Psychological Risk Factors and Coronary Artery Disease: Epidemiology, Pathophysiology, and Management

Key Points

  • During the last three decades, epidemiologic studies have demonstrated a consistent relationship between psychosocial variables and coronary artery disease.

  • Psychosocial risk factors for coronary artery disease include depression; various anxiety syndromes, such as phobias, panic, and post-traumatic stress syndrome; anger/hostility; negative cognitive patterns such as pessimism; and chronic stress, including work stress, marital stress, social isolation, and low socioeconomic status.

  • Important new data indicate that positive psychological factors, including positive emotions and having a sense of purpose, improve physiology and increase longevity.

  • The pathophysiology governing these relationships stems from chronic activation of the autonomic nervous system and the hypothalamic-pituitary-adrenal axis.

  • Both the brain itself, which is remodeled by chronic stress, and the heart are end-target organs of chronic psychosocial stress.

  • Psychosocial factors such as depression and chronic stress lead to widespread peripheral effects that promote coronary artery disease, such as inflammation, hypercoagulability, metabolic syndrome and diabetes, hypertension, endothelial dysfunction, and enhanced physiologic reactivity to environmental stimuli.

  • Acute psychological stress can increase the risk for cardiac events among patients with coronary disease through widespread pathophysiologic effects.

  • Mixed results among sparse large-scale studies have inhibited the adoption of formal guidelines for management of psychosocial stress in clinical practice.

  • New developments within medical psychology hold promise for optimizing behavioral interventions in the future.

Since antiquity, there has been a strong lay notion concerning the relationship between psychological stress and the development of heart disease. However, it was not until the late 1970s that a significant scientific evidence base began to coalesce in this arena. By the early 1960s, the Framingham study had already identified key clinical risk factors for coronary artery disease (CAD), including smoking, hypertension, hyperlipidemia, diabetes, and a family history of premature CAD. At that time and into the following decade, interest in potential psychosocial risk factors for CAD was principally dominated by a construct proposed by Friedman and Rosenman. They conceived of a coronary-prone type A personality pattern, characterized as hard driving, time impatient, and prone to hostility. Initial research linked type A personality to a higher risk of cardiac disease compared with type B. Whereas this concept proved popular and became part of the cultural vernacular for many years, ultimately the construct was largely abandoned because of the lack of sufficient confirmatory evidence during prospective study.

Subsequent research, however, began to define strong links between various psychosocial risk factors and CAD. For instance, in 1979, the landmark Alameda County study was published, noting a stepwise gradient between the size of individuals’ social network and all-cause mortality. New and more sophisticated epidemiologic studies, which for the first time corrected for concurrent risk factors such as smoking and involving increasing, more representative sample sizes, began to demonstrate a similar stepwise gradient between the magnitude of depressive symptoms and adverse cardiac outcomes. At the same time, important animal work began to define pathophysiologic links between chronic stress and atherosclerosis. With respect to acute stress, newly available imaging techniques were able to show for the first time a relationship between acute stress and myocardial perfusion, ischemic wall motion abnormalities, and coronary vasoconstriction. Since then, a markedly vast literature has developed to link a large variety of psychosocial risk factors to CAD.

This chapter reviews current understanding of the epidemiologic link between psychosocial risk factors and CAD. Both negative factors that cause CAD and more recently defined positive factors that help buffer against the development of CAD are reviewed. The pathophysiologic basis for these links is broadly examined, and treatment considerations that are relevant to cardiologists are considered.

The Epidemiology Linking Psychosocial Risk Factors to Cardiovascular Disease

During the last few decades, a number of negative psychosocial factors have been linked to the development of cardiovascular disease (CVD). More recently, various positive psychological factors, such as the presence of positive emotions and having a sense of purpose, have been demonstrated to protect against CVD and to promote longevity. The psychological factors that have been shown to be epidemiologically linked to CVD are listed in Box 34-1 . These factors can be divided into three broad categories: (1) chronic negative thought patterns and emotions; (2) chronic stress; and (3) positive psychosocial factors that promote health and buffer against CAD. Each of these categories is reviewed.

BOX 34-1

Psychosocial Factors That Have Been Linked to the Progression or Prevention of Cardiovascular Disease

Negative Thought Patterns and Emotions

  • Depressive syndromes

    • Mild to moderate depressive symptoms

    • Major depression

    • Hopelessness

  • Anxiety syndromes

    • Generalized anxiety disorder

    • Phobic anxiety

    • Panic disorder

    • Post-traumatic stress disorder

  • Hostility and anger

  • Worry

  • Pessimism

Chronic Stress

  • Work stress

  • Marital stress and dissatisfaction

  • Social isolation and lack of social support

  • Low socioeconomic status

  • Caregiver strain

  • Adverse childhood experience

  • Perceived injustice

Positive Psychological Factors

  • Positive emotions

  • Optimism

  • Social support

  • Sense of purpose

Thought Patterns and Emotions

Thought patterns and emotions are logically linked because they are bidirectionally related. Thoughts commonly generate concomitant emotional responses, but our moods and emotional states can also affect the quality of our thinking. Depression, anxiety, and anger/hostility have been most commonly studied; but in recent years, other thought patterns, such as pessimism/optimism, have also emerged as an important area of study.


Depression has been particularly studied as a risk factor for CAD, for a variety of clinical reasons. First, depressive symptoms are common in society; major depression, alone, affects approximately 5% of the U.S. population at any time, with an increased frequency in cardiac cohorts. Second, depression is painful and debilitating, leading to loss of productivity and high economic costs. Third, various effective quantifiable tools exist to measure depressive symptoms, and epidemiologic studies that use these tools have consistently demonstrated that depression is associated with a heightened frequency of atherosclerotic heart disease and adverse clinical events.

Validated standardized scales, such as the Beck Depression Inventory and the Center for Epidemiologic Studies Depression Scale (CES-D), allow depressive symptoms to be characterized along a full spectrum, ranging from mild to severe depressive symptoms. At the severe end of this spectrum is major depression, which is a specific clinical psychiatric disorder that is established by formal diagnostic interview according to the following Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV) criteria: the presence of severely depressed mood and/or inability to take pleasure in all or most things that were previously considered enjoyable, lasting ≥2 weeks and accompanied by functional impairment and somatic complaints, such as fatigue or loss of energy nearly every day, insomnia or hypersomnia, change in appetite, diminished ability to concentrate, feelings of worthlessness or inappropriate guilt, and recurrent thoughts of death or suicidal ideation.

Studies have consistently demonstrated that the prevalence of major depression is increased by at least threefold in cardiac populations, occurring in at least 15% to 20% of patients who are status post–myocardial infarction, unstable angina, angioplasty, bypass surgery, or valve surgery or who have congestive heart failure. In addition, at least another 15% of cardiac patients can be expected to manifest minor to moderate depressive symptoms that do not meet the criteria for major depression according to DSM-IV criteria. This high prevalence of depressive symptoms in cardiac populations is believed to be due to a bidirectional relationship between depression and CAD.

This high frequency of depressive symptoms in cardiac patients is of particular concern given repeated evidence of a strong gradient relationship between the frequency of depressive symptoms and the occurrence of adverse cardiac events both among community cohorts who are observed prospectively and among cardiac disease cohorts. Evidence indicates that even mild depressive symptoms are associated with an increased frequency of cardiac events compared with patients who have no depressive symptoms ( Fig. 34-1 ).


Grouping of post–myocardial infarction (MI) patients according to their Beck Depression Inventory (BDI) scores, ranging from those with no depressive symptoms (BDI < 5) to those with moderately severe depression (BDI ≥ 19). A gradient relationship was observed for frequency of death according to the magnitude of depressive symptoms. Notably, increased events occurred even among patients with mild depressive symptoms (BDI scores of 5 to 9).

Various meta-analyses have been performed, and each has found a prognostic association between depression and adverse outcomes. For example, meta-analyses of community-based cohorts conducted by Rugulies and by Wulsin and Singal found depression to be associated with a relative risk ratio of 1.64 for the development of CAD. Similarly, Van Melle and coworkers and Barth and colleagues found depression to be associated with at least a twofold increase in event risk for cardiac populations among meta-analyses involving 22 studies and 20 studies, respectively.

Anxiety Syndromes

Like depression, feelings of anxiety vary across a wide spectrum, but anxiety disorders are more common than depression. For instance, within the National Comorbidity Survey, the 12-month prevalence of diagnosed anxiety disorders within a representative national survey was approximately 20%. Transient experience of anxiety is a universal phenomenon that is often an adaptive warning of threat or danger. When anxiety is chronic or cannot be controlled, however, it becomes maladaptive. Many studies have now assessed the epidemiologic significance of both self-reported anxious symptoms and various pathologic forms of anxiety, as characterized in the DSM-IV.

In early study, conflicting results were obtained regarding anxiety as a CVD risk factor, but in recent years, increasing study has clarified the status of anxiety syndromes relative to cardiac risk. Table 34-1 lists the outcomes associated with measurements of anxiety in community samples for studies reported since 2003. In general, these studies found anxiety to be associated with significant risk ratios for adverse clinical events, although in some studies, anxiety was not a significant predictor after full adjustment for covariates of risk. This sensitivity to covariate adjustment may help explain some of the conflicting data concerning anxiety among earlier epidemiologic studies.

TABLE 34–1

Outcomes by Anxiety in General Population Samples

Study, Year No. of Subjects Follow-up Scale Endpoints Adjusted Risk Ratios
McCarron et al, 2003 9,239 20 years Physician impression ACM 1.36 (1.07-1.72)
Eaker et al, 2005 3,682 10 years Framingham anxiety scale ACM, male 1.22 (1.08-1.38)
ACM, female 1.27 (1.05-1.55)
Ringbäck Weitoft and Rosén, 2005 (3 different cohorts) 10,733 5 years Hospital ICD codes ACM 1.71 (1.1-2.5)
10,035 ACM 2.3 (1.6-3.3)
9,100 ACM 1.8 (1.1-1.3)
Ostir and Goodwin, 2006 506 5 years Zung ACM 1.52 (1.02-2.28)
Mykletun et al, 2007 61,349 4.4 years HADS ACM 0.90 (0.83-0.98)
CD 0.89 (0.67-3.38)
Fan et al, 2008 129,499 N/A History of anxiety on PHQ-8 Self-reported
CVD 1.46 (1.37-1.54)
Shen et al, 2008 735 12.4 years MMPI anxiety subscales MI 1.43 (1.17-1.75)
Phillips et al, 2009 4,256 15 years DSM-III (interview) ACM 1.80 (1.16-2.80)
CD 1.84 (0.98-3.45)

ACM, all-cause mortality; CD, cardiac death; CVD, cardiovascular disease; HADS, Hospital Anxiety and Depression Scale; MI, myocardial infarction; MMPI, Minnesota Multiphasic Personality Inventory; PHQ, Patient Health Questionnaire.

There have also been many studies concerning anxiety within cardiac populations, but most of these have involved small sample sizes. Among three studies with cardiac populations >500 individuals, each demonstrated an increased risk for adverse clinical events in the presence of anxiety. In an interesting study in this arena, Frasure-Smith and Lesperance compared self-reported anxiety symptoms versus the identification of generalized anxiety disorder (GAD) according to use of the Structured Clinical Interview for DSM-IV. The diagnostic criteria for GAD include excessive anxiety or worry for more days than not for >6 months, with difficulty in controlling the worry and symptoms of functional impairment and distress (e.g., fatigue, insomnia). The study assessed 804 patients with stable CAD observed for a composite index of major adverse cardiac events. Whereas both self-reports of anxiety and GAD were associated with increased events before covariate adjustment, only GAD remained a significant predictor of risk after covariate adjustment. These results are indicative of a gradient relationship, whereby more pathologic anxiety exerts greater pathophysiologic effects.

In addition to GAD, most of the other DSM-IV anxiety syndromes have been assessed relative to their clinical sequelae. Among the anxiety disorders that will be encountered by cardiologists is panic disorder, because the symptoms associated with panic attacks, including chest pain, palpitations, and dyspnea, can lead to frequent emergency department presentations. Panic disorder involves the presence of recurrent and unexpected panic attacks, with the occurrence of anticipatory anxiety or worry and a significant change in behavior related to the attacks. Among studies concerning panic attack, Smoller and coworkers observed 3369 women for 5.3 years and found panic attacks to be associated with a substantial hazard ratio for the development of CVD and stroke as well as all-cause mortality. Panic disorder was also found to be associated with an increased frequency of acute myocardial infarction in the follow-up of 9641 patients with panic disorder who were matched to 28,923 controls, and similar findings were noted in other large studies.

The most common of the DSM-IV forms of anxiety disorders is presence of phobic disorders. The 12-month prevalence of simple phobias was approximately 9% among the individuals within the National Comorbidity Survey, and social phobias were found within approximately 8%. The DSM-IV criteria for phobia include the presence of excessive and persistent fear that is cued by the presence or anticipation of exposure to a specific object or situation, with the exposure usually invoking an immediate anxiety response. Notably, phobic anxiety, as assessed by the Crown-Crisp Index, has also been linked to cardiac death in large cohorts, including the follow-up of 33,999 men from the Health Professionals Follow-up Study and 72,359 women from the Nurses’ Health Study.

Another anxiety disorder that has recently been strongly linked to cardiovascular sequelae is post-traumatic stress disorder (PTSD). PTSD is considered present if, after the exposure of an inciting traumatic event, subjects report re-experiencing of the traumatic event, hyperarousal, and avoidance of traumatic reminders and emotional numbing. In a prospective study of men who had served in the military, a stepwise relationship was noted between increasing symptoms of PTSD and both cardiac death and nonfatal myocardial infarction, and a second study found a relationship between PTSD and incident CAD among a community cohort of 1059 women who were observed for 14 years. In a third study, Boscarino evaluated 4328 men who had served in the Vietnam War. The presence of PTSD was associated with a more than twofold increase in the frequency of subsequent cardiac mortality, with the effect being independent of the presence of depression. PTSD patients are more prone to depression, but evidence indicates that pathophysiologic effects from PTSD, such as hypertension, may occur independently of depressive symptoms.

Anger and Hostility

Anger and hostility are often grouped together in studies of psychological risk factors because of overlapping characteristics. Hostility refers to an entrenched cognitive trait of easily precipitated resentfulness, cynicism, or suspicious negative thoughts about others. This cognitive style leads to frequent expressions of anger and negative social exchanges. Anger is an acute negative emotion, but individuals may by temperament or due to life experience be predisposed to a cognitive style of angry thoughts that may result in either expressed or suppressed feelings of anger. Hostile individuals are frequently angry, but anger can be experienced without hostility.

Unlike for depression or anxiety, no psychiatric system of classification has been developed for patients manifesting syndromal hostility or anger. Also, research in this arena has been characterized by the use of widely varying scales, which may have contributed to disparate findings over the years. Moreover, whereas anxiety and depression can be assessed both by self-report and by independent structured interview, the use of a standard structured interview approach has not been commonly applied for the study of anger and hostility. This represents a potential limitation in this arena because lack of self-awareness or self-denial may potentially limit the accuracy of self-reports concerning anger and hostility. Nevertheless, despite such limitations, an increasing literature regarding anger and hostility has emerged.

Chida and Steptoe recently conducted a meta-analysis of published prospective cohort studies concerning anger and hostility. Among 25 studies involving community samples, the hazard ratio for CVD events in initially healthy populations in association with anger and hostility was 1.19 (95% CI, 1.05-1.35). Similarly, the hazard ratio was increased to 1.24 (95% CI, 1.08-1.42) among 19 studies involving patients with known CVD. Further support for the link between hostility or anger and CVD comes from a number of studies that have found these psychological variables to be associated with greater presence or progression of objectively measured atherosclerosis.

Pessimism Versus Optimism

The tendency that people have toward thinking in pessimistic versus optimistic patterns has been closely linked to health outcomes. Optimism and pessimism have most commonly been measured in medical research in two ways. One characterization, developed by Seligman and colleagues, defines optimism versus pessimism in terms of the way individuals attribute the causes to life events. Optimists tend to see negative events as temporary and positive events as more permanent; they tend to attribute specific causes to negative events while viewing positive events in more global terms; and they tend to attribute external causes to negative events rather than employing self-condemnation. Pessimists have the opposite “explanatory style.” Another characterization of optimism versus pessimism, as developed by Scheier and Carver, defines optimism versus pessimism in terms of individuals’ disposition toward expecting positive or negative future outcomes. This disposition toward optimism or pessimism is seen as influencing peoples’ behaviors, how they pursue goals, and their resilience to life stress.

Both approaches toward measurement of pessimism and optimism have been linked to adverse clinical events. A 10-year follow-up of 1306 men from the Normative Aging Study found that those with a pessimistic explanatory speaking style had the highest frequency and those with an optimistic speaking style had the lowest frequency of incident angina and coronary events (myocardial infarction or cardiac death) during follow-up, whereas those who had a neutral speaking style had an event rate that was intermediate ( Fig. 34-2 ).


Occurrence of incident angina and hard cardiac events among 1306 healthy subjects in the Normative Aging Study, followed up for 10 years. A gradient relationship was noted for outcomes for subjects classified as having a pessimistic, neutral, or optimistic explanatory speaking style.

In a study of 7216 subjects who were observed during four decades, those who tested pessimistic on a personality inventory showed a significant linear trend toward greater all-cause mortality with increasing pessimism. Among elderly subjects observed for approximately 9 years, those subjects who tested high for optimism, compared with high scores for pessimism, had a substantially reduced hazard ratio for all-cause mortality (0.55; 95% CI, 0.42-0.74) and cardiac mortality (0.23; 95% CI, 0.10-0.55). In another elderly cohort, the degree of measured dispositional optimism was found to be inversely associated with the frequency of cardiovascular death in a cohort of 545 men observed for 15 years.

Most recently, Tindle and associates assessed the relationship between optimism and CVD in the largest cohort yet studied, 97,253 women from the Women’s Health Initiative. Optimistic women, compared with pessimistic women, again manifested a reduced adjusted risk ratio for cardiac mortality (0.70; 95% CI, 0.55-0.90) in this study. Together, these studies provide consistent evidence for a strong relationship between pessimism and optimism and adverse clinical events.

Worry, Rumination, and Other Negative Thought Patterns

A variety of other negative thought patterns, which may relate to anxiety or depression or serve as precursors to these psychological states, have received attention in the cardiovascular literature but not with a sufficient breadth of epidemiologic study to date. One of these cognitive states is chronic worry, which represents a cognitive component or precursor of anxiety. Among 1757 men in the Normative Aging Study, worry was associated with an increased risk ratio for nonfatal myocardial infarction (RR, 2.41; 95% CI, 1.40-4.13), with a dose-response relationship noted for both nonfatal myocardial infarction and total cardiac events.

Rumination is another negative cognitive state involving the tendency to repetitively think about negative events. Whereas epidemiologic study is lacking, ruminators have been shown to have heightened heart and blood pressure reactivity to acute stress and delayed recovery of these responses. Perfectionism also represents a negative cognitive state. Perfectionism is characterized as the tendency to set excessively high standards for performance, accompanied by a highly critical style of self-examination. Preliminarily, perfectionism has been linked both to mortality in one study and to excessive cortisol secretion during psychological stress in a second study. The cognitive tendency to be forgiving or unforgiving is another cognitive pattern that was found to influence the degree of cardiovascular reactivity to stress in one study. Clearly, much more study is needed in this arena, but combined, these studies are suggestive of a broad link between chronic negative cognitive states and cardiovascular sequelae.

Chronic Stress

The study of chronic stress is unique within the context of psychological risk factors for CVD because chronic stress can be readily studied in controlled animal models of stress, complementing the epidemiologic study in humans. A set of investigations performed in cynomolgus monkeys (Macaca fascicularis) has been particularly insightful for elucidating the relationship between chronic stress and atherosclerosis.

Cynomolgus monkeys are an apt model for correlative study because they develop coronary atherosclerosis when under stress or when fed fatty diets, with similarities to humans in terms of coronary pathology and pathophysiology. Cynomolgus monkeys also have definable and quantifiable social characteristics, and their social environment can easily be modified to create conditions of chronic stress. One approach to inducing chronic stress in these monkeys is to take advantage of the fact that they form well-defined social status hierarchies, with the most dominant monkeys reliably defeating more subordinate monkeys during competitive interactions. Once dominance rank is established, these monkeys form stable hierarchical societies. However, if monkeys are constantly placed into new social groups, a chronically stressful environment is created as the monkeys continually reinitiate their attempt to establish dominance within their new social groups. When male monkeys within stable versus unstable groups were kept on a low-cholesterol diet, the dominant male monkeys within the unstable social groups developed endothelial injury. When both the stable and unstable social groups of monkeys were fed a high-cholesterol diet, both groups developed atherosclerosis, but the magnitude of disease was greater in the dominant male monkeys that were in the unstable environments.

Chronic stress has been well studied within the epidemiologic literature. The most commonly studied chronic stress in this regard has been work stress. Estimates as to the frequency of chronic work stress vary widely, but it is a common stressor. The leading models of work stress are the model of job strain as developed by Karasek and associates and the effort-reward imbalance model as developed by Siegrist ( Fig. 34-3 ). The job strain model posits that individuals experience job strain when experiencing work that is highly demanding but associated with low job latitude. Lack of control, however, appears to be a more toxic factor than high job demand. In the effort-reward imbalance model, stress occurs when high demand is associated with low “reward,” either in terms of financial or professional reward (e.g., opportunities for promotion) or in terms of psychological reward (e.g., sense of security or self-esteem). Notably, both models have been linked to adverse CAD outcomes, as summarized in a review of 13 prospective cohort studies. Compared with each other, the job strain and effort-reward imbalance models were comparable predictors for adverse outcome.


Two common conceptual models of job stress. The job strain model (left) is based on assessment of the amount of job demand and decision latitude at work. The presence of high demand but low decision latitude is characteristic of job stress. The effort-reward imbalance model (right) is based on assessment of job demand versus “reward” at work, whether financial or in terms of nonfinancial factors such as recognition, advancement, and prestige at work. High effort with low reward is characteristic of job imbalance.

(From Rozanski A, Blumenthal JA, Davidson KW, et al: The epidemiology, pathophysiology, and management of psychosocial risk factors in cardiac practice: the emerging field of behavioral cardiology. J Am Coll Cardiol 45:637, 2005.)

Other studies have also linked work stress to the presence of accelerated atherosclerosis during serial carotid ultrasound study. The effects of work stress tend to be accentuated in individuals with lower socioeconomic status and those with lower social support. However, work stress may also affect white-collar and higher socioeconomic status workers with seemingly more autonomous jobs as high work demand or low latitude commonly occurs because of internal psychodynamic factors, such as the need for recognition or perfectionistic tendencies.

Although marital stress or dissatisfaction is common in society, for years it was less studied than work stress relative to its association with CVD. Many studies over the years have studied clinical outcomes relative to marital status (e.g., remaining married, single, divorced, or widowed), but only recently have investigators begun to focus on outcomes as a function of marital quality. Among community cohorts, the quality of marital communications and marital conflict were found to strongly influence the frequency of adverse cardiac outcomes during a 10-year follow-up of 3682 subjects in the Framingham Offspring Study. For example, women who self-silenced themselves during marital conflicts were noted to have a fourfold increase in mortality during follow-up. Similarly, an increased frequency of cardiac events was noted among those reporting more negative marital interactions during a 12.2-year follow-up of 9011 British civil servants. Among patients with CAD, Orth-Gomer and coworkers found marital stress to be associated with a 2.9-fold increased risk of recurrent events during a 5-year follow-up of women who were status post–myocardial infarction, and Coyne and associates found that marital quality influences mortality rates among patients with congestive heart failure. Marital quality has also been found to influence the rate of progression of atherosclerosis as measured during both serial carotid ultrasound and coronary artery calcium scanning.

Data garnered during three decades provide strong epidemiologic proof for the health damaging of poor social support and social isolation. Social support in epidemiologic study is commonly divided into the amount of tangible or instrumental support (e.g., the size of one’s social network) and the amount of emotional social support. Inadequacy in both types of support is linked to a heightened incidence of CVD and reduced longevity, both among community-based cohorts and among patients with known CAD. Within these studies, an inverse gradient has consistently been noted between the amount of social support and the development of CAD or cardiac events ( Fig. 34-4 ).


Mortality in the Alameda County study according to the size of subjects’ reported social network for men and for women, ranging from smallest to largest network size. An inverse gradient was observed in both men and women.

Two other common stressors are low socioeconomic status and caregiver stress. Low socioeconomic status is associated with such psychosocial factors as higher anxiety resulting from hardship and environmental threat, poorer social support, demoralized feelings, and lower self-esteem. There is strong evidence to suggest that the impact of socioeconomic status on outcomes is due not only to material disadvantage and poorer health habits but also to psychological and stress-related mechanisms. By contrast, whereas caregiver strain is commonplace in today’s society, it has been much less studied, especially in terms of cardiac outcomes. In the Nurses’ Health Study, caregiving for >9 hours per week was found to be associated with significantly increased risk of fatal CAD or nonfatal myocardial infarction. More study in this arena is required. Recently, other forms of chronic stress have begun to receive attention as potential risk factors for CVD, including the long-lasting influence of adverse childhood experiences and the experience of organizational injustice.

Positive Psychological Factors

Until recently, most investigation regarding psychosocial factors and CVD has focused on just the negative relationships associated with factors such as depression and anxiety. However, two psychosocial factors have been persistently associated with a full spectrum of both negative and positive effects: optimism and social support, as reviewed before. More recently, new research has focused on the buffering and health-promoting effects of other positive factors, including positive emotions and the satisfaction of basic psychological needs that can induce a state of relative flourishing. Positive emotions in the context of this research have been defined broadly to include both emotions such as happiness and states of being that reflect a positive engagement with the environment, such as the presence of curiosity and interest.

In important work, Fredrickson developed the broaden-and-build model concerning emotions. This work has demonstrated an important directional relationship between the quality of emotions and cognitive functions, such as a broader scope of attention, more flexibility, and better problem-solving and creativity skills. In positive emotional states, individuals are seen as more likely to be friendly and optimistic. Thus, positive emotion is seen as increasing individuals’ personal resources and providing them with increased resilience to cope with stress. A recent meta-analysis has assessed the relationship between positive well-being and longevity among 35 studies involving initially healthy populations and 35 other studies involving patients. In both cohorts, the presence of positive well-being was associated with reduced mortality. To date, however, there has been little study regarding the effect of positive emotions on cardiac outcomes. Certain attitudinal states have also received recent attention as to their impact on overall well-being. These include the practice of gratitude and social altruism. These parameters, however, have not yet been well assessed in terms of cardiac epidemiology.

In addition to positive emotions, various theorists have postulated that individuals have psychological needs that must be met for optimal satisfaction and that the absence of these needs leads to psychological tension. Whereas these theorists differ in what constitutes these needs, there is general agreement among virtually all theorists that social connection is a basic psychological need. This may help explain why measured levels of social support have consistently been a very strong disease predictor. In addition, some theorists also posit that individuals have a basic need for meaning or sense of purpose, which when unmet leads to chronic tension.

Recent data support this assertion. For example, in a study of 1238 elderly individuals, purpose in life was assessed according to a 10-item scale, and the subjects were then observed for a mean of 2.7 years. Those who were identified as having a higher level of purpose in life had a substantially adjusted reduced risk of mortality (0.60; 95% CI, 0.42-0.87) ( Fig. 34-5 ). Similarly, in the MacArthur Study of Successful Aging, older adults in the age range of 70 to 79 years who reported being more useful to friends and family during a baseline interview had reduced mortality and disability during a 7-year follow-up compared with those who reported never or rarely feeling useful. Similar findings were also noted in a third study involving a 6-year follow-up of elderly subjects in Japan. More study is needed to assess the epidemiologic effects of having a sense of purpose in younger subjects and its specific relationship to cardiovascular disease.


Cumulative hazard among 1238 older community-dwelling individuals observed for 5 years, according to the presence of a low versus high sense of purpose. The hazard rate for mortality in persons with high scores for purpose in life was about 57% of that for persons with low scores.

(From Boyle PA, Barnes LL, Buchman AS, et al: Purpose in life is associated with mortality among community-dwelling older persons. Psychosom Med 71:575, 2009.)

The influence of positive and negative psychological factors as well as the influence of biologic and behavioral factors has been postulated to be contained within a composite variable, one’s sense of emotional vitality, as conceptualized in Figure 34-6 . The emotional aspect of vitality is augmented by positive emotions, positive thought patterns like optimism, and the satisfaction of basic psychological needs like social support and a sense of purpose. By contrast, chronic negative emotions, negative thought patterns like rumination and worry, dissatisfaction of basic psychological needs, and chronic stress are all seen as depleting vitality.


Paradigm for emotional vitality. Emotional vitality is fostered by positive emotions, healthy thought patterns such as optimism, and satisfaction of basic psychological needs, such as social connectedness. High emotional vitality provides energy and resilience for dealing with stress and negative emotions. In turn, successful stress management and resilience in the face of negative emotions build self-esteem and preserve emotional vitality.

(From Rozanski A, Blumenthal JA, Davidson KW, et al: The epidemiology, pathophysiology, and management of psychosocial risk factors in cardiac practice: the emerging field of behavioral cardiology. J Am Coll Cardiol 45:637, 2005.)

Support for this concept comes from a study by Kubzansky and Thurston, who assessed the link between the emotional aspects of vitality and incident CAD among 6025 subjects in the National Health and Nutrition Examination Survey (NHANES) I. During a mean follow-up of 15 years, those who manifested higher levels of emotional vitality manifested less CAD.

Comparison of Cardiovascular Disease Risk Factors

In summary, epidemiologic studies have identified an increasing number of psychosocial risk factors associated with CAD. Although these risk factors have been studied in isolation, very frequently they cluster together. For example, the frequency of depression has been found to be increased threefold among individuals experiencing high job stress. The risk for CAD and cardiac events associated with many psychological risk factors for CAD generally demonstrates a gradient relationship and a level of risk that is comparable to conventionally studied CAD risk factors ( Fig. 34-7 ).


Comparison of the risk ratios for traditional Framingham risk factors in men versus the risk ratios for depressive symptoms and clinical depression as reported by Rugulies. The risk ratios for traditional risk factors are for death due to cardiac causes (death, myocardial infarction, coronary artery insufficiency, development of angina). For depressive symptoms and clinical depression, the risk ratios are for death due to cardiac disease and myocardial infarction. CI, confidence interval; HT, hypertension; HDL, high-density lipoprotein; LDL, low-density lipoprotein.

(From Rozanski A, Blumenthal JA, Davidson KW, et al: The epidemiology, pathophysiology, and management of psychosocial risk factors in cardiac practice: the emerging field of behavioral cardiology. J Am Coll Cardiol 45:637, 2005.)

A unique study that compared psychosocial risk with other CAD risk factors was the INTERHEART case-control study. This study examined a variety of CAD risk factors among an international population of 12,461 acute post–myocardial infarction patients, matched to 14,637 control subjects. A simple psychological index in this study was comparable to other CAD risk factors in terms of myocardial infarction risk ( Fig. 34-8 ). This psychosocial index remained a robust predictor of myocardial infarction independent of geographic or ethnic context.


The risk of acute myocardial infarction results from the international INTERHEART case-control study for each of nine CAD risk factors were evaluated, adjusted for age, gender, and geographic location. The prevalence of each CAD risk factor is presented for controls and cases in the third and fourth columns. The prevalence rates for the psychosocial index were not calculated as it was derived from a statistical odds ratio. PAR, population attributable risk.

(From Yusuf S, Hawken S, Ounpuu S, et al: Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case-control study. Lancet 364:937, 2004.)

Of note, within epidemiologic study, the risk associated with psychological risk factors is adjusted for other CAD risk factors. However, because behavioral and metabolic CAD risk factors tend to aggregate disproportionately among individuals with psychosocial stress, in causative fashion, the true cardiovascular risk posed by psychosocial risk factors may be even greater than that reported within the literature.


For many years, the pathophysiologic mechanisms responsible for the morbidity and mortality associated with psychosocial risk factors for CVD were obscure. However, this is no longer the case. Both animal study and sophisticated research in humans have elucidated a complex pathophysiology that is unleashed by negative psychosocial factors, resulting in widespread systemic effects. The pathophysiology associated with depression and chronic stress is summarized in Figure 34-9 . These two psychosocial conditions have been most studied relative to pathophysiology and are thus reviewed herein. However, there is also ample research into the pathophysiologic effects associated with other common psychosocial risk factors for CVD, such as poor social support, anxiety, and anger/hostility, as well as increasing study of the beneficial effects associated with positive factors such as positive emotions and optimism that are not reviewed in this chapter but are well documented in the literature.


Schematic representation of the principal pathways by which chronic stress and affective disorders promote disease. The primary effect of perceived stress on chronic negative emotions is chronic activation of the hypothalamic-pituitary-adrenal (HPA) axis and the sympathetic nervous system (SNS). Widespread peripheral effects derive from this central nervous system activation (see text). In addition, there are direct effects on central nervous system remodeling that may help drive altered behaviors. Another consequence is a heightened physiologic responsivity to acute environmental stimuli. ANS, autonomic nervous system.

(From Rozanski A, Blumenthal JA, Davidson KW, et al: The epidemiology, pathophysiology, and management of psychosocial risk factors in cardiac practice: the emerging field of behavioral cardiology. J Am Coll Cardiol 45:637, 2005.)

Effects of Chronic Stress

The brain serves as a constant sentinel for all stimuli that are perceived as physically or psychologically threatening in our environment. Perceived threat induces an acute stress response that involves activation of both the autonomic nervous system and the hypothalamic-pituitary-adrenal (HPA) axis. When perceived or real threat remains persistent, a chronic stress response ensues, involving profound dysregulation of both systems.

Under conditions of acute stress, the secretion of cortisol serves as a negative feedback system, helping to terminate the physiologic response to acute stress. By contrast, chronic stress results in turn-off of this negative feedback loop, with resultant high cortisol secretion. In addition, there is elevated sympathetic nervous system stimulation during chronic stress. The chronic stimulation of the HPA and sympathetic nervous systems within the brain during chronic stress leads to the cascade of peripheral pathophysiologic effects that are outlined in Figure 34-9 , including increased inflammation, hypercoagulability, metabolic syndrome, central obesity, and hypertension.

In addition, an important and consistent effect of chronic stress is that it leads to impaired and exaggerated physiologic reactivity, as characterized by augmented heart rate and blood pressure responses to physiologic stimuli. A variety of clinical data indicate that such enhanced physiologic reactivity leads to accelerated atherosclerosis. For example, in experimental study, cynomolgus monkeys manifesting higher heart rate responses to a threatening experimental stimulus demonstrated atherosclerotic lesions in coronary and carotid arteries that were twice as large as those of less reactive monkeys. Pretreatment with beta blockade abolished the excess atherosclerosis that is observed in dominant males housed in unstable social environments, indicating that sympathetic activation is an important mediator of this stress-induced atherosclerosis. In parallel to this animal work, individuals with greater physiologic reactivity have been found to manifest accelerated atherosclerosis during serial carotid ultrasound study.

An area of recent interest is an apparent relationship between chronic stress and premature aging, as assessed by the effect of stress on the length of leukocyte telomeres. Telomeres are repetitive DNA sequences complexed with proteins that cap and protect chromosome ends. They decline in length with age, but limited study has found shortened telomere length among those with more perceived stress ( Fig. 34-10 ) and in caregivers, and a study suggests an association between pessimism and shorter telomeres as well.

FIGURE 34-10

Comparison of mean telomerase length and telomerase enzyme activity in subjects with high versus low perceived stress.

Only gold members can continue reading. Log In or Register to continue

Jul 10, 2019 | Posted by in CARDIOLOGY | Comments Off on Psychological Risk Factors and Coronary Artery Disease: Epidemiology, Pathophysiology, and Management
Premium Wordpress Themes by UFO Themes