, … Clark AL (2007). Prognostic value of the 6 min walk test and self-perceived symptom severity in older patients with chronic heart failure. European Heart Journal, 28(5), 560–568. doi: 10.1093/eurheartj/ehl527 [DOI] [PubMed] [Google Scholar]
  27. Jehn M, Schmidt-Trucksass A, Schuster T, Weis M, Hanssen H, Halle M, & Koehler F (2009). Daily walking performance as an independent predictor of advanced heart failure: Prediction of exercise capacity in chronic heart failure. American Heart Journal, 157(2), 292–298. doi: S0002–8703(08)00887–9 [pii]; 10.1016/j.ahj.2008.10.006 [DOI] [PubMed] [Google Scholar]
  28. Jenner RC, Strodl ES, & Schweitzer RD (2009). Anger and depression predict hospital use among chronic heart failure patients. Australian Health Review, 33(4), 541–548. doi: 10.1071/AH090541 [DOI] [PubMed] [Google Scholar]
  29. Joffe SW, Webster K, McManus DD, Kiernan MS, Lessard D, Yarzebski J, … Goldberg RJ (2013). Improved survival after heart failure: A community-based perspective. Journal of American Heart Association, 2(3), e000053. doi: 10.1161/jaha.113.000053 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Kaplan RM (2003). The significance of quality of life in health care. Quality of Life Research, 12(Suppl 1), 3–16. [DOI] [PubMed] [Google Scholar]
  31. Keith F, Krantz DS, Chen R, Harris KM, Ware CM, Lee AK, … Gottlieb SS (2017). Anger, hostility, and hospitalizations in patients with heart failure. Health Psychology, 36(9), 829–838. doi: 10.1037/hea0000519 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Kelkar AA, Spertus J, Pang P, Pierson RF, Cody RJ, Pina IL, … Butler J (2016). Utility of patient-reported outcome instruments in heart failure. JACC Heart Failure, 4(3), 165–175. doi: 10.1016/j.jchf.2015.10.015 [DOI] [PubMed] [Google Scholar]
  33. Kop WJ, Synowski SJ, & Gottlieb SS (2011). Depression in heart failure: Biobehavioral mechanisms. Heart Failure Clinics, 7(1), 23–38. doi:S1551–7136(10)00106–6 [pii]; 10.1016/j.hfc.2010.08.011 [DOI] [PubMed] [Google Scholar]
  34. Krantz DS, Olson MB, Francis JL, Phankao C, Bairey Merz CN, Sopko G, … Matthews KA (2006). Anger, hostility, and cardiac symptoms in women with suspected coronary artery disease: The Women’s Ischemia Syndrome Evaluation (WISE) Study. Journal of Womens Health (Larchmt), 15(10), 1214–1223. doi: 10.1089/jwh.2006.15.1214 [DOI] [PubMed] [Google Scholar]
  35. Kubzansky L, & Winning A (2016). Emotions and health. In Barrett LF, Lewis M, & Haviland-Jones JM (Eds.), Handbook of emotions (4th ed., pp. 613–628). New York: The Guilford Press. [Google Scholar]
  36. Littell RC, Milliken GA, Stroup WW, Wolfinger RD, & Schabenberger O (2006). SAS® for mixed models (2nd ed.). Cary, NC: SAS Institute Inc. [Google Scholar]
  37. Mittleman MA, Maclure M, Sherwood JB, Mulry RP, Tofler GH, Jacobs SC, … Muller JE (1995). Triggering of acute myocardial infarction onset by episodes of anger. Determinants of Myocardial Infarction Onset Study Investigators. Circulation, 92(7), 1720–1725. doi: 10.1161/01.CIR.92.7.1720 [DOI] [PubMed] [Google Scholar]
  38. Mostofsky E, Maclure M, Tofler GH, Muller JE, & Mittleman MA (2013). Relation of outbursts of anger and risk of acute myocardial infarction. American Journal of Cardiology, 112(3), 343–348. doi: 10.1016/j.amjcard.2013.03.035 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Mostofsky E, Penner EA, & Mittleman MA (2014). Outbursts of anger as a trigger of acute cardiovascular events: A systematic review and meta-analysis. European Heart Journal, 35(21), 1404–1410. doi: 10.1093/eurheartj/ehu033 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Myers ND, Brincks AM, Ames AJ, Prado GJ, Penedo FJ, & Benedict C (2012). Multilevel modeling in psychosomatic medicine research. Psychosomatic Medicine, 74(9), 925–936. doi:PSY.0b013e3182736971 [pii]; 10.1097/PSY.0b013e3182736971 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Pelle AJ, Gidron YY, Szabo BM, & Denollet J (2008). Psychological predictors of prognosis in chronic heart failure. Journal of Cardiac Failure, 14(4), 341–350. doi:S1071–9164(08)00022–5 [pii]; 10.1016/j.cardfail.2008.01.004 [DOI] [PubMed] [Google Scholar]
  42. Perlman LV, Ferguson S, Bergum K, Isenberg EL, & Hammarsten JF (1971). Precipitation of congestive heart failure: Social and emotional factors. Annals of Internal Medicine, 75(1), 1–7. [DOI] [PubMed] [Google Scholar]
  43. Pocock SJ, Wang D, Pfeffer MA, Yusuf S, McMurray JJ, Swedberg KB, … Granger CB (2006). Predictors of mortality and morbidity in patients with chronic heart failure. European Heart Journal, 27(1), 65–75. doi:ehi555 [pii]; 10.1093/eurheartj/ehi555 [DOI] [PubMed] [Google Scholar]
  44. Prior A, Fenger-Gron M, Larsen KK, Larsen FB, Robinson KM, Nielsen MG, … Vestergaard M (2016). The association between perceived stress and mortality among people with multimorbidity: A prospective population-based cohort study. American Journal of Epidemiology, 184(3), 199–210. doi: 10.1093/aje/kwv324 [DOI] [PubMed] [Google Scholar]
  45. Richardson S, Shaffer JA, Falzon L, Krupka D, Davidson KW, & Edmondson D (2012). Meta-analysis of perceived stress and its association with incident coronary heart disease. American Journal of Cardiology, 110(12), 1711–1716. doi:S0002–9149(12)01929–7 [pii]; 10.1016/j.amjcard.2012.08.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Rod NH, Andersen I, & Prescott E (2011). Psychosocial risk factors and heart failure hospitalization: A prospective cohort study. American Journal of Epidemiology, 174(6), 672–680. doi: kwr144 [pii]; 10.1093/aje/kwr144 [DOI] [PubMed] [Google Scholar]
  47. Rod NH, Gronbaek M, Schnohr P, Prescott E, & Kristensen TS (2009). Perceived stress as a risk factor for changes in health behaviour and cardiac risk profile: A longitudinal study. Journal of Internal Medicine, 266(5), 467–475. doi:JIM2124 [pii]; 10.1111/j.1365-2796.2009.02124.x [DOI] [PubMed] [Google Scholar]
  48. Rumsfeld JS, Havranek E, Masoudi FA, Peterson ED, Jones P, Tooley JF, … Spertus JA (2003). Depressive symptoms are the strongest predictors of short-term declines in health status in patients with heart failure. Journal of American College of Cardiology, 42(10), 1811–1817. doi:S0735109703011781 [pii] doi: 10.1016/j.jacc.2003.07.013 [DOI] [PubMed] [Google Scholar]
  49. Russell MA, Smith TW, & Smyth JM (2016). Anger expression, momentary anger, and symptom severity in patients with chronic disease. Annals of Behavioral Medicine, 50(2), 259–271. doi: 10.1007/s12160-015-9747-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Samartzis L, Dimopoulos S, Tziongourou M, & Nanas S (2013). Effect of psychosocial interventions on quality of life in patients with chronic heart failure: A meta-analysis of randomized controlled trials. Journal of Cardiac Failure, 19(2), 125–134. doi:S1071–9164(12)01395–4 [pii]; 10.1016/j.cardfail.2012.12.004 [DOI] [PubMed] [Google Scholar]
  51. Sherwood A, Blumenthal JA, Hinderliter AL, Koch GG, Adams KF Jr., Dupree CS, … O’Connor CM (2011). Worsening depressive symptoms are associated with adverse clinical outcomes in patients with heart failure. Journal of American College of Cardiology, 57(4), 418–423. doi: 10.1016/j.jacc.2010.09.031 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Sherwood A, Blumenthal JA, Trivedi R, Johnson KS, O’Connor CM, Adams KF Jr., … Hinderliter AL (2007). Relationship of depression to death or hospitalization in patients with heart failure. Archives of Internal Medicine, 167(4), 367–373. doi: 10.1001/archinte.167.4.367 [DOI] [PubMed] [Google Scholar]
  53. Smith OR, Pedersen SS, Van Domburg RT, & Denollet J (2008). Symptoms of fatigue and depression in ischemic heart disease are driven by personality characteristics rather than disease stage: A comparison of CAD and CHF patients. European Journal of Cardiovascular Prevention & Rehabilitation, 15(5), 583–588. doi: 10.1097/HJR.0b013e3283083e17 [DOI] [PubMed] [Google Scholar]
  54. Smith TW, & Frohm KD (1985). What’s so unhealthy about hostility? Construct validity and psychosocial correlates of the Cook and Medley Ho scale. Health Psychology, 4(6), 503–520. [DOI] [PubMed] [Google Scholar]
  55. Spertus J, Peterson E, Conard MW, Heidenreich PA, Krumholz HM, Jones P, … Rumsfeld JS (2005). Monitoring clinical changes in patients with heart failure: A comparison of methods. American Heart Journal, 150(4), 707–715. doi:S0002–8703(04)00906–8 [pii]; 10.1016/j.ahj.2004.12.010 [DOI] [PubMed] [Google Scholar]
  56. Spielberger CD, Johnson EH, Russell SF, Crane RJ, Jacobs GA, & Worden TJ (1985). The experience and expression of anger: Construction and validation of an anger expression scale. In Chesney MA & Rosenman RH (Eds.), Anger and hostility in cardiovascular and behavioral disorders (pp. 5–30). New York: Hemisphere/McGraw-Hill. [Google Scholar]
  57. Suls J, & Bunde J (2005). Anger, anxiety, and depression as risk factors for cardiovascular disease: The problems and implications of overlapping affective dispositions. Psychological Bulletin, 131(2), 260–300. doi: 10.1037/0033-2909.131.2.260 [DOI] [PubMed] [Google Scholar]
  58. Wegrzynowska-Teodorczyk K, Rudzinska E, Lazorczyk M, Nowakowska K, Banasiak W, Ponikowski P, … Jankowska EA (2013). Distance covered during a six-minute walk test predicts long-term cardiovascular mortality and hospitalisation rates in men with systolic heart failure: An observational study. Journal of Physiotherapy, 59(3), 177–187. doi: 10.1016/S1836-9553(13)70182-6 [DOI] [PubMed] [Google Scholar]
  59. Zaya M, Phan A, & Schwarz ER (2012). Predictors of re-hospitalization in patients with chronic heart failure. World Journal of Cardiology, 4(2), 23–30. doi: 10.4330/wjc.v4.i2.23 [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES