346 Coronary Heart Disease and Hypertension
cynicism and mistrust of others, was shown to be related to
occurrence of coronary disease (Barefoot & Lipkus, 1994).
One study involved a 25-year follow-up of physicians who
completed the Minnesota Multiphasic Personality Inventory
(MMPI; a precursor to the Cook-Medley scale) while in med-
ical school. High scores on the MMPI predicted incidence of
coronary disease and mortality from all causes, independent
of smoking, age, high blood pressure, and other risk factors
(Barefoot, Dahlstrom, & Williams, 1983). Another study has
shown evidence that low hostility scores are associated with
decreased death rates during a 20-year follow-up of nearly
1,900 subjects in the Western Electric Study (Shekelle, Gale,
Ostfeld, & Paul, 1983). Later research indicated that hostility
scores on the Cook-Medley are higher among certain groups,
particularly men and non-Caucasians in the United States, and
are positively correlated to smoking prevalence (Siegler,
1994). These “ndings make it possible to hypothesize that
hostility may account for some of the gender and socioeco-
nomic differences in mortality rates from cardiovascular dis-
eases (Stoney & Engebretson, 1994).
Presence of the emotional trait of anger has also been stud-
ied as a possible risk factor for coronary disease. One study used
various scales tailored speci“cally to anger traits and found
a signi“cant gradient between anger levels and the frequency
of subsequent cardiac events (Kawachi, Sparrow, Spiro,
Vokonas, & Weiss, 1996). More recently, researchers studied
nearly 13,000 individuals (including African American and
Caucasian individuals of either gender) andmeasured anger
using the Speilberger Trait Anger Scale. Each individual was
classi“ed as either having high, middle, or low anger traits,
with high scorers tending to be slightly younger males. Indi-
viduals who were the most anger-prone were 2.7 times more
likely to have MI than those with the lowest anger ratings
(J. Williams et al., 2000).
Clinical and Subclinical Depression
Approximately one in “ve cardiac patients can be diagnosed
with the signs and symptoms of clinical depression. Depres-
sive symptoms (not limited to major depression) following
MI has been associated with a three- to fourfold increase
in risk of cardiac mortality (Frasure-Smith, Lesperance, &
Talajic, 1993). These and other “ndings have enabled the
medical community to label depression as the most prevalent
and epidemiologically relevant psychosocial risk factor for
cardiovascular disease (Wulsin, Vaillant, & Wells, 1999).
Clinical depression can be diagnosed if a patient experiences
sadness or loss of interest or pleasure in most usual activities
that often interferes with his or her personal, occupational, or
social activities. Other symptoms such as sleep dif“culties, loss
of appetite or weight, fatigue, and thoughts of suicide or death
are often present as well (APA, 1994). Depression that coexists
with cardiac disease is often hard to diagnose, for patients often
attribute their symptoms, such as fatigue and other unexplained
somatic symptoms, to their heart disease. Also, cardiac patients
often replace typical symptoms such as sadness and guilt with
less typical symptoms such as irritability and anxiety (Fava,
Abraham, Pava, Shuster, & Rosenbaum, 1996). In addition,
studies suggest that symptoms that fall short of frank clinical
depression may also confer increased risk of poor outcomes in
CAD patients (Anda et al., 1993; Hans, Carney, Freedland, &
Skala, 1996; Schliefer, Macari-Hinson, & Coyle, 1989).
The presence of a major depressive episode in coronary
patients is associated with poor psychosocial rehabilitation
and increased medical morbidity (Carney, Freedland, Rich, &
Jaffe, 1995). Several studies have followed the clinical course
of depressed versus nondepressed cardiac patients and
have found an increase in events and lower mortality rates
associated with depression. Frasure-Smith et al. (1993)
prospectively followed 222 post-MI patients and found that a
diagnosis of major depression has a strong association with
mortality in the six months following hospital discharge. An-
other study followed patients for one year and found diagno-
sis of major depressive disorder at the time of angiography to
be the best predictor of a signi“cant cardiac event, including
such things as death, reinfarction, and bypass (Carney et al.,
1987). Schleifer, Keller, Bond, Cohen, and Stein (1989) found
that depressed patients had higher rates of rehospitalization
and reinfarction than their nondepressed peers. More recently,
two studies reported that initially healthy populations who
begin to experience a major depressive episode (Pratt et al.,
1996) or worsening of depressive symptoms (Wassertheil-
Smoller et al., 1996) are more likely to develop cardiac events
in the future. A similar set of studies (e.g., Appels, 1990; Kop,
Appels, Mendes de Leon, de Swart, & Bar, 1994) suggested
that symptoms of exhaustion or fatigue, even in the absence of
other clinical symptoms or depressive affect, are predictive of
subsequent development or worsening of cardiovascular
events or symptoms. This concept of a fatigue syndrome has
been termed •vital exhaustionŽ(Appels, 1990) and its predic-
tive value cannot be explained by the effects of illness on
mood or energy level (Kop et al., 1994).
There are several mechanisms that may explain the
link between depression and mortality in coronary patients.
Carney, Freedland, et al. (1995) suggest that depressed car-
diac patients are less likely to comply with medical therapeu-
tic and exercise regimens. Amick and Ockene (1994) believe
that a lack of compliance can often be attributed to an
unsupportive social network. Others attributed depression in
cardiac patients to the use of beta-blockers. However, over
the “rst 30 months of the Beta-Blocker Heart Attack Trial, no
difference was found between placebo and treatment groups