Brent N Henderson & Andrew Baum. Handbook of Health Psychology. Editor: Stephen Sutton, Andrew Baum, Marie Johnston. Sage Publications, 2008.
The emergence and success of behavioral medicine and health psychology have been due in part to the changing nature of health and disease over the past 100 years. This is evident in several aspects of modern medicine. The nature of health threats has changed, from infectious diseases like influenza to chronic illness and cancer. Life expectancy has increased substantially and gains of 50 per cent or more have been attributed to the elimination of polio, tuberculosis, influenza, and smallpox (Matarazzo, 1984). A 1979 Surgeon General’s Report in the US estimated that, in 1900, for every 100,000 people there were about 480 deaths per year due to influenza, diphtheria, pneumonia, tuberculosis, and other infectious illnesses (Califano, 1979). In 2000, this figure dropped to only 30 deaths per 100,000 people. However, this dramatic decline in infectious illness was matched by a steep climb in more chronic diseases that have substantial behavioral causes. Cancer deaths have more than tripled, and heart disease, cancer, and AIDS have become prominent causes of death and disability. These diseases have no cure, no vaccine to prevent them, but are caused in part by lifestyle and behavior, which can be modified. Diet, tobacco use, drug use, stress, and exercise are intertwined with people’s lifestyles, and modification of these processes should help us control and manage these diseases. Health psychology offers unique insight and promise in controlling these modern ‘epidemics’.
Health psychology is a biobehavioral discipline, focusing on behavioral and biological mechanisms by which environmental or social experiences are translated into physiological changes and changes in one’s health. Interest in biobehavioral interactions is not new, but its most recent emergence in health-related areas has been both catalytic and controversial. The extent to which health and wellness are biological versus psychological states remains a point of debate and contention despite the recognition that biological events are immediate, ‘proximate’ causes while psychological or behavioral variables promote or impair health by influencing these biological events. Disease is essentially a biological event; it typically involves dysfunction or damage to bodily tissue, organs or systems, and whether it has behavioral causes or not it remains a biological process. For example, tobacco use certainly affects health and is a cause of cancers, heart disease and other illnesses. However, it affects these outcomes by causing biological damage, such as making cells in the lungs more susceptible to mutations and malignancies or by promoting atherosclerosis in the circulatory system. Stress also has a range of effects on disease etiology or progression, but it conveys these effects by affecting changes in the immune, endocrine, cardiovascular, gastrointestinal, and other bodily systems’ activity. ‘Disease-bearing variables’ or behavioral pathogens (Matarazzo, 1984) must be translated into biological changes in order to contribute to physical disease. Biobehavioral pathways or mechanisms, then, are sets of related behavioral and biological processes that can modify one another and provide ways to explain and transmit behavioral influences on health and illness. Research in health psychology is focused on the pathways by which behavioral effects are conveyed, amplified, and/or modified, and herein lies the promise of health psychology for understanding and controlling disease.
This chapter is about these biobehavioral mechanisms, with an emphasis on how various bodily systems are altered by behavioral variables and how these changes contribute to pathophysiology There are several ways to categorize and cluster these systems. One such taxonomy centers on function: systems can be regulatory systems (e.g., nervous system, endocrine system), transport systems (e.g., cardiovascular system, lymph system), resource systems (e.g., respiratory system, gastrointestinal systems), and effector or defense systems (e.g., immune system, DNA repair system). Such a classification permits evaluation of different levels of behavioral influences; effects on regulatory systems, for example, are likely to reverberate in other systems since they exert some control over these other systems. It also permits generation of testable hypotheses about channels or pathways through which behavioral effects are transmitted. It should be noted that our coverage of biological systems includes most but not all bodily systems; readers are referred elsewhere for a comprehensive and more thorough review of biological systems or divisions relevant for health and disease (West, 1991).
There are three main ways in which behavioral variables influence health-related outcomes. First, some of these variables or conditions exert direct effects on the functioning of a system or systems. These direct effects are considered to be primary outcomes of a set of conditions. For example, stress is thought to exert direct effects on most systems of the body, many of which can be pathogenic. In addition, behavioral or social variables may affect other behaviors that in turn have direct consequences for health. These indirect pathways would include, for example, any variables or conditions that influenced smoking or other tobacco use (smoking and tobacco have direct effects on several pathogenic processes), diet, exercise, sleep, and so on. A third set of pathways includes behavioral variables or conditions that affect treatment once one is ill. Access to care, adherence, lifestyle change, and other factors influencing access, compliance, and maintenance of appropriate treatments/changes are representative of these pathways. Both chronic burdens and acute stressors would be expected to exert effects in these ways, suggesting generally that behavioral and social variables can create vulnerabilities as well as exacerbate or ‘realize’ pre-existing vulnerabilities.
Bodily Systems and Behavior
In describing these systems, we provide a brief overview of how each functions and where behavioral influences are most likely to be manifest, review representative research linking behavioral and biological changes to one another and to health outcomes, and describe some potential ameliorative strategies for minimizing health impairing effects.
Regulatory systems, primarily the nervous and endocrine systems, serve communication and integration functions by stimulating or inhibiting the activity of other systems. These activities are typically in the service of maintenance of homeostasis or balance, or of coping or adaptation in response to environmental inputs signaling threats or opportunities. They include detection and interpretation of external and internal events as well as direction of responses to these events and evaluation of the efficacy of these responses. Changes in external or internal environments are detected by sensory systems that relay information to the brain, a central processing unit that interprets the information sent to it and sends general and specific directions back out to the periphery that determine how the organism copes or reacts. These incoming signals, called afferents, constitute a major determinant of how events are appraised or resolved. They evoke both novel and well-established appraisals of responses, influenced by past experience and severity of threat or demand, and send action directives to other regulatory centers and specific organs. These instructions for coping and adaptation are called efferents and typically terminate in action by target systems.
The hypothalamus plays a primary regulatory role for the nervous system. Among its many important functions is its control of the sympathetic nervous system (SNS) and parasympathetic nervous system (PNS), which are coordinated to regulate arousal and reduction of arousal in response to change or threat. The SNS and PNS interface with nearly every organ or organ system in the body and help to control such diverse activity as respiration, digestion, heart rate, and energy storage or release. Given these broad connections, it is not surprising that changes in activity in the nervous system have been linked to various disease states and processes.
Of some interest is the close interaction between the two major regulatory systems in the body, the nervous and endocrine systems. The central nervous system (CNS) consists of the brain and spinal cord and serves an integration function. The peripheral nervous system includes all ganglia, neurons, and synapses that detect and relay information to the spinal cord and brain. Excitation of a pathway stimulates activity that eventually sends regulatory messages to effector organs such as skeletal muscles. Afferent excitation sends these messages to the CNS; efferent excitation involves information from the CNS going to the periphery.
The endocrine system is closely connected to the nervous system and often works in concert with it to achieve desired objectives. For example, neurally mediated activation of the sympathetic nervous system lasts a relatively short time and needs to occur periodically to maintain an aroused state. However, endocrine activity can augment sympathetic activity and is reliably associated with the same modulation of organ systems activity as nervous system activation. The chief difference is that endocrine-mediated arousal has the potential to last longer and most cases of simultaneous activation of nervous system and endocrine pathways appear to be designed to intensify and extend arousal.
These regulatory systems are organized in ways that allow them to play key gatekeeper and regulatory roles in orchestrating response to nearly any stimulus or set of conditions. Most incoming information from external sensors (e.g., eyes, ears, touch) is transmitted to the brain through the spinal cord, permitting immediate local activity or reflexive responses as well as more reflective, cognitively mediated responses. The efferents that emanate from the CNS in turn lead to both neural and endocrine changes that inhibit activity in some systems and stimulate activity in others. Control of blood pressure, for example, is responsive to external and internal events and often involves both local, non-CNS-mediated regulation as well as CNS-derived messages that inhibit some cardiovascular activity and increase other activity. This complex orchestration of responses is but one example of the extraordinarily adaptive nature of the body.
Stress and Regulatory Systems
A good example of how these systems work together is stress, a complex, biobehavioral process that heightens adaptive capacity and motivates action that will eliminate or accommodate threats or demands. Stress involves nearly every organ system in the body, is clearly mediated by the CNS, and is an often-cited pathway by which environmental events may affect health. It is also a useful exemplar of many of the biobehavioral mechanisms discussed in this chapter. For our purposes, we focus on the nature of regulatory signals in stress. Popular theory suggests that stress is appraised and experienced as a function of detection and interpretation of environmental changes or conditions. External events, ranging from disasters, weather, noise, crowding, and job stress or loss, to interpersonal conflicts and social stressors, pose threats to people, require some form of adaptation, and motivate and support coping that will reduce arousal and perceived danger. These events or changes are detected and relayed to the CNS where they are interpreted and responses formulated. Such appraisals trigger neural- and endocrine-mediated SNS activation, which can be relatively long-lasting and generalized. Many of the health effects linked to stress are associated with SNS activity mediated primarily by the endocrine system, which interfaces with nervous system activity at various points.
The cascade of biological changes associated with stress is accomplished primarily through two major endocrine systems: the sympathetic-adrenomedullary (SAM) and the hypothalamic-pituitary-adrenocortical (HPA) axes. The SAM is activated when sympathetic innervation stimulates the adrenal medulla and causes the release of epinephrine. Some norepinephrine is also released from the adrenal medulla but most norepinephrine is released by sympathetic nerve endings concentrated in several areas of the body. These hormones, also known as catecholamines, enter the bloodstream and, as described, can have wide-ranging and relatively long-lasting effects, augmenting and supporting direct SNS effects on cardiovascular and respiratory function, digestion, metabolism, skeletal muscle tone, and other activities. The HPA axis is not a component of sympathetic activation, representing a separate system that has effects on metabolism, inflammation and immune system activity. The system is activated by release of corticotropin-releasing factor (CRF) from the hypothalamus, which stimulates release of adrenocorticotropin hormone (ACTH) from the pituitary gland, which in turn travels to the adrenal cortex and stimulates the release of corticosteroids such as cortisol from the adrenal cortex. Neurons in the hypothalamus monitor the rate of change of cortisol in the blood, and when this monitoring indicates that circulating levels of cortisol are sufficiently high, exert inhibitory effects on further release of CRF from the hypothalamus. Receptors that mediate this negative feedback loop are also present in pituitary tissue, permitting control over ACTH release.
Secreted glucocorticoids facilitate the conversion of fats and carbohydrates to immediately usable forms of energy and other steroids help to govern mineral balance in several systems. Changes in glucocorticoids appear coordinated to mobilize or liberate stored energy in preparation for behavioral response to threat. They are also anti-inflammatory agents and appear to help regulate some immune system activities. It should be noted that the functioning of major neuroendocrine systems is more complex than described here. Although we have presented them as relatively autonomous, they overlap considerably, have some redundant functions, and regulate one another in a variety of ways. For example, CRH also influences activity associated with SNS such as epinephrine and norepinephrine release, and HPA products affect the synthesis and activity of other SNS components (Griffin & Ojeda, 1992).
Stress, Regulatory Systems, and Disease
Stress-related changes in nervous and endocrine system activity are believed to be adaptive in the sense that they can facilitate survival during acute stress, particularly when alertness, strength, or speed is critical. However, this benefit does not come without a cost. When activation of these systems is chronic or excessive, damage can occur. Chronic or excessive activation of nervous or endocrine systems may occur as a result of dispositional hyperre-sponsivity or as a result of ‘normal’ response systems operating under conditions of unusually frequent or intense stress. Stress can also affect endocrine activity by disrupting negative feedback loops involved in HPA regulation, as when chronically high levels of cortisol affect these feedback mechanisms (e.g., Sapolsky Krey & McEwen, 1984). Stress can also affect endocrine activity by altering gene expression. For example, the release of catecholamines during SNS activity can lead to an increase in the expression of genes that encode enzymes involved in the production of more catecholamines (Sabban & Kvetnansk, 2001). Because stress effects occur in systems that are actively influencing other systems, they may reverberate and directly or indirectly cause dysfunction in other systems. The damage stemming from chronic or excessive activation can contribute to pathophysiology as well as to the progression or exacerbation of existing disease conditions.
One example is cardiovascular disease. High levels of sympathetic or HPA activity can influence the development of cardiovascular disease as well as trigger coronary events among patients with pre-existing coronary artery disease (e.g., Rozanski et al., 1988). Evidence suggests that direct effects of sympathetic activation may be involved in arterial wall damage, that stress may increase platelet aggregation and clotting and may even precipitate ischemia or myocardial infarction (Smith & Ruiz, 2002). The mechanisms underlying these influences are described below (see section ‘Stress, cardiovascular function, and cardiovascular disease’).
Diabetes is another useful example of the clinical importance of stress-related endocrine changes. Diabetes mellitus can be divided into two forms, type I (insulin dependent) and type II (insulin independent). Both disorders are characterized by a buildup of glucose levels in the blood as a consequence of the absence or dysfunction of a hormone called insulin. Short-term symptoms can include increased thirst and urination, fatigue, and blurred vision, while diabetes can eventually cause severe damage to the retina, kidneys, and other tissues, potentially leading to loss of sight, kidney failure, leg and foot ulcers or gangrene, or diabetic coma (Guyton, 1991).
Glucose is a primary fuel that cells use to function. Its production is a major task of the digestive system as glucose is extracted or derived from foods that have been eaten. Glucose is taken up into cells through the action of insulin. In type I diabetes, pancreatic cells fail to produce enough insulin for glucose to be properly taken up and used in cells throughout the body; in type II diabetes, bodily cells gradually become resistant to the effects of insulin, causing similar problems with glucose uptake. As discussed, HPA activity facilitates the conversion of stored energy into available energy, which results in increased release of glucose into the bloodstream. Glucocorticoids can also inhibit insulin production. These effects present obvious problems for diabetic individuals who are already compromised in their ability to take glucose from the bloodstream into cells. By increasing blood glucose levels, or reducing insulin levels, stress-related HPA activity can affect symptom expression or onset as well as disease course and management in both type I and type II diabetes (Surwit & Schneider, 1983; Surwit & Williams, 1996).
Elevated glucocorticoids can also exert inhibitory effects on various metabolic processes, including protein and triglyceride synthesis and glucose and calcium transport. These processes can lead to loss of bone and muscle mass (Griffin & Ojeda, 1992). Excessive HPA activity can also have deleterious CNS effects. High levels of cortisol or other corticosteroids can damage the hippocampus, a critical brain area for memory storage, and have been associated with some cognitive deficits (e.g., Davis et al., 1986; Ling, Perry & Tsuang, 1981; Starkman, Giordani, Berent, Schork & Schteingart, 2001). Finally, recent evidence has also shown a positive association between cortisol and disease progression in HIV disease (Leserman et al., 2002).
As noted, excessive or pathogenic endocrine responses can result from ‘normal’ systems operating under extreme conditions, or from ‘normal’ conditions operating on hyperrespon-sive systems. While the former has been the focus of considerable research into links between endocrine functioning and types of stressful experiences or other environmental risk factors, dispositional differences are the focus of efforts to identify genetic or biological risk factors. Genetic or biologically based differences in patterns of endocrine responses between people of different genders, ages or ethnicities have been suggested to account for some differences in disease risk observed in these groups (Schooler & Baum, 2000).
Implications for Adaptation
Most descriptions of the functional utility of regulatory systems converge on a single theme: preparing and enabling the body to respond to challenges or threats in ways that will increase the likelihood of surviving this threat. From an evolutionary perspective, this speaks to the concept of adaptive value or reproductive fitness and may help explain why these survival-enhancing systems evolved. At the same time, an evolutionary perspective can provide insight into why biobehavioral processes contribute to disease vulnerabilities. In considering the evolutionary significance of biological and biobehavioral systems, it is a mistake to assume that the products of evolution by natural selection have been perfected over time for optimal health. Current biological or biobehavioral systems do not represent a pinnacle of evolutionary processes or a finished product. Natural selection does not result in perfectly designed solutions to adaptive challenges, but rather proceeds by selecting among randomly generated and often suboptimal alternatives that are bound by various forms of constraints. Furthermore, what is being selected for is not the maintenance of health per se, but rather reproductive success. Although health is obviously important for reproducing and leaving viable descendants, natural selection can and does at times favor biological or biobehavioral features or processes that have the potential to impair health. Examples include adaptive features or processes that have health-impairing costs or byproducts, those that provide immediate fitness benefits at the expense of health consequences that emerge after reproductive consequences have diminished or ceased, those that result from dysregulation or chronic activation of otherwise adaptive processes, or those that may have been adaptive in the past but compromise health when they operate in the context of a modern, evolutionarily novel environment.
These considerations can offer insight into why the physiological machinery underlying many biobehavioral processes is imperfectly designed with respect to human health, helping to explain the presence of biobehaviorally mediated disease vulnerabilities. They also suggest that the effects of these vulnerabilities may be modifiable and can be identified and isolated. To the extent that these vulnerabilities are species-wide, controlling disease risk will involve identification and management of behavioral or psychosocial correlates that may shift these vulnerabilities in one direction or another or cause clinical manifestations of these vulnerabilities to occur in the first place. For example, the flexible and modifiable nature of cognitive appraisal processes offers opportunities to lessen the health-impairing effects of some pathogenic biobehavioral processes, as illustrated by the use of cognitively based intervention strategies to successfully modify perceptions and effects of stress among at-risk individuals (e.g., Antoni et al., 2000).
The immune system serves a primary defense role for the human body against foreign materials, or antigens, such as invading bacteria and viruses or more difficult to detect altered or irregular host cells such as malignancies. It is a complex system consisting of natural barriers such as skin and mucous membranes as well as an array of immune organs and cells. As described, lymphatic nodes and vessels also play a complementary role for the immune system. Immune cells are white blood cells or leukocytes, and include lymphocytes (T cells, B cells, and natural killer cells), monocytes (which become macrophages), and granulo-cytes (neutrophils, basophils, and eosinophils). These cells have effector actions and participate in the production of intercellular messengers called cytokines. Together, these barriers, cells, and cell products attack pathogens, keep them out, or alter the bodily environment to make it inhospitable.
Deficient immune function is thought to create susceptibility to disease, much as any weak defense structure is vulnerable to infiltration. These deficiencies can occur at any point in the process of bodily defense. A cut provides a path into the body through the skin, a primary immune system barrier. Weakness in local or initial systemic defenses can allow infections to develop, and weakened cellular activity can result in persistence and spread of that infection. Immune function clearly influences susceptibility to and control of infectious diseases. There is also evidence that it may be involved in the etiology or course of cardiovascular disease (by affecting relevant inflammatory processes) and cancer (by affecting host immunosurveillance). Growing evidence of behavioral and psychosocial influences on immune function has led to the emergence of psychoneuroimmunology and given it prominence in health psychology research.
The immune system consists of at least two kinds of agents, those associated with the innate immune system and those associated with acquired immunity. Natural killer (NK) cells, macrophages, monocytes and neu-trophils, which comprise innate or non-specific immunity, need no prior exposure in order to target antigens. These cells represent a first line of defense against pathogens. Other immune cells, including T cells and B cells, require prior exposure to antigens before they can recognize them as targets. This represents what is known as specific or acquired immunity. After an initial exposure to an antigen, T cells are able to retain a specific memory of the antigen and mount a powerful defense against the same antigens in future encounters.
One important function of innate immunity is to mediate inflammation. Inflammation is a local response to tissue damage or the presence of microbial invasion or infection. An adaptive benefit of inflammation is that it can limit the extent of damage at the site of injury, either by promoting destruction and elimination of invading pathogens or by initiating tissue repair processes. In this way, these processes represent a first line of defense against various invasive agents. However, under some circumstances these same inflammatory mechanisms can contribute to rather than protect against disease.
Immune cells originate in bone marrow from pluripotent stem cells. Some of these cells migrate to the thymus where they develop into T cells, and are agents of cell-mediated immunity. These cells then circulate through the blood or lymph and reside in other immune organs such as the spleen or lymph nodes. B cells mature elsewhere and become agents of humoral immunity, which provides defense by producing antibodies. For the purposes of this chapter, it will suffice to say that cell-mediated and humoral immunity are sophisticated systems that require complex coordination and interaction between various types of cells. What will be emphasized here is the growing recognition of the relevance of behavior and emotion for the functioning of these defense systems.
Considerable research with both humans and animals demonstrates that the immune system shares bidirectional relationships with the central nervous system, through direct sympathetic innervation as well as endocrine system pathways (Ader, Felten & Cohen, 1991). Sympathetic connections communicate with a variety of immune organs including spleen, thymus, bone marrow and lymph nodes, representing direct physical links between the CNS and immune system. Indirect connections are suggested by evidence showing that immune cells, including T and B cells, contain receptors for several hormones and neuropeptides including corticosteroids, catecholamines, and opioid peptides (Plaut, 1987). At the same time, some immune system changes have corresponding effects on how people feel or how well they perform; increases in some cytokines, for example, are associated with feeling ill, increased inflammation, and general malaise (e.g., Watkins, Maier & Goehler, 1995). Nervous and endocrine system influences on some aspects of immune function have been confirmed by several areas of research (e.g., Irwin, Hauger, Jones, Provencio & Britton, 1990). Injections of epinephrine that in some ways mimic sympathetic arousal have been found to cause changes in the proportions of lymphocyte subsets in peripheral blood (Crary et al., 1983). This is consistent with evidence that high SNS reactors show the largest stress-related immune changes (Zakowski, McAllister, Deal & Baum, 1992) and with a number of studies suggesting that sympathetic stimulation causes increased migration of some immune cells from storage in lymphoid tissue into the bloodstream (e.g., Delahanty et al., 1996; Manuck, Rabin, Muldoon & Bachen, 1991). In addition, when medications are administered that block adrenergic receptors, stress-related immune effects are reduced or eliminated (e.g., Bachen et al., 1995). Biological bases for an interface between regulatory systems and the immune system have thus been demonstrated.
Stress and Immune System Activity
Assessing the effects of stress or other psychosocial influences on immune function and the measurement of immune competence in general are complicated by the complexity of the immune system and the related difficulty involved in selecting and interpreting samples. Quantitative and functional assessments of immune parameters can be performed in the laboratory, and several in vivo parameters can also be assessed. But it can be difficult to generalize or draw conclusions about the clinical or even biological significance of observed changes or fluctuations in these measures. Moreover, there are significant differences in the magnitude of immune changes following stress from one individual to the next, although there is some evidence that these responses are stable across time and stressor type for any given individual (Marsland, Henderson, Chambers & Baum, 2002; Marsland, Manuck, Fazzari, Stewart & Rabin, 1995). Nevertheless, research has begun to identify the types of psychosocial conditions that can elicit nervous- and endocrine-mediated immune changes, as well as those that may be associated with clinically relevant outcomes such as disease risk (Kiecolt-Glaser & Glaser, 1995).
Loneliness, poor social support, negative mood, disruption of marital relationships, bereavement, natural disasters and other forms of stress have been associated with changes in various aspects of immune function (reviewed in Cohen & Herbert, 1996). For the most part, these naturalistic stressors have been associated with suppression of both functional and quantitative parameters, particularly when the stressor is severe and sustained (e.g., McKinnon, Weisse, Reynolds, Bowles & Baum, 1989). On the other hand, acute laboratory stressors often precipitate immediate and short-lived increases in the numbers of circulating T cells and NK cells (e.g., Wang, Delahanty, Dougall & Baum, 1998) as well as transient increases in NK cell activity (Naliboff et al., 1991). Stress-related changes in some immune parameters may occur in a biphasic fashion, exhibiting temporary increases during or immediately following stress, followed by a more sustained post-stressor drop below baseline levels (Cohen, Delahanty, Schmitz, Jenkins & Baum, 1993). At least two studies have prospectively documented this pattern within a single sample, finding increased NK cell activity during acute stress and subsequent below-baseline reductions within 1 hour after stressor termination (Breznitz et al., 1998; Schedlowski et al., 1993). Stress-related suppression of NK cell activity may be especially relevant because natural killer cells can spontaneously destroy cancer cells (e.g., Herberman & Orlando, 1981) and may play an important role in defense against the progression of some cancers (Tajima, Kawatani, Endo & Kawasaki, 1996; Whiteside & Herberman, 1995).
Stress, Immune Activity, and Disease
The relationship between stress and disease risk has been more difficult to establish than have simple stress effects on immune function. Not all biological manifestations of stress will necessarily translate into clinically observable effects. Nevertheless, there is considerable support for the likelihood that stress increases risk of upper respiratory infections such as colds (e.g., Cohen et al., 1998; Cohen, Tyrell & Smith, 1991) and some research supporting the possibility that stress or other psychosocial factors increase the risk or progression of cancer or other illnesses (Cole, Kemeny, Taylor & Visscher, 1996; Ramirez et al., 1989). Studies that have intervened to reduce stress among cancer patients have found a variety of benefits associated with stress reduction, including some evidence of slowed disease course (Baum & Andersen, 2001; Fawzy et al., 1993; Spiegel, Bloom, Kraemer & Gottheil, 1989). Although the immune system is considered a likely mediator of these observed relationships, verification of immune mediation of stress-related health effects has been elusive (e.g., Cohen et al., 1998), and clinical relevance of stress-related immune changes has not been empirically established for any diseases. Researchers are still attempting to confirm the possibility that stress-related immune changes affect the incidence or progression of colds and influenzas (e.g., Cohen et al., 1991), HIV disease (e.g., Leserman et al., 1997), and cancer (e.g., Andersen, Kiecolt-Glaser & Glaser, 1994). In addition, asthma, arthritis, irritable bowel syndrome, and psoriasis may be influenced through similar psychoimmune mechanisms. Stress-related changes in immune function remain an unconfirmed but plausible biobehavioral pathway through which psychosocial factors influence disease etiology or progression.
Given the plausibility of this pathway, substantial research has been directed towards evaluating the extent to which the psychosocial milieu or psychological interventions can favorably modulate immune function. Although it may not be possible or even desirable to enhance immune function beyond normal or baseline levels, preventing or buffering stress-related immune decrements is considered beneficial. Research suggests that some psychosocial influences are influential in this regard. High levels of reported social support have been positively associated with functional immune parameters in a number of populations at risk for stress-related immune suppression, including cancer patients (Levy et al., 1990), spouses of cancer patients (Baron, Cutrona, Hicklin, Russell & Lubaroff, 1990), and individuals reporting high levels of general stress (Schlesinger & Yodfat, 1991). These effects have been attributed to the ability of a strong social support network to minimize or buffer stress-related decreases in immune function, perhaps by modulating stress effects on biological activities like endocrine function or effects on behaviors such as sleep or diet.
Such findings have stimulated a number of attempts to deliver beneficial psychosocial interventions to populations who are at risk for immune suppression in order to evaluate whether similar immune benefits might be generated. There have been some promising results. For example, among a sample of newly diagnosed malignant melanoma patients, a structured psychiatric intervention was associated with increased interferon-augmented NK cell activity at 6-month follow-up (Fawzy et al., 1990). Increased NK cell activity was also observed in geriatric subjects who underwent a psychosocial intervention (Kiecolt-Glaser, Glaser et al., 1985). In addition, a stress-management intervention attenuated the reduction in immunocompetence associated with notifying people of their HIV seropositive status (Antoni et al., 1991). These studies provide evidence that psychosocial interventions have the potential to buffer the immunosuppressive effects of stress. However a recent meta-analysis of the effects of psychological interventions on immune function indicated that these effects are modest, with intervention strategies emphasizing hypnosis and conditioning showing the strongest immune effects (Miller & Cohen, 2001). Although evidence for the effects of stress on immune function is well established, evidence for the clinical relevance of these effects and for the ability to modulate these effects is considerably less compelling.
Subcellular Defense Systems
The immune system may represent a primary mode of biological defense, but there are subcellular levels of defense that are also relevant. Cells periodically sustain damage from external agents or random events. This damage is ordinarily repaired and DNA restored to its prior state. These molecular defense systems act upstream of immune surveillance and serve to protect DNA by preventing and repairing damage that could lead to deleterious changes in cell structure or function. Changes in cellular structure or function are particularly relevant for the development of cancer, although they may also contribute to other forms of illness or disability. DNA damage can result from a variety of external insults as well as from biochemical interactions with endogenous chemicals, and may contribute to tumor initiation and progression (Jackson & Loeb, 2001). Autonomous DNA repair systems operate through a variety of pathways to maintain genomic stability by fixing this damage before it gives rise to somatic mutations (Wood, Mitchell, Sgouros & Lindahl, 2001). Failure to repair DNA damage potentially allows the propagation of mutations in future generations of cells leading eventually to clinical disease. Immune-mediated processes and apoptosis become the major defense against continued replication of altered cells once these mutations have occurred. Apoptosis is the process of programmed cell death, signaled by the nuclei in normally functioning cells. It is a process that requires metabolic activity and results in disintegration of cells into membrane-bound particles that are later eliminated by the body. In some ways apoptosis reflects a ‘failsafe’ system that is needed when basic defenses against propagation of mutations fall short and potentially dangerous mutations are poised to replicate and in some cases become malignant. Ironically, however, one of the characteristic features of cancer cells is their ability to avoid apoptosis; cancer cells appear to be resistant to these processes.
Research suggests that DNA repair systems and/or the extent of DNA damage may be moderated by stress. Various forms of stress have been associated with altered repair of damaged DNA in both human and animal studies (Cohen, Marshall, Cheng, Agarwal & Wei, 2000; Forlenza, Latimer & Baum, 2000; Glaser, Thorn, Tarr, Kiecolt-Glaser & D’Ambrosio, 1985; Kiecolt-Glaser, Stephens, Lipetz, Speicher & Glaser, 1985). These studies suggest additional pathways through which stress or other psychosocial factors may influence carcinogenesis beyond immune surveillance (Forlenza & Baum, 2000). In addition to processes like stress and mood states, overt behaviors such as tobacco use, alcohol use and sun exposure are also known to affect this defense system (Nakajima, Takeuchi, Takeshita & Morimoto, 1996; van Zeeland, de Groot, Hall & Donato, 1999). Exposure to ultraviolet radiation in sunlight inflicts genetic damage and is considered to be responsible for the majority of skin cancers (International Agency for Research on Cancer, 1992). Alcohol consumption has been found to inhibit endogenous DNA repair processes (Brooks, 1997). Tobacco smoke is also known to contain a variety of substances that initiate genetic damage, and may indirectly promote carcinogenesis through affecting activation and detoxification of xenobiotic compounds and the generation of oxidative damage from free radicals in cigarette smoke (Pryor, 1993).
Several systems can be considered in this category because they are wholly or partly involved in movement of nutrients, waste products, chemical messengers, and other elements of normal bodily function. The cardiovascular system consists of the heart, arteries, capillaries, and veins. It falls under the rubric of a transport system because it carries a variety of substances to places in the body that need them and transports waste products to places where they can be removed. The heart functions as a central pumping and collection unit, moving blood through blood vessels. This system transports nutrients absorbed through digestion, oxygen absorbed through respiration, and regulatory hormones and peptides released during endocrine activity. These substances are often destined for distant organs and tissues, and the pervasiveness and reach of this system is critical for timely and targeted delivery. The system reaches every region of the body and has some regulatory function, typically due to innervation or hormone receptors of the nervous and endocrine system.
There are four chambers in the heart: the right atrium, left atrium, right ventricle, and left ventricle. Once blood is oxygenated in lung tissue, it travels into the left atrium and left ventricle. The heart pumps blood from the left ventricle into the arteries, where it is dispersed throughout the body, where oxygen and nutrients are delivered and waste products are taken up. Venules and veins then return the oxygen-depleted blood to the right atrium and right ventricle, where it is pumped into the pulmonary region, and carbon dioxide in the blood is taken up prior to being exhaled. Heart valves ensure the continuous forward movement of this system.
Breakdown or dysfunction of the cardiovascular system can threaten the viability of cells, organs, and ultimately the organism. Some of the conditions that cause dysfunction have increased over the past 100 years, and cardiovascular disease has been the most common cause of death among people living in industrialized countries during the past century. This suggests that stress, diet, exercise, and a few other biobehavioral variables are major sources of risk for heart disease. There are several syndromes and kinds of cardiovascular disease that differ in dynamics and locus. Coronary heart disease (CHD), which involves arteries that provide nutrients to the heart muscle itself, and the narrowing of arteries in the brain or other areas of circulatory activity both involve the accumulation of plaque on the inner lining of arteries, a process known as atherosclerosis. Atherosclerosis typically develops insidiously over the course of many years, as blood flow to the heart is increasingly restricted. When coronary arteries are sufficiently restricted they are not able to deliver oxygen to the heart, causing ischemia, and this previously asymptomatic condition can result in clinical manifestations of CHD including chest pain, heart attack and sudden cardiac death. Other diseases and disorders involve blood clotting, hemorrhage, or other dynamics that can cause stroke, irregular heart rate, and other syndromes.
Stress, Cardiovascular Function, and Cardiovascular Disease
The pathogenesis of CHD is complex, involving various biochemical, inflammatory, and hemodynamic processes (Black & Garbutt, 2002; Ross, 1999). Likewise, biobehavioral mechanisms in the etiology or course of CHD may intersect with CHD pathophysiology at multiple levels. As noted, nervous and endocrine systems interface with the heart, causing changes that are ostensibly adaptive in the short term when faced with acute stress. However, increases in the frequency of exposure to glucocorticoids or catecholamines or other stress-related changes can precipitate a number of hemodynamic and immune inflammatory changes that can be pathogenic over the long run.
Stress-related hemodynamic changes may contribute to CHD processes by influencing heart rate, cardiac output, blood pressure, and clotting processes including coronary vasoconstriction and platelet aggregation (Smith & Ruiz, 2002). Increases in heart rate and blood pressure increase the force and pressure within vessels and arteries, which can contribute to shear stress that may damage the endothelium (Traub & Berk, 1998). Once the interior lining of the blood vessel has been damaged, the resulting lesions may constitute vulnerabilities for subsequent buildup of atherosclerotic plaque at these sites. In addition to causing hemodynamic changes that can damage arteries and coronary vessels, stress-related endocrine effects may be pathogenic by promoting the formation of blood clots or altering neural transmissions to the heart (e.g., Kamarck & Jennings, 1991). Stress-related SNS or HPA activity may also contribute by mediating immune and inflammatory processes, and growing evidence suggests that these processes are centrally involved in CHD. For example, products of HPA and SNS activity such as corticosteroids and cytokines can mediate inflammatory processes at sites of endothelial damage, promoting the adhesion of immune cells to the arterial wall (Black & Garbutt, 2002). This, then, contributes to narrowing of the interior of the vessel.
Behavioral and psychosocial variables have been clearly linked to CHD risk and progression in a variety of epidemiological and animal studies (Smith & Ruiz, 2002). Men who are especially impatient, hostile, and antagonistic or have recently endured stressful life events appear to have an increased risk for CHD (e.g., Helmer, Ragland & Syme, 1991; Mittleman et al., 1995). These individuals have been shown to have higher circulating levels of epinephrine and less favorable cholesterol profiles, and respond to stress with greater SNS and HPA activity (e.g., Williams, Suarez, Kuhn, Zimmerman & Schanberg, 1991). Experimental studies with non-human primates suggest that disruption of the social environment, heightened cardiovascular reactivity, and behavioral dominance increase CHD risk (Manuck, Marsland, Kaplan & Williams, 1995). The use of medications that block endocrine signals confirms that neuroendocrine pathways play a mediating role between psychosocial conditions and CHD processes (Manuck et al., 1995).
Identification of behavioral and emotional correlates of cardiovascular disease outcomes or processes, such as anger and hostility, has led to efforts to manage or reduce these processes and associated disease risk. The fact that behavioral risk is largely modifiable is encouraging because it permits a possible pathway to reduce overall risk for disease and disability. A variety of behavioral and cognitive strategies have been used among cardiac patients or those at risk, including relaxation training, cognitive behavioral stress management, meditation, group emotional support, and cognitive therapy (reviewed in Linden, Stossel & Maurice, 1996; Schneiderman, Antoni, Saab & Ironson, 2001). Despite some reports that such approaches can indeed reduce psychological and biological risk factors as well as the incidence of recurrent myocardial infarctions (e.g., Friedman et al., 1986; Linden et al., 1996), negative or inconsistent findings have also been reported (Rozanski, Blumenthal & Kaplan, 1999). While stress and some of its physiological consequences have clearly been linked to the etiology of CHD, the ability to successfully reduce CHD-associated morbidity or mortality by modifying these processes, while promising, is less certain.
Other Transport Systems
A second transport system is the lymphatic system, a network of nodes connected by low-pressure vessels that carry a fluid called lymph to and from bodily tissues and back into circulation again. As an alternative circulatory system, it conveys immune cells, growth factors, and the like, and removes spent immune cells and debris from invading pathogens. Lymph contains lymphocytes (immune cells) along with protein and fats. Approximately 3 liters of lymph seep from blood vessels into body tissues every day before being carried by lymphatic vessels, passing through lymph nodes, and being delivered back into the bloodstream.
The lymphatic system comprises lymphoid organs, which include the spleen, tonsils, Peyer’s patches, thymus, and lymph nodes. Lymph nodes contain a mesh of tissue in which lymphocytes and macrophages are housed. These nodes are a staging area for immune cells, are involved in the production of antibodies, and serve to filter, attack, and destroy antigens such as cancer cells, bacteria, and viruses. Lymph nodes are spread throughout the body, clustering where lymphatic vessels branch off such as in the armpits, neck, and groin. Although this is a transport system, it plays a crucial supporting role in bodily defense, representing another example of the overlapping and complementary nature of many of these systems. The development of new lymphatic vessels is also centrally involved in tissue repair and inflammatory reactions throughout the body, as healing of damaged bodily tissue requires the successful regrowth and reconnections of lymphatic vasculature (Oliver & Detmar, 2002).
As noted, acute stress can cause rapid changes in concentrations of immune cells in circulation, reflecting stimulation of nodes and lymphoid organs to release immune cells. There is also some animal research showing increases in flow of lymph fluid and immune cell output from lymphatic nodes following administration of acute pain or adrenaline injections, suggesting that lymphocytes in regional tissues can be mobilized in response to acute stress (Shannon, Quin & Jones, 1976). Additional research is needed to better understand the clinical implications of stress-related alterations in lymphatic system activity and how this interfaces with the well-documented stress-related changes in various immune parameters.
The digestive or gastrointestinal (GI) system transforms consumed food into usable resources. It includes the mouth, salivary glands, esophagus, stomach, and intestines. Digestion takes place in stages, beginning in the mouth where food is broken down through chewing and the action of salivary enzymes. After food is swallowed, it is pushed to the stomach by contractions called peristalsis, where gastric acids and enzymes further break down the food. Breakdown is completed as food then passes through the small and large intestines, where secretions from the pancreas, liver, and gallbladder aid in the process of absorption through the lining of the intestines and into the bloodstream.
The respiratory system is a second resource system. It works closely with the cardiovascular system to remove carbon dioxide from the blood and replenish it with oxygen. These processes occur in the lungs, the primary organ in the respiratory system. The lungs are large organs that contain about 1000 square feet (100 m2) of surface area. A normal breath brings in approximately 0.5 liter of gas, while maximum adult capacity is between 5 and 6 liters. The nose, mouth, pharynx, trachea, diaphragm and abdominal muscles also play supporting roles in respiration.
Air is inhaled through the nose or mouth with the help of the pharynx, a muscular organ at the back of the throat, and the diaphragm, a muscle at the bottom of the rib cage. The activity of the diaphragm causes the rib cage to raise, increasing lung volume and causing a low-pressure area. This pressure brings air into the lungs through the trachea or windpipe, which branches into bronchial tubes and then smaller bronchioles, terminating in small air sacs called alveoli that have permeable membranes allowing the exchange of oxygen and carbon dioxide. The exchange of these gases is carried out through diffusion. The diffusion gradient is maintained by inhalation, which renews air in the alveoli with an oxygen concentration near levels of atmospheric air, and alveolar capillaries, which supply blood from circulation with a low oxygen concentration and high carbon dioxide concentration.
The respiratory system is controlled by both voluntary and involuntary mechanisms, overlapping to some extent with regulatory systems that function to promote optimal levels of blood gases. Part of the brain known as the medulla monitors carbon dioxide levels in the blood in order to maintain respiration at this optimal rate, and can independently initiate respiration if necessary.
The leading respiratory disease is chronic obstructive pulmonary disease (COPD). COPD is a progressive, largely irreversible disorder marked by airflow limitation associated with inflammatory responses in the lungs to noxious substances (National Institutes of Health, 2001). These disorders include or involve emphysema, small airway inflammation and fibrosis, chronic bronchitis, and mucus gland hyperplasia. Cigarette smoking is the primary risk factor for COPD, lung cancer and other respiratory disorders (National Institutes of Health, 2001). Despite clear evidence that there is a causal relationship between this preventable behavior and illness and death, efforts to get people to quit smoking have largely been discouraging. The best current approaches to cessation among healthy individuals, which combine pharmacological and behavioral strategies, result in modest 1-year cigarette abstinence rates of 20-25 per cent (Centers for Disease Control, 2000). The effectiveness of these interventions is likely hampered by the strong reinforcing and addictive properties of nicotine (e.g., Gamberino & Gold, 1999), and by psychosocial and behavioral factors that play a role in maintaining smoking behavior (e.g., Hiatt & Rimer, 1999). For most individuals who smoke, the possibility of remote health benefits may not be sufficient to outweigh the immediate reward and alleviation of withdrawal symptoms associated with continued use of tobacco products.
Stress, Resources System Activity, and Disease
In addition to smoking, stress and other psychosocial or behavioral influences such as negative emotional states can affect the activity of these resource systems, often at multiple levels or stages. For example, these influences can alter choice of food intake, disrupt digestion by inhibiting saliva production, slow the flow of food into the stomach, alter the concentration of digestive acids or enzymes, alter blood flow to the stomach, slow nutrient absorption through the intestinal lining, and alter contractile or other digestive activities in the intestines (Astertita, 1985). These changes can contribute to disease or disease processes throughout the digestive tract, including an increased risk of developing ulcers, ulcerative colitis, and other GI problems (Sapolsky, 1998).
Models of Biobehavioral Influence
As described, ambient or stable environmental conditions and chronic conditions such as stress affect a number of aspects of normal human functioning. Temperature, weather, social conditions such as support or interpersonal conflict, socioeconomic status, ethnicity and minority status, gender, age, and conditions characterizing work or home can affect mood, behavior, and biological processes. Different models order these elements differently, so one theory might argue that these conditions produce negative emotional states or cognitive appraisals that then affect biological and behavioral responses, while others would suggest that negative affective states occur in part because of experienced arousal of biological systems and serve a motivational function. Behavior change is ordinarily due to efforts to adapt to these conditions, either through direct action and manipulation, flight, or accommodation to them, but may also occur as a ‘byproduct’ of other dynamics.
General Biobehavioral Model
Biological changes in a stress model would be more central to the process and would trace back to detection and appraisal processes in the CNS. This central processing would be a source of the negative affective states and peripheral biological activation that often appear to emerge simultaneously. Placed in this context, biological changes would also be modified by personal attributes of people acting in a particular setting, either by modulation of biological systems themselves (as in, for example, genetic modulation of defense or transport system strengths or weaknesses) or by altering the ways in which environmental or social conditions are appraised or experienced.
There are several important implications of viewing biobehavioral influences in this way. First, the importance of considering personal attributes in these equations cannot be overstated. Basic genetic predispositions are key elements of tendencies to experience particular moods, to behave in specific ways, and to have predispositions for vulnerability or ‘immunity’ to certain illnesses. Ethnic genetic variability is known; some diseases, such as sickle cell anemia among African Americans or Tay Sachs disease in European Jews, are far more prominent in susceptible groups. Age affects both biological responses (e.g., response systems may become inelastic, or total response may be diminished as in immunosenescence); gender also influences these systems. Clearly, behavioral and affective differences are important as well, and, together with well-known differences in disease vulnerabilities, biological reactivity, and response to treatment, contribute to disparities in the burden of chronic diseases. African Americans generally experience more cancer morbidity and mortality as well as greater risk of hypertension than do white people in the US (Jemal, Thomas, Murray & Thun, 2002). Men have greater risks for most diseases and exhibit larger regulatory and transport system changes when provoked than do women (e.g., Canto et al., 2000). Age increases the likelihood of mutations in cells that can produce malignancies, and is associated with declines in immune defenses that may heighten vulnerability to infectious illness. These kinds of deficits are correlated with systemic wear and tear that can eventually produce dysfunction.
Another source of personal influence includes habitual behavior patterns or personality variables, including hostility, optimism, and emotional expressiveness. To some extent these styles probably reflect genetic variability but to some degree they appear to be learned as well. Hostility can be conceptualized as a set of cynical attitudes that increase proneness to anger (Smith, 1992), and is commonly assessed with the Cook-Medley Hostility Inventory (Cook & Medley, 1954). Hostility has been clearly linked to differences in cardiovascular reactivity and to risk for cardiovascular disease (Smith & Ruiz, 2002). Early studies demonstrated that these effects are independent of hypertension or other cardiovascular risk factors (e.g., Barefoot, Dahlstrohm & Williams, 1983; Barefoot, Dodge, Peterson, Dahlstrom & Williams, 1989).
Optimism is another dispositional variable that has received considerable research attention in studies of coping and illness. Most of this research has used the Life Orientation Test (LOT: Scheier & Carver, 1985) to measure optimism. The LOT is an instrument that assesses the extent to which people expect or believe that things will work out for the best. These studies generally show that optimism is associated with better physical and psychological wellbeing (Scheier & Carver, 1992). The mechanisms through which optimism influences health have not been established. There is some evidence that differences in stress responses or cognitive styles may be involved (Carver et al., 1993; MacLeod, Williams & Bekerian, 1991). Optimistic individuals demonstrate a bias towards control and efficacy, variables that appear to sustain favorable expectations in the face of ambiguity, as well as use action-oriented problem solving strategies.
Some evidence also suggests a link between styles of emotional expression and disease risk. For example, women with a repressive coping style who fail to express strong negative emotions such as anger appear to be at moderately increased risk for cancer (Greer & Morris, 1975; McKenna, Zevon, Corn & Rounds, 1999). Other forms of behavioral or emotional inhibition have also been associated with unfavorable health outcomes (e.g., Cole et al., 1996). Haynes and colleagues found relationships between suppressed hostility and coronary heart disease in analyses of Framingham Study data (Haynes, Feinleib & Kannel, 1980). Conversely, interventions aimed at eliciting written expression of emotionally upsetting events have shown some health benefits (Smyth, 1998). Additional research is necessary to establish the extent to which disposition or personality influences disease and the biological mechanisms that mediate these possible effects.
Joint determination of behaviors that are beneficial or harmful for health is also important in this context. Behavioral responses to environmental change, for example, are motivated by the negative affective states that are associated with threatening, harmful, or dangerous events. These affective states are a joint function of negative cognitions and physical discomfort associated with arousal or activation of key regulatory systems. Because the organism ‘feels bad’, it is motivated to cope, either by attacking the problem or by fleeing from it. At the same time these responses are shaped and supported by biological sequelae of threatening or harmful appraisal. Coping, whether adaptive or maladaptive (or simply ineffective), can contribute to pathophysiology as byproducts of coping or as poor fit between coping repertoires and stressors; to the extent that coping is unsuccessful, that the readying responses associated with stress are not helpful or are harmful to adaptation to a given set of conditions, or that systems are hyperengaged for long periods of time, specific damage and non-specific wear and tear that can precipitate or contribute to illness are likely.
These models also generate a number of testable hypotheses and offer multiple pathways for intervention and moderation of risk. However, there are other useful models and/or perspectives in the general health and behavior field that provide key insights into mind-body interactions and health and that feature specific influential biobehavioral interactions as central pathways.
Diet and Disease
Diet is a behavior that intersects with a number of biological systems, with important implications for health and disease. The amount and type of resources taken into the body can obviously affect the ability of resource and transport systems to process and deliver critical substances or optimal levels of these substances to places in the body that need them. Food intake plays a particularly strong role in cardiovascular and cerebrovascular disease (e.g., Huijbregts et al., 1997), diabetes (e.g., Feskens, 1992) and certain cancers (e.g., Willett, 1996). The importance of diet in disease and mortality is suggested by the estimate that diet accounts for 35 per cent of all cancer deaths (Doll, 1992). Although the pathways through which diet affects disease risk are still being uncovered, general dietary effects on disease risk can be categorized as: (1) overconsumption of foods that can be health impairing when consumed beyond certain levels; (2) underconsumption of potentially health promoting or protective foods; and (3) the balance between amount of calories consumed and expended.
Consumption of specific foods is one key dietary behavior. Saturated fat and trans-fatty acids clearly increase cardiovascular and cerebrovascular disease risk (e.g., Hu et al., 1999). A high ratio of omega 6 to omega 3 fatty acids may also contribute to these diseases (Simopoulos, 1999). Specific types of dietary fat may also influence risk of non-insulin-dependent diabetes (Feskens, 1992) and certain cancers (Guthrie & Carroll, 1999), although these relationships are less clear. Similarly, sodium intake is known to increase the risk of hypertension, which is a risk factor for cardiovascular disease, stroke, and renovascular disease (He & Whelton, 1999). Members of remote rural populations, such as the Xingu Indians of Brazil, who do not add salt to food, have lower blood pressure, reduced lifetime incidence of hypertension, and increased stability of blood pressure over the lifespan compared to controls (Carvalho et al., 1989). Sodium intake may also be associated with risk for some cancers (Joossens et al., 1996).
Diets high in plant foods appear to reduce the risk of a number of diseases, including cancer (Kelloff et al., 1996; Potter & Steinmetz, 1996) and cardiovascular disease (Ness & Powles, 1997). However, many individuals do not consume these foods at optimal levels for protective benefits, and suffer disease risk as a result (e.g., American Institute for Cancer Research, 1997). In some Western societies, plant foods are consumed at less than one-third the level estimated among recent human ancestors (Peters & O’Brien, 1981).
Overweight and obesity promote a variety of chronic diseases and disease processes (reviewed in Must et al., 1999; Pi-Sunyer, 1998), contributing to approximately 300,000 deaths in the United States annually (Allison, Fontaine, Manson, Stevens & Van Itallie, 1999). Diet can contribute to overweight and obesity by leading to excess energy intake relative to expenditure, or a positive energy balance. A positive energy balance may promote cardiovascular and cerebrovascular disease by contributing to high levels of circulating free-fatty acids, excess fat stores, which increase hyper-lipidemia and hyperglycemia and promote oxidation and glycoselation processes, and elevated LDL cholesterol. A positive energy balance appears to promote the development of type II diabetes through contributing to insulin resistance. Muscle is more efficient at taking up glucose in response to insulin than is adipose tissue (DeFronzo, 1997). In genetically susceptible individuals, high levels of adiposity and low levels of lean muscle mass therefore contribute to the failure of insulin secretion to restore glucose homeostasis, leading to insulin resistance, glucose intolerance, and clinical type II diabetes (Eaton, Eaton & Cordain, 2002). Although more research is needed to clarify the relationship between energy balance and cancer risk, a positive energy balance may promote breast carcinogenesis due to the effects of adipose tissue on epithelial cell growth (Guthrie & Carroll, 1999), or on the production of estrogen among postmeno-pausal women (Mezzetti et al., 1998). A positive energy balance has also been associated with an increased risk of colon, endometrial, gall bladder, pancreatic, and other forms of cancer (Ford, 1999; Pi-Sunyer, 1998; Wolk et al., 2001).
Exercise or physical activity appears to reduce disease risk by helping to restore a balance between energy intake and expenditure. Additional ways in which exercise may be beneficial may include immediate reduction of stress (Scully, Kremer, Meade, Graham & Dudgeon, 1998), reduction of subsequent physiological reactivity to stress (Hinde, Moraska, Gaykema & Fleshner, 1999), reduction of LDL cholesterol or increases in HDL cholesterol (Williams, 1998), reduction of blood pressure (Williams, 1998), increases in endogenous free radical scavengers (Ji, 1999), alteration of hormonal levels (Thune, Brenn, Lund & Gaard, 1997) or alteration of aspects of innate immune system functioning (Woods, Davis, Smith & Nieman, 1999). The clinical relevance of these effects is suggested by studies that link physical activity to reduced risk of cardiovascular disease (Bijnen et al., 1998) and diabetes (Baan, Stolk, Grobbee, Witteman & Feskens, 1999) after controlling for body mass. Exercise may have independent protective effects for certain cancers as well (Longnecker, Gerhardsson le Verdier, Frumkin & Carpenter, 1995; Thune et al., 1997).
A cluster of conditions referred to as ‘metabolic syndrome X’ (Hansen, 1999) may represent another biological mechanism through which behavior or lifestyle contributes to disease risk. As typically characterized, this syndrome includes glucose intolerance (including type II diabetes), hyperinsulinemia/insulin resistance, abdominal or visceral obesity, dys-lipidemia, and hypertension (Hansen, 1999). There is still some debate about whether this syndrome actually represents one distinct risk factor or process, what its core pathogenic elements are, and the relative contribution of genetic and environmental influences on its development (e.g., Matsuzawa, Funahashi & Nakamura, 1999; Zimmet, Boyko, Collier & de Courten, 1999). Nevertheless, there is evidence that this syndrome or its representative conditions play an etiological role in cardiovascular disease and contribute to neuropathy and liver and kidney damage or dysfunction (Hansen, 1999; Lempiäinen, Mykkänen, Pyörälä, Laakso & Kuusisto, 1999). As reviewed, exercise, diet, and other behaviors can influence important components of this syndrome such as blood lipid profiles and obesity, suggesting that the disease mechanisms involved in metabolic syndrome X may transmit some behavioral influences on disease risk.
Evolutionary and Genetic Considerations
Biological systems typically operate in the service of homeostasis and survival, but they can also represent pathways to disease and disability, and biobehavioral influences can contribute substantially to these outcomes. The mechanisms by which biobehavioral processes exert these effects, as well as the reasons why some individuals are more at risk than others, are being uncovered by research in fields like health psychology, developmental biology, and molecular and behavioral genetics. The reason why human biological systems appear universally vulnerable to pathogenic biobehavioral processes is a different sort of question, and one that is likely to require a better understanding of the development and interaction of these systems over evolutionary time. Of particular interest is the process by which regulatory systems became interconnected with other biological systems, since these networks seem to constitute disease vulnerabilities as well as adaptive capabilities. While the overlapping of different biological systems is beneficial in that it allows for networks with extensive and sophisticated communication and regulation capacities, this engineering scheme seems less optimal when disease processes are initiated as a consequence of system design or inelasticity.
The fact that many of these overlapping systems are layered on top of one another belies a fundamental reality of natural selection that may help explain the presence of these vulnerabilities: meaningful increases in phylogenetic complexity proceed by building on top of or adding to existing structures. That is, species cannot scrap their existing design and start again when environmental change calls for new modes of responding or when individual members begin to wish for improved health and longevity or other outcomes that natural selection has not optimized. Existing design can impose significant constraints on the subsequent structure and adaptability of organisms and the systems that comprise them. It remains to be determined whether biobehaviorally mediated disease vulnerabilities ultimately reflect design constraints imposed during the course of our evolutionary heritage, costs or byproducts of adaptive processes, effects of evolutionarily novel inputs, or other manifestations of the imprecision of evolutionary processes. It can be argued that stress responses and some behavioral motivational systems can simply operate at cross-purposes with health and longevity, reflecting natural selection pressures that do not place premiums on these outcomes. Many physiological consequences of stress appear adaptive in response to acute threats, but carry costs that are especially evident when this activation is prolonged or severe. Inflammation and related immune processes are adaptive and health protective in many instances of acute injury or microbial threat, but can promote cardiovascular disease when operating in response to coronary vessel lesions or in conjunction with evolutionarily novel blood triglyceride or cholesterol levels. Appetites for sugary, salty, and fatty foods may promote sustenance in environments of scarcity, but can contribute to disease in evolutionarily novel environments that provide an abundance of these foods. These accounts comprise considerations of species-typical vulnerabilities or ultimate (evolutionary) causes that complement and inform descriptions of proximate relationships. Proximate explanations, which center on mechanism and development, and evolutionary explanations, which center on function and phylogeny, are complementary but independently necessary components of thorough explanatory models (Nesse, 1999).
The selective pressures faced by our human ancestors are reflected in the modern genome, which transmits evolutionarily shaped vulnerabilities. In this sense, current human genes place fundamental constraints on the health and longevity of all people, including those imposed by apoptosis and other normal cellular processes involved in ageing or senescence (Wick, Jansen-Durr, Berger, Blasko & Grubeck-Loebenstein, 2000). Many forms of disability and disease that manifest late in life are related to these cellular processes and are considered consequences of normal but suboptimal somatic mutation or repair capabilities.
Another way that genes are relevant for health and disease is by contributing to genetic differences between individuals. These individual differences can shift disease vulnerabilities in one direction or another, and clearly contribute to disease risk for a range of diseases including diabetes, myocardial infarction, Alzheimer’s disease, Parkinson’s disease, and asthma (Ruse & Parker, 2001). These influences are the subject matter of molecular genetics and epidemiology, which can identify genetic variants through biochemical and cellular studies, linkage and positional cloning, candidate gene studies, or genome-wide studies (Day, Gu, Ganderton, Spanakis & Ye, 2001). Genetic differences between individuals also define the parameters within which biobehavioral systems function (Nesse & Berridge, 1997). For example, genetic differences appear to contribute to individual differences in cardiovascular responses to stress (Hewitt & Turner, 1995), diet and activity patterns (Faith, Johnson & Allison, 1997; Reed, Bachmanov, Beauchamp, Tordoff & Price, 1997), resting energy expenditure and substrate utilization (Goran, 1997) and substance use (Blum et al., 1996; Heath & Madden, 1995). As such, genetic differences between individuals can influence risk for biobehaviorally mediated disease vulnerabilities.
The range of genetic variation that exists among modern humans may reflect a variety of processes, including variation in selective pressures exerted over differing evolutionary environments or passive maintenance of variation that has had little or inconsistent fitness consequences. In addition, the presence of clinically meaningful genetic differences in biobehavioral systems may reflect a number of processes at work during human evolution, such as ecological imperatives for substantial physical activity, constraints on availability of health-impairing foods or substances of abuse, or earlier death by unrelated causes. Such factors may have masked potentially deleterious gene effects, contributing to the maintenance of genes that may have otherwise been selected out.
This chapter has broadly described some biological mechanisms underlying health and behavior relationships. We have emphasized biobehavioral interactions for several reasons, most related to the complexity of the relationships that are being modeled. The explanatory power of these interactions, particularly when unpacking the layers of influence in the human body and the array of determinants of behavior, lies in careful consideration of the multiple, partially overlapping effects of biobehavioral interactions on outcomes such as health and wellbeing or disease and disability. It is clear that the interaction of biobehavioral processes with biological systems can affect pathophysiology in many instances, and efforts to control or undermine these disease pathways, while promising, have generally achieved modest success. Understanding ultimate as well as proximate causes of these processes requires a broader perspective than is typically applied in health psychology, but a more thorough understanding of causality may provide novel ways of thinking about solutions, and looking for ways to subvert evolutionarily shaped vulnerabilities should complement efforts to control empirically based psychosocial or biobehavioral antecedents of disease.