Relationship of Depression, Anxiety, and Social Isolation to Chronic

Heart Failure Outpatient Mortality

Erika Friedmann, PhDa,b; Sue A. Thomas, RN, PhD, FAAN a; Fang Liu, MSa; Patricia G. Morton, RN,
PhD, CANP, FAAN a; Deborah Chapa, RN, MS, CRNPa; Stephen S. Gottlieb, MD, FACCc
Am Heart J. 2006;152(5):940.e1-940.e8. 2006 Mosby, Inc.
Posted 01/30/2007

Abstract and Introduction

Abstract
Objectives: The Psychosocial Factors Outcome Study (PFOS) investigated the prevalence of
depression and anxiety and the relationship of psychosocial factors to mortality in outpatients with heart
failure (HF).
Background: Considerable evidence links psychosocial factors to coronary heart disease mortality and
sudden cardiac death (SCD). The contribution of psychosocial factors independent of disease severity
to HF outpatient mortality is not well elucidated.
Methods: Patients (N = 153) from 20 Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT) sites
participated in the PFOS. SCD-HeFT provided demographic, medical history, and cardiac data.
Participants completed questionnaires to assess psychosocial status at PFOS entry.
Results: Depression and anxiety were common in HF outpatients (36% Beck Depression Inventory-II
13; 45% State Trait Anxiety Inventory 40). Depression, anxiety, and social support amount did not
differ in the SCD-HeFT treatment groups: implantable cardioverter defibrillator, amiodarone, and placebo
medication. Fifteen (9.8%) patients died during mean follow-up at 23.6 months (SD = 8.2). In Cox
regression controlling for treatment, depression, anxiety, and social isolation separately predicted
mortality; perceived HF-specific functional status did not. Depression (ln) [P = .04, hazard ratio (HR) =
1.81] and social isolation (P = .04, HR = 2.25), but not anxiety, predicted mortality independent of
demographics, clinical predictors, and treatment. When simultaneously including significant
demographic, clinical, and psychosocial predictors and treatment groups, depression (ln) (P = .022, HR
= 2.2) and social isolation (P = .094, HR = 1.75) predicted mortality. All-cause mortality was 12% for
depressed patients and 9% for others.
Conclusion: This study finds a high prevalence of anxiety and confirms the high prevalence of
depression in the HF outpatient population. Depression and social isolation predicted mortality
independent of demographic and clinical status in HF outpatients.
Introduction
Considerable evidence links psychosocial factors such as depression,[1-5] anxiety,[2,4-6] life stress,[2,7-9] social
isolation,[3-5,10] and lack of pet ownership[11-13] to coronary heart disease (CHD) mortality and sudden
cardiac death (SCD). Previous studies have documented an association between depression and
mortality in outpatients with heart failure (HF),[14-17] inpatients with HF,[18] and hospitalized patients with HF.
[19,20]
The importance of the other psychosocial factors in mortality of outpatients with HF who are not
awaiting cardiac transplantation has not been well evaluated.
The Psychosocial Factors Outcome Study (PFOS) in Sudden Cardiac Death was the first study to
examine the independent contributions of psychosocial factors and disease severity to mortality in HF
outpatients. We hypothesized that depression, anxiety, life stress, social isolation, and lack of pet
ownership, the psychosocial factors that are associated with coronary heart disease prognosis, would
also be related to mortality independent of demographic and clinical status for HF outpatients.

Methods
All Sudden Cardiac Death Heart Failure Trial (SCD-HeFT) sites were invited to participate in the PFOS;
20 sites completed human subjects approval, signed contracts, and contributed at least 1 patient. Sites
participating in PFOS were located in the United States (n = 18), Canada (n = 2), and New Zealand (n =
1). PFOS started after SCD-HeFT recruitment had begun. SCD-HeFT participants who were
randomized to the quality of life substudy of SCD-HeFT were excluded because of concerns about
patient burden. Patients who entered SCD-HeFT after PFOS began or returned for SCD-HeFT follow-up
and who could read and write English were eligible to participate. Separate informed consent was
required for the PFOS.

SCD-HeFT participants had New York Heart Association (NYHA) class II or III HF and left ventricular
ejection fraction (EF) 35%, and were randomly assigned to implantable cardioverter defibrillator (ICD),
amiodarone, or placebo medication. SCD-HeFT inclusion and exclusion criteria and recruitment
methods are published elsewhere.[21] SCD-HeFT provided demographic and health data.
Participants completed questionnaires to assess psychosocial status including depression, anxiety,
social support, pet ownership, and life stress at PFOS entry after 1 month and every 6 months.
Perceptions of impairment in HF-specific functional status was assessed before randomization and at 6month intervals.
Instruments
The State Trait Anxiety Inventory (STAI), which has well-established validity and reliability,[22] was used to
assess anxiety. Patients with scores >40 are classified as anxious.[23] Both state and trait anxiety were
assessed. Trait anxiety reflects how a person generally feels. State anxiety, which reflects how the
patient was feeling at the time of the assessment, was used for all analyses. The Beck Depression
Inventory-2 (BDI-II) was used to measure depressive symptoms. The reliability and validity of the BDI-II
are well established.[24] Depression severity was categorized according to the BDI-II manual:[24] absent
(0-12), mild (13-19), moderate (20-28), or severe (29). The Social Support Questionnaire-6 (SSQ-6)
was used to measure social support. The SSQ-6 provides a list of 6 situations potentially requiring social
support and asked the respondent who they could rely on for help (amount) and their perceived
satisfaction with help (satisfaction) in each situation. The reliability and validity of the SSQ-6 were
reported previously.[25] The Social Readjustment Rating Scale, a checklist of recent life events, was used
to evaluate patients' life stress in the last 6 to 12 months and expectations of stress in the next 6
months. Individuals check each of 46 life events, positive or negative, occurring during the specified
period.[26] Test-retest reliability was established.[8] Pet ownership was assessed with 5 demographic-type
questions.[27]
Perceived impairment in HF-specific functional status was assessed with the Minnesota Living with HF
Scale (MLWHF). It includes 21 questions measuring the patient's perception of physical and
psychosocial impact of HF-related symptoms and treatments in adults.[28,29] Higher scores indicate that
patients perceive more impairment.
Data Analysis
2 Tests and one-way analyses of variance were used to examine differences in demographic and
medical history variables among patients in the SCD-HeFT treatments. Before multivariate analyses,
skew and kurtosis of predictors were examined. Cox proportional hazards regression analysis was used
to examine the contributions of predictors to all-cause mortality. Multivariate analyses used SAS version
9.1 (SAS Institute, Cary, NC). The analysis began with a series of bivariate analyses of the contribution
of each predictor to survival after controlling for assigned treatment. The demographic and clinical
variables that predicted survival were included in a series of separate survival analyses with each of the
significant psychosocial variables from the initial Cox model and with impairment in function. The
psychosocial variables that predicted survival with P < .20 in the initial Cox model were then combined
in one survival analysis. An iterative approach was used to create the most parsimonious final model by
combining the psychosocial and demographic and clinical variables and eliminating predictors from
previous models with P > .20. Dummy variables comparing each treatment group with the placebo
group were retained in all analyses because of differences observed in the larger SCD-HeFT clinical
trial.[21] A priori expectations that anxiety, depression, functional impairment, being diabetic, older, NYHA
class III HF, having a history of myocardial infarction, and receiving placebo medication would be
associated with worse survival and higher EF, and social support would be associated with better
survival were tested with 1-tailed tests. Race and sex were tested with 2-tailed tests.

Results
Sites contributed from 1 to 19 patients (median = 5). Thirteen sites contributed 7 patients.
Patients were largely men and white. Ages ranged from 35.5 to 85.1 years. Fifteen patients died during
a follow-up of 1.6 to 49.2 months, a 9.8% all-cause mortality rate. Survivors were followed up for at least
1 year (median 27 months). SCD-HeFT mortality was 26.4% with a minimum follow-up of 24 months
(median 54.5 months). Patients (N = 153) entered PFOS 0 to 54.4 months after they entered the SCDHeFT (mean 17.5 months; SD 16.0). Lag between SCD-HeFT and PFOS entries was <1 month for 47
patients, 1-12 months for 21 patients, and >1 year for 85 patients. For demographic and medical history
information for PFOS and SCD-HeFT, see Table 1 .

Summaries of psychosocial measures at initial assessment are in Table 2 . Skewed distributions of

depression, social support amount, social support satisfaction, past life events, and future life events
necessitated logarithmic transformations (ln), which generated normal distributions. Transformed
variables [identified with "(ln)" suffix] were used for parametric analyses.
The demographic, clinical, and psychosocial characteristics of the PFOS participants from the SCDHeFT treatment groups did not differ significantly. In univariate analyses, mortality did not differ
significantly among PFOS participants in the treatment (placebo 6/51, 11.8%; amiodarone 3/45, 6.7%;
and ICD 6/57, 10.5%). Detecting differences in treatment group mortality required a larger sample.[21]
Psychosocial Distress
BDI-II scores at initial PFOS assessment ranged from 0 to 39. Symptoms of depression were common
in the sample; 36% of the patients reported at least mild depression. Mortality among patients who were
depressed was 12% compared with 9% for those who were not.
State anxiety scores at initial PFOS assessment ranged from 20 to 63. Anxiety was common in the
sample; 45% of the patients report significant state anxiety. Mortality among patients who were anxious
was 12% compared with 8% for those who were not. Mortality was highest for patients who were
depressed and anxious (14%), intermediate for patients who were either depressed or anxious (10%),
and lowest for patients who were neither depressed nor anxious (8%).
Social support amount as scored on the SSQ-6 at initial PFOS assessment ranged from 0 to 54.
Mortality was 7% among patients above the median on social support and 14% among those below it.
Mortality was 8% among patients with high social support and without depression or anxiety, 16% for
socially isolated patients with anxiety or depression, and 20% for socially isolated patients with anxiety
and depression.
Ischemia and Psychosocial Distress
More than half of the participants (54.9%) experienced HF from ischemic cardiomyopathy. Depression
[t(134) = -.567, P = .572], anxiety [t(147) = .404, P = .687], and social support [t(146) = -1.604, P = .111]
did not differ between patients with ischemic and nonischemic HF. Differences in survival according to
psychosocial status do not appear to reflect ischemic status.
Perceived Impairment in HF-specific Functional Status
At initial PFOS assessment, total MLWHF scores, including both psychosocial and physiologic
components, ranged from 0 to 91 (mean 40 1.97). Scores did not differ among treatment groups (F2,132
= 0.30, P = .74). Mortality was 10.3% for patients with functional impairment above the median and
7.5% for those below it.
Relationship of Psychosocial Distress to Perceived Functional Status
At initial PFOS assessment, perceived impairment in functional status was correlated with depression
[r(138) = 0.438, P < .001] and state anxiety [r(131) = 0.421, P < .001] but not social support amount
[r(144) = .039, P = .60]. Patients with more functional impairment were more depressed and more
anxious.
Time to Event Analysis
The contribution of each demographic, clinical, and psychosocial variable, lag time and perceived HFspecific functional status to mortality beyond the effect of treatment group was examined via Cox
regression (see Table 3 ). In each analysis, the dummy variables for treatment group with placebo as the
reference group and the predictor were entered simultaneously. Demographic and clinical predictors of
mortality (P < .20) included older age, atrial fibrillation/flutter, diabetes, lower EF, and NYHA class.
Psychosocial predictors of mortality (P < .20) included anxiety, depression, and social isolation.
Perceived impairment in HF specific functional status predicted mortality (P < .2) and was included with
psychosocial factors as it assesses perceptions rather than physical function or clinical characteristics.
Hazard function graphs for the relationship of depression, anxiety, and social support to mortality are
included in Figure 1, Figure 2, and Figure 3.

Figure 1.
Cumulative hazard functions comparing patients who scored below the median (<10) and those who are
above the median (10) on the BDI-II.

Figure 2.
Cumulative hazard functions comparing patients who are anxious (state anxiety score on the STAI >40)
and those who are not.

Figure 3.
Cumulative hazard functions comparing patients who report below median and above median amounts
of social support on the SSQ-6.

The demographic and clinical predictors of mortality (P < .20) beyond the effect of treatment group were
entered simultaneously with treatment group into a Cox regression model. Atrial fibrillation/flutter and
NYHA class were significant independent predictors of mortality (see Table 4 ). When lag time was
entered into the model, it made no significant contribution to survival or substantive change in the
findings. Patients with atrial fibrillation or flutter and NYHA class III HF were significantly more likely to
die than those with sinus rhythm and NYHA class II HF, independent of the effects of treatment group
and diabetic status.
Each of the 3 psychosocial variables that predicted mortality beyond the effect of treatment group and
MLWHF was entered with atrial fibrillation/flutter, NYHA class, and treatment group into a Cox
regression model. In these models, depression [n = 136, hazard ratio (HR) 2.25, CI 1.047-4.728, P = .
04) and social isolation (n = 145, HR 0.55, CI 0.418-0.949, P = .036) were significant independent
predictors of mortality; anxiety (n = 149, HR 1.03, CI 0.989-1.072, P = .12) and MLWHF (n = 135, HR
1.02, CI 0.990-1.041, P = .16) were not.
The psychosocial, demographic, and clinical variables predicting mortality (P < .20) from the separate
analyses were then combined as predictors of mortality. A series of analyses resulted in the final model
(see Table 5 ). In this model, depression was a significant independent predictor of mortality, and social
isolation tended to predict mortality, independent of demographic and clinical status, treatment group,
and each other. When lag time was entered into the model, it made no significant contribution to survival
or substantive change in the findings. Patients who were more depressed and more socially isolated
were more likely to die.

Discussion
This study finds a high prevalence of anxiety and confirms the high prevalence of depression in the HF
outpatient population. Depression and social isolation predicted mortality independent of demographic
and clinical status and each other in this population. Whether depression and anxiety reflect worse HF
or cause changes that increase mortality is not known. The psychosocial variables are not proxies for
ischemia or ejection fraction. The findings of the present study suggest that these psychosocial factors
could be causative. Because the prevalence of depression and anxiety is high, this study also suggests
that interventions to improve psychological status could impact mortality in HF outpatients.
Prevalence of Depression and Anxiety
In this chronic HF outpatient population, depression was common. The current study confirms our earlier
work demonstrating the high prevalence of depression in the outpatient HF population.[30] Many studies

documented depression in acute HF,[18,20] and, increasingly, studies show depression in a large
proportion of patients with chronic HF.[31] The prevalence of depression in this study may be an
underestimate of the true prevalence among HF outpatients. The demands of a large complex clinical
trial such as SCD-HeFT may discourage depressed patients from participating.
Similar to studies of depression in hospitalized patients, we found depression prevalent in chronic HF
outpatients. The prevalence of depression is likely to increase as the duration of HF increases.
Havranek et al[32] demonstrated prospectively that 21% of HF outpatients who are not depressed at first
assessment were depressed 1 year later.
In the only previous study of anxiety in HF outpatients not awaiting transplantation, the women in the
Studies of Left Ventricular Dysfunction (SOLVD) trial had worse Profile of Mood States anxiety scores
than a normative sample; men did not.[33] Anxiety occurred in 45% of HF outpatients participating in the
current study: 54% of the women and 43% of the men. Our findings suggest that anxiety scores in HF
outpatients (mean 36.7) are similar to other patients with chronic heart disease.[2,34] In the current study,
anxiety was more than 1.5 times as frequent in HF outpatients as in the Jiang et al study of inpatients
with HF.[19] The causes of anxiety may differ for patients who are in chronic and acute HF, or outpatients
may be more anxious about their HF because they are less confident about their care than inpatients.
Predictors of Mortality
In the current study, the Cox regression including only individual psychosocial predictors and treatment
group dummy variables suggested that depression, state anxiety, and social isolation amount predict
mortality in HF outpatients. Both depression and social isolation were predictors of mortality after
controlling for the demographic and clinical predictors of mortality. The importance of anxiety related to
mortality has not been assessed previously in this patient population, nor have the independent effects
of the 3 psychosocial variables been assessed.
The finding that depression predicts mortality is consistent with the Murberg et al study[14-16] of 119 stable
HF outpatients, and the Faris et al study[17] of 396 nonischemic HF outpatients. Depression also predicts
mortality in HF inpatients[18] and hospitalized patients who have HF.[19,20] Depressive symptoms are a
strong predictor of worsening HF health status.[35] Depression is reported more frequently with the BDI,
which assesses the presence of depressive symptoms, some of which are also characteristic of HF,
than with some other measures of depression.[36] However, the independent impact of depression and
clinical severity in our multivariate analysis indicates that depression contributes beyond the contribution
of HF symptoms. An increase of 1 in the natural log of the depression scale increased the risk of dying
2.35 times. Patients with mild, moderate, and severe depression were 2.2, 3.0, and 5.4 times as likely to
die as those who were not depressed based on the Cox depression model and BDI-II scores at the
midpoint of each depression category.
In the current study, anxiety was a univariate predictor of mortality in HF outpatients. In multivariate
analysis, its contribution was not independent of depression. The considerable overlap in the affective
symptoms of depression and anxiety was confirmed by the significant correlation between anxiety and
depression. Depression was the stronger predictor of mortality. The contribution of anxiety to mortality in
HF patients has only been examined in one previous study of inpatients; it did not predict mortality.[19]
The conceptualization of social support can affect its relationship to mortality. We found that the number
of people the patients could ask for support in stressful situations was a predictor of survival in
outpatients independent of demographic and clinical status and tending to be independent of
depression. The interrelationship of depression and social isolation is consistent with HF outpatients
who live alone having a higher likelihood of developing depression within 1 year.[32] Emotional support
(number of people available to talk over problems) was an independent predictor of mortality within 1
year in patients hospitalized for HF after controlling for demographics, clinical status, and depression.[37]
Social isolation was related to mortality in HF outpatients after controlling for depression and HF
severity.[14] In the same study, perceived social support defined as the likelihood of receiving help from
others during illness was not related to mortality. Similarly, in our study expectations of satisfaction with
social support received from others was not related to mortality.
HF, depression and anxiety include common symptoms and hyperactivity of the sympathoadrenalmedulla system, the hypothalamic-pituitary-adrenal axis, and abnormal platelet reactivity.[45] Thus,
establishing a causal relationship between depression and HF would require longitudinal analysis of
both psychologic and physiologic status.

The small sample and low mortality in the current study limited the number of variables examined and
the power of our findings. Additional variables may have made significant contributions to mortality in a
larger study. Because of the low number of minority group members and women enrolled in this study,
examination of psychosocial predictors in these groups requires further investigation. Furthermore,
participants in clinical trials and those willing to complete psychosocial questionnaires are self-selected
and may be less distressed than the population of outpatients with HF. They also may represent a less
distressed or healthier group within the SCD-HeFT trial. Although the PFOS patients did not differ
significantly in any one clinical characteristic from the SCD-HeFT patients, their EFs were slightly better
and there were slightly fewer women, minorities, and patients with NYHA class II HF, diabetes, and atrial
fibrillation/flutter. The combination of these differences may account for the lower overall mortality in
PFOS.
Conclusion
Depression and anxiety are common in HF outpatients. Psychosocial distress, including depression,
anxiety, and social isolation, are related to mortality in this patient population; depression and social
isolation predict mortality independent of demographic and clinical status and perceived functional
status.
Discrepancies between findings about the relationship of psychosocial factors to mortality of HF
inpatients and outpatients suggest that psychosocial distress differs in patients during acute and chronic
HF. Psychosocial distress is associated with mechanisms that increase mortality; improving depression,
anxiety, or social support might reduce chronic HF progression or mortality. Attention to assessment and
management of psychosocial issues is important in patients with chronic heart failure. Larger
longitudinal studies are required to examine the independent contributions of depression, anxiety, and
social isolation to mortality in this population and to evaluate the most appropriate targets for
intervention. Additional studies are needed to determine if these independent contributions differ based
on sex or race.

Table 1. Demographic and Clinical Characteristics of the Patients Enrolled in PFOS

(N = 153) and in SCD-HeFT (N = 2521) at Baseline

Afib, atrial fibrillation or flutter; ACE, angiotensin-converting enzyme; ARB, antiotensin II receptor
blocker.

Table 2. Psychosocial Status of the Patients at First Assessment for all PFOS
Participants (N = 153) and for Participants Randomized to Receive
Amiodarone (n = 45), ICD (n = 57), or Placebo (n = 51)

STAI, state anxiety; BDI-II, depression; SSSAT, satisfaction with social support; SSAMT, social support
amount; PLE, number of stressful life events in the past; FLE, number of stressful life events expected in
future.

Table 3. Summary of Cox Regression Models Using Individual Baseline Predictors

and Treatment Group to Predict Mortality

Afib, atrial fibrillation/flutter; FLE, number of stressful life events expected in future; ln, natural log
transformed; SSAMT, social support amount; SSSAT, satisfaction with social support; lag time reference
group is <1 month.
*Two-tailed probability, no a priori directional hypothesis.
95% CI.
One-tailed probability, in opposite of predicted direction.

Table 4. Cox Regression Model Predicting Mortality From Demographic and

Clinical Predictors and Treatment Group (N = 153)

Placebo is the reference group for treatment group comparisons.

Table 5. Cox Regression Model Predicting Mortality From Demographic and

Clinical Variables, Depression, Social Support Amount and Treatment
Group (n = 132)

Placebo is the reference group for treatment group comparisons.

References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.

Silverstone PH. Depression and outcome in acute myocardial infarction. Br Med J (Clin Res
Ed). 1987; 294: 219-220
Thomas SA, Friedmann E, Wimbush F, et al. Psychological factors and survival in the cardiac
arrhythmia suppression trial (CAST): a reexamination. Am J Crit Care. 1997; 6: 116-126
Frasure-Smith N, Lesperance F, Ginette G, et al. Social support depression and mortality during
the first year after myocardial infarction. Circulation. 2000; 101: 1919-1924
Musselman DL, Evans DL, Nemeroff CB. The relationship of depression to cardiovascular
disease: epidemiology, biology, and treatment. Arch Gen Psychiatry. 1998; 55: 580-592
Rozanski A, Blumenthal JA, Kaplan J. Impact of psychological factors on the pathogenesis of
cardiovascular disease and implications for therapy. Circulation. 1999; 99: 2192-2217
Denollet J, Brutsaert DL. Personality disease severity and the risk of long term cardiac events
with decreased ejection fraction after myocardial infarction. Circulation. 1998; 97: 167-173
Engle GL. Sudden and rapid death during psychological stress. Folklore or folk wisdom? Ann
Intern Med. 1971; 74: 771-782
Rahe RH, Romo M, Bennett L, et al. Recent life changes, myocardial infarction, and abrupt
coronary death. Studies in Helsinki. Arch Intern Med. 1974; 133: 221-228
Ruberman W, Weinblatt E, Goldberg JD, et al. Psychosocial influences on mortality after
myocardial infarction. N Engl J Med. 1984; 311: 552-559
Sterling P. Principles of allostasis: optimal design, predictive regulation, pathophysiology and
rational therapeutics. In: Shulkin J editors. Allostasis, homeostasis, and the costs of adaptation.
Cambridge (Mass): MIT Press; 2003; p. 17-64
Friedmann E, Thomas SA. Pet ownership, social support, and one-year survival after acute
myocardial infarction in the Cardiac Arrhythmia Suppression Trial (CAST). Am J Cardiol. 1995;
76: 1213-1217
Friedmann E, Thomas SA, Stein PK, et al. Relation between pet ownership and heart rate
variability in patients with healed myocardial infarcts. Am J Cardiol. 2003; 91: 718-721
Friedmann E, Katcher AH, Lynch JJ, et al. Animal companions and one-year survival of patients
after discharge from a coronary care unit. Public Health Rep. 1980; 95: 307-312
Murberg TA, Bru E. Social relationships and mortality in patients with congestive heart failure. J
Psychosom Res. 2001; 51: 521-527

15. Murberg TA, Bru E, Aarsland T, et al. Functional status and depression among men and women
with congestive heart failure. Int J Psychiatry Med. 1998; 28: 273-291
16. Murberg TA, Bru E, Svebak S, et al. Depressed mood and subjective health symptoms as
predictors of mortality in patients with congestive heart failure: a two-years follow-up study. Int J
Psychiatry Med. 1999; 29: 311-326
17. Faris R, Purcell H, Henein MY, et al. Clinical depression is common and significantly associated
with reduced survival in patients with non-ischaemic heart failure. Eur J Heart Fail. 2002; 4: 541551
18. Vaccarino V, Kasl SV, Abramson J, et al. Depressive symptoms and risk of functional decline
and death in patients with heart failure. J Am Coll Cardiol. 2001; 38: 199-205
19. Jiang W, Kuchibhatla M, Cuffe MS, et al. Prognostic value of anxiety and depression in patients
with chronicheart failure. Circulation. 2004; 110: 3452-3456
20. Jiang W, Alexander J, Christopher E, et al. Relationship of depression to increased risk of
mortality and rehospitalization in patients with congestive heart failure. Arch Intern Med. 2001;
161: 1849-1856
21. Bardy GH, Lee KL, Mark DB, et al. Amiodarone or an implantable cardioverter-defibrillator for
congestive heart failure. N Engl J Med. 2005; 352: 225-237
22. Spielberger CD, Lushene RE, Gorsuch RL. STAI manual. Palo Alto (CA): Consulting
Psychology Press; 1972
23. Frasure-Smith N, Lesperance F, Talajic M. The impact of negative emotions on prognosis
following myocardial infarction: is it more than depression? Health Psychol. 1995; 14: 388-398
24. Beck AT, Stear RA, Brown GK. The Beck Depression Inventory second edition. Boston (Mass):
Houghton Mifflin; 1996
25. Sarason IG, Sarason BR, Shearin EN, et al. A brief measure of social support-practical and
theoretical implications. J Soc Pers Relatsh. 1987; 4: 497-510
26. Holmes TH, Rahe RH. The social readjustment rating scale. J Psychosom Res. 1967; 11: 213218
27. Friedmann E. Psychological status, pet ownership, and attachment among post MI patients. In:
Paper presented at the 6th International conference on Human Animal Interactions, Montreal,
July. 1992
28. Rector TS, Kubo SH, Cohn JN. Patient's self-assessment of their congestive heart failure:
content, reliability, validity of a new measure, the Minnesota Living with Heart Failure
questionnaire. Heart Fail. 1987; 3: 198-209
29. Rector TS, Cohn JN. Assessment of patient outcome with the Minnesota Living with Heart
Failure questionnaire: reliability and validity during a randomized, double-blind, placebocontrolled trail of pimobendan. Am Heart J. 1992; 124: 1017-1025
30. Gottlieb SS, Khatta M, Friedmann E, et al. The influence of age, gender, and race on the
prevalence of depression in heart failure patients. J Am Coll Cardiol. 2004; 43: 1542-1549
31. Westlake C, Dracup K, Fonarow G, et al. Depression in patients with heart failure. J Card Fail.
2005; 11: 30-35
32. Havranek EP, Spertus JA, Masoudi FA, et al. Predictors of the onset of depressive symptoms in
patients with heart failure. J Am Coll Cardiol. 2004; 44: 2333-2338
33. Riedinger MS, Dracup KA, Brecht ML. Quality of life in women with heart failure, normative
groups, and patients with other chronic conditions. Am J Crit Care. 2002; 11: 211-219
34. Lane D, Carroll D, Ring C, et al. Do depression and anxiety predict recurrent coronary events
12 months after myocardial infarction? QJM. 2000; 93: 739-744
35. Rumsfeld JS, Havranek E, Masoudi FA, et al. Depressive symptoms are the strongest
predictors of short-term declines in health status in patients with heart failure. J Am Coll Cardiol.
2003; 42: 1811-1817
36. Thomas SA, Friedmann E, Khatta M, et al. Depression in patients with heart failure: physiologic
effects, incidence, and relation to mortality. AACN Clin Issues. 2003; 14: 3-12
37. Krumholz HM, Butler J, Miller Jeal. Prognostic importance of emotional support for elderly
patients hospitalized with heart failure. Circulation. 1998; 97: 958-964
Funding Information

This research was partially supported by grants R01 NR07613 from the National Institute of Nursing
Research, National Institutes of Health, Bethesda, MD and grants UO1 HL55766, UO1 HL55297, and
UO1 HL55496 from the National Heart Lung and Blood Institute, National Institutes of Health, Bethesda,
MD and by Medtronic, Minneapolis, MN, Wyeth-Ayerst Laboratories, Sanford, NC and Knoll
Pharmaceuticals, Mount Olive, NJ.
Reprint Address

Erika Friedmann, PhD, School of Nursing, University of Maryland, 655 W Lombard St, Baltimore, MD
21201

Erika Friedmann, PhDa,b, Sue A. Thomas, RN, PhD, FAAN a, Fang Liu, MSa, Patricia G. Morton, RN,
PhD, CANP, FAAN a, Deborah Chapa, RN, MS, CRNPa, and Stephen S. Gottlieb, MD, FACC c
School of Nursing, University of Maryland, Baltimore, MD
Brooklyn College of CUNY, Brooklyn, NY
c
School of Medicine, University of Maryland, Baltimore, MD
a
b