Research from animal and human studies has clarified that pathways to many disorders provide a basis for understanding and altering biobehavioral trajectories before disease is extant. One example is cardiovascular disease where a great deal has been revealed regarding the role of elevated cholesterol, homocysteine blood levels, hypertension, and abdominal fat, among other factors, in increasing risk for coronary atherosclerosis, stroke, and myocardial infarction. The detection of early warning signs with a high predictive value for later disease is an important aspect of prevention.
We use the term “predisease pathway” to describe the biological influences (e.g., genetic factors, endocrine and immune factors) and related links to behavioral, psychological, and social influences that precede morbidity and mortality. A primary message is the need to assess precursors to disease at points more distant temporally than has been examined in most previous research. Wider time horizons are required to understand early antecedents to later risk factors as well as the long-term etiological processes involved in multiple disease outcomes. Predisease pathways thus include a broad array of factors that affect the individual from conception (or before) through development and adulthood into later life. Illustrative influences include prenatal and early risk factors, along with a diverse array of psychological factors (e.g., control and efficacy, temperament, optimism, cognitive states, emotion regulation), behavioral factors (e.g., diet, exercise, smoking, alcohol consumption, drug abuse, sexual activity), and familial
and environmental influences (e.g., social ties and support, family stress, work conditions, community supports).
Equally important is the need to integrate these biological, behavioral, psychological, and social precursors to disease. The importance of such long-term developmental integration has recently been underscored by Worthman (1999): “We do not have an integrated model of human developmental physiology that can assist us in thinking about how ontogenetic changes may mediate environmental effects on adult health outcomes.” Emphasis on temporally distant and integrated assessment of psychological, social, and behavioral risk requires counterpart assessment of cumulative risk across multiple physiological systems, a topic that is addressed below.
CUMULATIVE PHYSIOLOGICAL RISK
Building on the guiding theme of integrative research (see Introduction), there is a need to assess physiological risk across multiple systems simultaneously. Illustrative of this objective is the broad framework of allostatic load (McEwen, 1998; McEwen and Stellar, 1993; McEwen and Seeman, 1999), which maintains that either repeated or continuous exposure to challenge or chronic underexposure and social isolation disrupts basic biological regulatory processes central to the maintenance of homeostasis and health. This model suggests that individuals (human or animal) exposed to challenge at vulnerable times (e.g., during early stages of pre-and postnatal development) or repeatedly during any period of life may experience overexposure to physiological responses that are outside normal operating ranges. Such overexposure comes about either because there are many challenges or because the turning on and turning off of the physiological responses is inefficient. This exacts a wear and tear termed “allostatic load.” Factors that may increase allostatic load include genetic predispositions, adverse experiences from early development, poor health behaviors (e.g., diet, exercise, and substance abuse), and exposure to stressful environmental conditions across the life span.
Through repeated efforts to adapt to stressful circumstances, the organism experiences a cumulative multisystem physiological toll, leading to cascading, potentially irreversible interactions between genetic predispositions and environmental factors. Over time, these cascades can contribute to large individual differences in dysregulation of the hypothalamic-pituitary-adrenal (HPA) axis, impaired immune function, altered cardiovascular reactivity, and ultimately stress-related physical and mental disorders (including chronic hypertension, coronary heart disease, diabetes, hippocampal atrophy and associated cognitive dysfunction; see Seeman et al.,
1997; Seeman and Robins, 1994; McEwen, 1998; Dhabhar and McEwen, 1996; McEwen et al., 1997).
Emphasis on allostatic load is not meant to overshadow the importance of diverse avenues of ongoing research on single systems of physiological risk (e.g., syndrome X and cardiovascular disease, dysregulation of the HPA axis and cognitive impairment). However, to date, there has been insufficient attention given to co-occurring risk across multiple physiological systems as well as to more temporally distant assessments of such risk. Because of its dual emphasis on cumulative risk across multiple physiological systems and cumulation across time, the concept of allostatic load is a particularly promising contributor to integrative health research, the theme of this report. Growing evidence of life course trajectories of comorbidity (i.e., individuals are ever more likely to suffer from multiple chronic conditions as they age) further underscores the importance of attending to co-occurring risk factors and their cumulation through time. Specifically, 45 percent of women and 35 percent of men ages 60-69 report two or more chronic conditions; figures rise to 61 percent of women and 47 percent of men ages 70-79 and 70 percent of women and 53 percent of men ages 80-89 (Jaur and Stoddard, 1999). Understanding the etiology of these co-morbid profiles requires attending to multiple co-occurring risk factors.
A provisional operationalization of allostatic load has been provided by Seeman et al. (1997, unpublished manuscript), using measures of the HPA axis (cortisol, dihydroepiandrosterone sulfate), sympathetic nervous system (epinephrine, norepinephrine), cardiovascular activity (systolic and diastolic blood pressure), metabolism and adipose tissue deposition (waist-hip ratio), glucose metabolism (glycosylated hemoglobin), and atherosclerotic risk (serum HDL and total cholesterol). Underscoring cross-system linkages, high allostatic load was found in the MacArthur Studies of Successful Aging to predict later life mortality, incident cardiovascular disease, and decline in cognitive and physical functioning. Looking to socioeconomic and psychosocial precursors, the same operationalization has been linked to cumulative economic and social relational adversity from childhood through age 59 in a 40-year longitudinal study (Wisconsin Longitudinal Study; Singer and Ryff, 1999). A slightly modified version of this operationalization of allostatic load has been associated with lower levels of education as well as greater hostility in the Normative Aging Study (Kubzansky et al., 1999). We underscore the provisional nature of the operationalization of allostatic load, and in Chapter 10 we discuss in detail the future research program needed to refine assessment of cumulative physiological risk. In Chapter 3 we also discuss the optimal functioning of these multiple physiological systems via the concept of allostasis.
CHARACTERIZING PREDISEASE PATHWAYS
In addition to indicators of allostatic load, markers of the existence of individual diseases are becoming better known, leading to expanded prospects for understanding predisease processes. Some of these prospects stem from genetic advances that enable researchers to focus on mechanisms of predisease in vulnerable populations (e.g., people with genetic risk for Huntington's disease or breast cancer). Other prospects come from expanding knowledge of the physiological mechanisms underlying disease and corresponding markers. For example, in the case of HIV infection, CD-4 T-cell counts and viral load are reliable indicators of asymptomatic infection progression. Consequently, studying factors that influence how quickly or slowly such indicators progress will lead to insights regarding disease processes, behavioral cofactors, and the dynamics of the infection trajectory.
Scientific advances such as these expand the window of time within which disease processes may be examined. They open investigations to studying interactions among genetic predispositions, prenatal and early life influences on health trajectories, behavioral factors (e.g., poor health habits), and social psychological states across the life span. Each of these are implicated in the early stages of single or multiple health disorders and point to protective factors that may delay the course of illness or potentially reverse these trajectories.
Prenatal and Early Life Risk Factors
Prenatal experience plays a critical role in interacting with the genome to shape brain development, and these epigenetic influences in intrauterine life confer a set of predispositions that act across the life span to affect vulnerability for many chronic diseases (see Chapter 4). Optimal prenatal environments produce beneficial effects and adverse environments produce deleterious effects on the developing brain. For example, third-trimester placental corticotropin-releasing hormone (CRH) predicts incidence of preterm birth and fetal growth restriction (Wadhwa et al., 1993). The concomitant HPA products (adrenocorticotropic hormone and beta-endorphins) may be related to dysregulation of the HPA axis in offspring and to fetal learning (Sandman et al., 1997).
Both animal and human studies document the important role that quality of caregiving and parenting behavior plays in the development of stress regulatory systems. Animal studies demonstrate that variations in maternal care permanently alter the expression of behavioral, endocrine, and cognitive responses to stress. For example, as adults the offspring of rat dams who provided more maternal care during their first 10 days of life showed
reduced plasma adrenocorticotropic hormone and corticosterone responses to acute stress, increased hippocampal glucocorticoid receptor messenger RNA expression, enhanced glucocorticoid feedback sensitivity, and decreased levels of hypothalamic corticotropin-releasing hormone messenger RNA (Caldji et al., in press). In humans, children exposed to parenting characterized by conflict, aggression, and neglect show disruptions in stress-responsive biological regulatory systems (sympathetic-adrenomedullary, HPA), poor health behaviors, and poor skills for emotional and social regulation (Repetti et al., in press). Abusive and neglectful family environments not only enhance risk for disease, injury, and premature death among children, they also interact with temperament to affect a broad array of mental and physical health disorders in adolescence and adulthood, including propensity for violence, depression, and risk for certain chronic diseases, including ischemic heart disease, some cancers, liver disease, and chronic obstructive pulmonary disease, a progressive disease process most commonly resulting from smoking (e.g.; Felitti et al., 1998; Walker et al., 1999).
Other early system dysregulations also are implicated in predisease pathways to adverse outcomes. Serotonergic dysregulation has been tied to enhanced risk for depression, suicide, and aggression, among other adverse outcomes. For example, depressed abused children showed elevated prolactin responses to a serotonin challenge, and those responses were correlated with clinical ratings of aggressive behavior and family history of suicide attempt (Kaufman et al., 1998). Animal studies suggest an important role for maternal behavior in moderating risk for serotonergic dysregulation. Here genetic risk factors play a role. For example, the short form of the 5-HTT allele, a gene related to serotonin transporter efficiency, confers low serotonin reuptake efficiency in monkeys, whereas the long form of the allele is associated with normal serotonin reuptake efficiency (Suomi, 1997). Monkeys with the short 5-HTT allele who were raised by peers (a risk factor for reactive, impulsive temperament) showed lower concentrations of the primary central serotonin metabolite (5-HIAA) than did monkeys with the long allele. But for monkeys raised by their mothers, primary serotonin metabolite concentrations were identical for monkeys with either allele. This pattern clearly suggests a protective effect of maternal behavior on expression of genetic risk for serotonin dysfunction.
Additional evidence for the interaction between genetic expression and early experience is provided by a study that randomly assigned rhesus monkey neonates selectively bred for differences in temperamental reactivity to foster mothers who were either unusually nurturant or within the normal range of mothering behavior (Suomi, 1987). Infants whose pedigrees suggested normal reactivity exhibited the expected patterns of biobehavioral development, independent of the relative nurturance of the fos-
ter mother. In contrast, dramatic differences emerged for the genetically highly reactive infants. Highly reactive infants cross-fostered to normal mothers exhibited deficits in early exploration and exaggerated behavioral and physiological responses to minor environmental perturbations. In adulthood they tended to drop and remain low in the dominance hierarchy (Suomi, 1991). Highly reactive infants cross-fostered to exceptionally nurturant females, in contrast, appeared to be behaviorally precocious. They left their mothers earlier, explored the environment more, and displayed less behavioral disturbance during weaning than both control (low-reactive) infants reared by either type of foster mother or highly reactive infants cross-fostered to normal mothers. In addition, when permanently separated from their foster mother and moved into larger groups, the highly reactive animals cross-fostered to nurturant mothers became adept at recruiting and retaining other group members as allies, and most became high dominant.
These primate studies are significant not only because they suggest an important role of parenting for modifying expression of genetically based temperamental differences but because they tie serotonergic dysfunction directly to behavioral attributes similar to those found in human offspring from abusive families. Specifically, monkeys raised without their mother (i.e., raised with peers) have difficulty moderating behavioral responses to rough-and-tumble play with peers, sometimes escalating those bouts into full-blown aggressive exchanges. Rearing with peers is also associated with a deficit of certain forms of prosocial behavior, such as less grooming among females. Peer-raised adolescent monkeys also show certain propensities for substance abuse, for example, requiring larger doses of the anesthetic ketamine to reach a state of sedation, and consistently consuming more alcohol and developing a greater tolerance for alcohol, compared to mother-raised monkeys. This pattern is predicted by their central nervous system 's serotonin turnover rates, and thus serotonin plays a key role in the normal regulation of these behaviors. In human studies, difficulty in moderating aggressive impulses, problems in developing and maintaining social relationships, and risk for substance abuse are among the outcomes most consistently seen in response to the family environment characteristics of hostility and conflict, deficient nurturing, and parental neglect (Repetti et al., in press).
The potential intergenerational transmission of these behavior patterns also warrants note. Monkeys raised by peers (rather than by their mothers) are significantly more likely to exhibit neglectful or abusive treatment of their own offspring (especially firstborns) compared with their mother-reared counterparts (Champoux et al., 1992; Suomi and Levine, 1998). Exposure to effective parenting reduces these behaviors in both animals (Suomi, 1987) and humans. For example, interventions that modify family
interaction patterns have demonstrated improvements in the behavioral concomitants of these dysregulations, such as drug abuse (Schmidt et al., 1996; see also McLoyd, 1998).
The underlying environmental conditions that give rise to these risky developmental pathways are increasingly understood and, again, parallels between human and animal studies are evident. Human conditions reliably associated with conflictual and abusive parenting include low or deteriorating socioeconomic status (SES), marital strife, and exposure to chronic stress. Comparable effects are seen in animal studies of caregiving under conditions of scarcity and exposure to other chronic stressors. A compelling example are the investigations of macaque mother-infant dyads maintained under one of three foraging conditions: low foraging demand (LFD), where food was readily available; high foraging demand (HFD), where ample food was available but required long periods of searching; and variable foraging demand (VFD), a mixture of the two conditions on a schedule that did not allow for predictability (Rosenblum et al., 1994). Exposure to these conditions over a period of months had a significant influence on mother-infant interactions, with the VFD condition the most disruptive. Mother-infant conflict increased in the VFD condition and infants of mothers housed under these conditions were significantly more timid and fearful. These infants showed signs of depression commonly observed in maternally separated macaque infants. As adolescents the infants reared in the VFD conditions were more fearful and submissive and showed less social play.
More recent studies have demonstrated the effects of these conditions on the development of neurobiological systems that mediate the organisms ' behavioral and endocrine/metabolic response to stress. As adults, monkeys reared under VFD conditions showed increased cerebrospinal fluid levels of corticotropin-releasing factor (CRF). Increased central CRF drive, which reflects increased activity of hypothalamic and extra-hypothalamic CRF systems, such as are found in the central nucleus of the amygdala, is consistent with the role of CRF in anxiety and depression, and this is exactly what was seen in adolescent VFD-reared animals. These findings provide a mechanism for the increased fearfulness observed in the VFD-reared animals, which may then be transmitted to a subsequent generation of offspring. In a recent study in humans (Heim et al., 2000b), women with a history of childhood abuse exhibited increased pituitary-adrenal and autonomic responses to stress in adulthood compared to controls. The findings suggest that early life stress results in persistent sensitization of the HPA axis to mild stress in adulthood, thereby contributing to vulnerability to psychopathological conditions.
The examples of predisease pathways that thus far are tied to interactions between genes and early environment concern primarily HPA, sympathetic-adrenal-medullary, and neurochemical functioning. Appropriately
timed measurements of serum or salivary cortisol, serum catecholamines, CSF levels of CRF and serotonin and catecholamine metabolites all provide clues as to the state of neural function in predisease states, and other measures need to be developed. The above-mentioned findings are well-researched examples of what is likely to be a broad array of system dysregulations in response to such gene-environment interactions. Other such system dysregulations may include but are not confined to alterations in dopaminergic regulation in noradrenergic and excitatory amino acid regulation, in the benzodiazapine system, in parasympathetic responses to stress, in metabolic functioning, and in reproductive functioning.
The parallels between animal and human investigations and the degree to which plausible pathways have already been mapped in these kinds of investigations underscores the importance of continued investigations of early routes to later disease from a perspective that examines the interactions among genetic, developmental, behavioral, social, and biological regulatory systems. For a superb review of symbiotic studies on humans and animals that exhibit such parallels, see Henry and Stephens (1977). For a lucid and more contemporary discussion, see Sapolsky (1994).
Three decades of psychosocial research have identified psychological states that are also implicated in predisease pathways. Such states interact with physiological predispositions, environmental stressors, and individual behavior to influence vulnerability to a broad array of illnesses, the trajectories of those illnesses, and their potential amelioration. These states include a sense of personal control or self-efficacy; the ability to regulate emotional experience; the development of social competence; temperamental states, such as optimism and neuroticism; cognitive states, such as positive or negative expectations regarding health; emotional states, such as depression and anxiety; and coping strategies, such as active versus avoidant coping. We highlight here some specific examples of the importance of these states in predisease processes.
A sense of personal control or self-efficacy is tied to a broad array of health-related outcomes. Self-efficacy is critically implicated in people's abilities to initiate and maintain good health habits, including exercise, breast self-examination, smoking cessation, and control of alcohol consumption (see Taylor, 1999, for a review). Interventions with patients awaiting noxious medical procedures, such as endoscopic examinations, reveal that even minimal interventions designed to enhance feelings of control (such as instructions in swallowing or attentional focus) permit these procedures to proceed with fewer complications (Johnson and Leventhal, 1974). Experimental investigations with institutionalized elderly have found
that control-enhancing manipulations (such as the ability to choose when to participate in activities or when to have visitors) are associated with improved health and longevity (Schulz, 1976; Langer and Rodin, 1976). One study found that rats implanted with a cancerous tumor preparation and exposed to inescapable electric shock were less likely to reject the tumor preparation than animals exposed to no shock or to escapable electric shock (Visintainer et al., 1983). Experiences of control or self-efficacy affect health outcomes across the life span and in manifold ways, ranging from initial vulnerability to adherence to treatment.
Emotion regulation is implicated in the early stages of health disorders, ties that have been most clearly demonstrated in investigations of anger and hostility. Hostility is a risk factor for coronary heart disease in adult men (Dembroski et al., 1985), and in women antagonistic hostility is related to high levels of low-density lipoprotein cholesterol, high levels of triglycerides, and a higher ratio of high-density lipoprotein cholesterol to total cholesterol (Suarez et al., 1998). Emotional suppression, a form of emotion regulation, reduced expressive behavior in adult men and women and produced a mixed physiological state characterized by decreased somatic activity and decreased heart rate, along with increased blinking and indications of increased sympathetic nervous system activity (Gross and Levenson, 1993).
The antecedents of emotion regulation are laid down in early childhood. A high frequency of negative interactions between parents and sons predicted the sons' later hostile attitudes and outward expressions of anger (Matthews et al., 1996). Children of hypertensive parents show elevated systolic blood pressure reactivity to angry exchanges between adults, which may be a precursor of later difficulties in stress management and risk for hypertension (Ballard et al., 1993). A large body of literature suggests that a genetic risk for heightened sympathetic-adrenal-medullary reactivity to stress is exacerbated by familial transmission of hypertension through repeated exposure to such hostile episodes, resulting in a hostile interpersonal style in adulthood (Ewart, 1991). The fact that hostility can be significantly modified in interventions for people diagnosed with coronary heart disease and the subsequent effects these interventions have on risk factors (Blumenthal et al., 1988) suggests the potential importance of modifying emotional regulation early.
Temperamental states such as optimism interact with physiological states to moderate predisease pathways. For example, optimism was associated with higher numbers of helper T cells and higher natural killer cell cytotoxicity in students undergoing the stress of first-year law school (Segerstrom et al., 1998). Because these immune changes can be precursors to significant clinical states, including depression, anxiety, and vulnerability to infectious disorders, these results suggest potentially protective effects of
optimism on predisease processes. This is a topic warranting further research, since the immune system can exhibit considerable fluctuation (plasticity) while remaining within normal operating range. A pessimistic explanatory style, namely the tendency to explain negative events in terms of internal, stable, global qualities in oneself, may represent a general risk factor for disease and early mortality. In one study, pessimistic explanatory style was measured at age 25 and health at ages 45 to 60 (Peterson et al., 1988). Those high in pessimistic explanatory style had significantly poorer health two to three decades later. A study of elderly people showed that those with a pessimistic explanatory style had compromised cell-mediated immunity (Kamen-Siegel et al., 1991), which may represent elevated risk for immune-related disorders in this vulnerable population.
Negative affective states adversely affect vulnerability and the courses of a range of health disorders. For example, chronic negative affect was associated with more severe respiratory illness (measured as mucous production) following experimentally induced exposure to a cold virus (Cohen et al., 1995). A meta-analysis revealed an association between prior depression and subsequent development of coronary heart disease (Booth-Kewley and Friedman, 1987). Among already diagnosed patients, those who became depressed in response to their diagnosed coronary artery disease were more likely to have a more debilitating course of illness and a repeat cardiac event, after controlling for other risk factors (Frasure-Smith et al., 1995).
The broad base of empirical evidence demonstrating the significance of these and related psychological states in pathways to many acute and chronic disease outcomes underscores the importance of continuing to explore their role in initiating, exacerbating, and moderating these predisease processes.
A set of behavioral risk factors, including poor diet, little exercise, promiscuous and/or unprotected sexual activity, smoking, alcohol abuse, and drug abuse, is associated with a broad array of diseases. Continued attention to the development and modification of these behaviors is essential.
Research suggests intergenerational transfer of risk behaviors that point to the necessity for early intervention. For example, mothers who smoked during pregnancy were more likely to have adolescent daughters who smoked, even after controlling for the mothers' postnatal smoking histories (Kandel et al., 1994); this suggests that nicotine or other substances in tobacco released by maternal smoking may have affected the fetus, perhaps through nicotinic input to the dopaminergic system. These changes appear to predispose the brain during a critical period of its development to the
subsequent addictive influence of nicotine more than a decade later. Other forms of substance abuse, such as alcohol and addictive drugs, appear to have similar prenatal effects, altering gene expression to produce lasting functional and structural changes in the brain. For example, intergenerational transmission of susceptibility to morphine and cocaine addiction has been demonstrated in animal studies (Beitner-Johnston et al., 1992).
People who use any type of drug are likely to use others (Capaldi et al., 1996; Donovan and Jessor, 1985; Kandel and Yamaguchi, 1993), which suggests that the drugs may share common biological substrates and/or serve common functions. Specifically, there is evidence that substance abuse helps individuals cope with dysregulated serotonin by facilitating release and/or impeding reuptake of the neurotransmitter. Dysregulation of serotonin and of the dopamine system is also tied to adverse early environments in animal and human studies. A likely pathway to clusters of poor health habits, then, is suggested by genetic risk interacting with challenging prenatal or early childhood conditions to produce serotonergic dysregulation, which is then “treated” through multiple poor health habits (especially those involving addictive substances) that represent efforts at self-medication.
Prospects for modifying these behaviors have expanded in recent years. For example, “stage models” of behavior change provide important insights into the modification of risk factors for disease and the corresponding predisease pathways they implicate. When trying to change health behaviors, people go through a series of stages that influence their receptivity to different kinds of interventions. This observation laid the groundwork for matching the type of intervention to the stage of readiness to change. For example, individuals still considering whether to change a behavior (such as stopping smoking) are best approached through persuasive communications that highlight the benefits of change, whereas those already committed to change may be best served by interventions that induce them to make explicit commitments to change and that provide training for bridging the gap between intentions and action. Similarly, strategies of relapse prevention are best directed to those facing the problem of long-term maintenance. This matching approach has been successfully applied to smoking cessation, quitting cocaine, weight control, modification of a high-fat diet, adolescent delinquent behavior, practice of safe sex, condom use, sunscreen use, exercise, and obtaining regular mammograms (Prochaska, 1994; Prochaska et al., 1992).
A lesson learned from secondary prevention concerns modification of multiple behavioral risk factors simultaneously (such as diet, exercise, and stress management). The strategy is to identify effective ingredients from studies of single-risk-factor behavior change programs and combine them into packages of effective multibehavior change programs. For example,
for people with insulin-dependent diabetes, multilevel programs aimed at the simultaneous problems of weight control, glucose management, diet control, exercise, and stress management have the strongest impact on disease course when they are put together in a package and the elements of each segment are linked to each other (Wing et al., 1986). Such findings have implications for understanding predisease pathways through the alteration of multiple health habits as well.
Even when effective intervention packages have been created in research settings, there may be a gap between scientific knowledge and clinical practice. Despite demonstrated efficacy of research therapies in treating a broad array of disorders in children, adolescents, and adults (e.g., depression, phobia, other anxiety disorders), factors demonstrated to be effective in research are often not incorporated into clinical practice, consequently producing far poorer outcomes (Weisz et al., 1995). What is needed is careful attention to the design and implementation of effective intervention packages, so that they can become the standard of care administered by primary care practitioners.
The social environment is critically important in influencing when disease processes will be initiated and what their course will be across the life span. Over 100 investigations of social ties and social support are testimony to the vital role these processes play in predisease pathways, as well as for disease course and recovery processes (Seeman, 1996).
The positive effects of the social environment on predisease processes begins very early. For example, in a review of more than 200 investigations of the determinants of adverse birth outcomes, social support from the baby's father or, in the case of unmarried girls, from their family members predicted beneficial birth outcomes, especially higher birth weight. These direct effects were indirectly mediated by less use of health-compromising substances such as cigarettes, alcohol, and drugs; better prenatal care; and lower stress (Dunkel-Schetter et al., in press). These health behaviors are inversely linked to risk of spontaneous abortion, fetal growth restriction, preterm delivery, and cognitive and motor deficits in offspring (Ness et al., 1999; Abrams, 1994).
The beneficial effects of social ties in adolescence and adulthood also are manifold. For example, individuals with higher levels of social ties were better able to avoid illness in response to experimentally induced exposure to a virus (Cohen et al., 1997). Such investigations suggest a significant role for social ties in predisease processes, specifically by reducing initial vulnerability, although the exact physiological and immunological mechanisms remain to be demonstrated. Nonetheless, there is substantial potential for
using such findings to develop effective support-based interventions for improving health across the life span.
The differential distribution of social resources across SES levels and by racial/ethnic groupings also merits additional study. So understudied and important are social processes to predisease and disease outcomes that we devote parts of three chapters to their consideration, one on personal social ties (see Chapter 5), one on the collective properties of health environments (see Chapter 6), and one on the health consequences of inequality (see Chapter 7).
Environmental stressors are known to have effects, often dramatic, across the life span. Chronic stress can precipitate insulin-dependent diabetes in animals with a genetic risk (Lehman et al., 1991), and in children stressful life events increase the symptoms of juvenile diabetes (Hagglof et al., 1991). Animals given a malignant tumor preparation and exposed to a chronic stressor (crowding) showed higher rates of malignant tumor development than those not exposed to stress (Amkraut and Solomon, 1977). Evidence like this implicates stress in predisease pathways and suggests the importance of both research designed to identify the specific mechanisms involved in these risks and the development of interventions to modify stressors that may precipitate risk conferred by genetic or behavioral vulnerability.
Adverse effects of chronic stress exposure can be seen as early as the prenatal environment. Maternal stress at 18 to 20 weeks' gestation has been found to significantly predict corticotropin-releasing hormone (CRH) at 28 to 30 weeks' gestation, even after controlling for CRH at 18 to 20 weeks (Hobel et al., 1999). Research on adult populations has identified potential early signs of disordered physiological functioning in populations undergoing intensely stressful events. Research on populations affected by Hurricane Andrew ties stress exposure to alterations in immune functioning, including reductions in natural killer cell cytotoxicity and numbers of CD-4 and CD-8 T cells (Ironson et al., 1997). There is, as yet, no clear evidence that these perturbations are linked to downstream health outcomes. This underscores, however, the importance of investigating the plasticity of the immune system and clarifying which immune system parameters should be included in future operationalizations of allostatic load. Post-traumatic stress disorder (PTSD) is associated in some studies with increased cortisol, epinephrine, norepinephrine, testosterone, and thyroxin functioning, changes that last over a long period of time (Wang and Mason, 1999). However, hypocortisolism has also been implicated in response to PTSD (Heim et al., 2000a; Thaller et al., 1999; Ehlert et al., 1999) and even for healthy individuals living under conditions of chronic stress (Heim et al., 2000b). Exposure to combat in war is tied to the development of deviant thyroid profiles. (Indeed, a history of traumatic stress is one of the
most reliable factors predicting thyrotoxicosis.) These alterations in thyroid functioning in response to traumatic stress are detectable more than 50 years later and thus appear to represent permanent alterations in physiological regulation in response to these intense stressors (Wang and Mason, 1999). Knowledge of such changes provides bases for hypothesizing and testing predisease pathways that may be implicated in adverse long-term health outcomes.
The workplace is an important source of both adverse and protective health effects across the life span. Among the workplace stressors that threaten health are low control over work activities, high work demands, low rewards, and combinations of these factors (Karasek and Theorell, 1990). Workers exposed to high job strain (high demands, low control) showed higher levels of ambulatory blood pressure three years later; moreover, changes in job strain were associated with changes in ambulatory blood pressure (Schnall et al., 1994). These patterns suggest that job strain can be an occupational risk factor implicated in the etiology of essential hypertension. A longitudinal study of 10,308 civil servants designed to identify work-related predictors of mental and physical health found that the work characteristics of high effort/low reward and poor social relationships were associated with poorer physical, psychological, and social functioning five years later, after adjusting for potential confounding factors (Stansfeld et al., 1998). These same work factors are especially implicated in the likelihood that workers will develop cardiovascular disease (Schnall et al., 1994).
Unemployment is a profoundly important stressor associated with high rates of depression, anxiety, and self-reported physical illness. Active coping strategies, that is the ability to respond to unemployment with plans and activities designed to alter the situation, can offset these ill effects (Turner et al., 1991). Unstable employment and employment at multiple unrelated jobs were associated with premature mortality, and the predisease mechanisms that underlie these effects merit exploration (e.g., Pavalko et al., 1993; Rushing et al., 1992). Workplace investigations also identified protective factors implicated in health trajectories. For example, the ability to develop social ties at work, including ties with supervisors, act as protective factors against adverse health and mental health effects of work stress (Buunk and Verhoeven, 1991).
Changing patterns of combining work and family roles have been understudied for their effects on health, especially their roles in predisease pathways. These include the expanding role of work and corresponding diminishing leisure time among workers and how work can compromise the development of personal resources, such as stable marriages and families and other social ties that might otherwise contribute to the protection of health. The effects on women of combining family and work roles and
the factors that influence whether and when these role combinations have adverse or beneficial effects on mental and physical health merits research attention (Repetti et al., 1989). For example, female caregivers for older demented adults experienced long-term changes in immunity and health as a result of combining this demanding role with existing home and work responsibilities (Kiecolt-Glaser et al., 1991). Nurses scoring higher on job demands had higher ambulatory blood pressure and heart rate during the workday and higher epinephrine in the evening, and married nurses had higher nighttime cortisol levels than did unmarried women (Goldstein et al., 1999). These findings suggest that both high work demands and a combination of job demands and role demands at home have identifiable potentially adverse effects on neuroendocrine and physiological profiles. These effects were stronger the longer these women had worked. For a superb discussion of women's work roles, their impact on mothers and their children and health consequences, all from an evolutionary perspective, see Hrdy (1999).
As these research examples attest, intensely stressful environmental conditions (e.g., the aftermath of a natural disaster, unemployment), geographic areas of intense chronic stress (such as crowded neighborhoods), and life domains in which intense change is occurring (e.g., changing patterns of combining work and family) present special opportunities to identify predisease pathways that, to date, have been understudied. These acutely and chronically stressful conditions create alterations in underlying physiology and neuroendocrine responses that may be precursors to adverse health and mental health conditions. Using such naturally occurring conditions and events as laboratories for investigating predisease pathways and identifying who is most at risk for adverse health effects can lead to substantial progress in understanding the parameters of predisease states, the mechanisms by which they develop, and subsequent progression to disease.
CONNECTING PREDISEASE PATHWAYS TO CUMULATIVE PHYSIOLOGICAL RISK
The preceding discussions of prenatal and early life risk factors, psychological states, behavioral factors, and environmental stressors address a wide array of precursors to illness and disease, focused largely on early life. As yet these diverse literatures have not been linked with cumulative physiological risk, as illustrated by the concept of allostatic load, in later life. Indeed, research on allostatic load has been conducted primarily in samples of adults and the aged. Thus, an important direction for future research on predisease pathways is to connect these diverse early life risks (biological, behavioral, psychological, environmental) to subsequent life course trajec-
tories of allostatic load. It is, in fact, such early precursors that are believed to initiate the cumulative physiological wear and tear that the concept of allostatic load is intended to capture. Investigations along these lines will require increased assessment of co-occurring components of physiological risk (e.g., function of the cardiovascular system, the HPA axis, sympathetic nervous system function, metabolic risk, immune function) early in life as well as their long-term sequelae in later life comorbidity.
A basic research initiative throughout NIH should focus on predisease pathways. Such an initiative would, of necessity, emphasize the unfolding interactions between environmental influences and gene expression over time. It should include the following topics:
identification of early markers of predisease states;
examination of their genetic and environmental origins through animal and human studies;
identification of behavioral risk factors in the exacerbation or amelioration of predisease pathways;
prioritization of experimental and longitudinal research to chart these trajectories across the life span;
focus on the mechanisms by which genetic influences, early life experiences, and behavioral and psychosocial risk factors across the life span interact, leading to accumulating physiological risk for a broad range of disease outcomes.
Abrams B. 1994 “Maternal nutrition” in Cresy RK , Resnik R (eds.). Maternal-Fetal Medicine: Principles and Practice, 3rd Ed. ( Philadelphia PA : WB Saunders ) 162-170 .
Amkraut A , Solomon GF. 1977 “From the symbolic stimulus to the pathophysiologic response: Immune mechanisms” in Lipowski ZJ , Lipsitt DR , Whybrow PC. (eds.). Psychosomatic Medicine: Current Trends and Clinical Applications ( New York Oxford University Press ).
Ballard M , Cummings EM , Larkin K. 1993 “Emotional and cardiovascular responses to adults' angry behavior and to challenging tasks in children of hypertensive and normotensive parents” Child Development 64 : 500-515 .
Beitner-Johnston D , Guitart X , Nestler EJ. 1992 “Neurofilament proteins and the mesolimic dopamine system: Common regulation by chronic morphine and chronic cocaine in the rat ventral tegmental area” Journal of Neuroscience 12 : 2165-2175 .
Blumenthal JA , Emery CF , Walsh MA , Cox DR , Kuhn CM , Williams RB , Williams RS. 1988 “Exercise training in healthy type A middle-aged men: Effects of behavioral and cardiovascular responses” Psychosomatic Medicine 50 : 418-433 .
Booth-Kewley S , Friedman HS. 1987 “Psychological predictors of heart disease: A quantitative review” Psychological Bulletin 101 : 343-362 .
Buunk BP , Verhoeven K. 1991 “Companionship and support at work: A microanalysis of the stress-reducing features of social interaction” Basic and Applied Social Psychology12 : 243-258 .
Caldji C , Francis D , Liu D , Plotsky PM , Meaney MJ. In press “The role of early experience in the development of individual differences in behavioral and endocrine responses to stress” in McEwen BS , Steller E (eds.). Handbook of Physiology: Coping with the Environment ( New York : Oxford University Press ).
Capaldi DM , Crosby L , Stoolmiller M. 1996 “Predicting the timing of first sexual intercourse for at-risk adolescent males” Child Development 67 : 344-359 .
Cohen S , Doyle WJ , Skoner DP , Fireman P , Gwaltney JM Jr , Newsom JT. 1995 “State and trait negative affect as predictors of objective and subjective symptoms of respiratory viral infections” Journal of Personality and Social Psychology 68 : 159-169 .
Cohen S , Doyle WJ , Skoner DP , Rabin BS , Gwaltney JM Jr. 1997 “Social ties and susceptibility to the common cold” Journal of the American Medical Association 277 : 1940-1944 .
Dembroski TM , MacDougall JM , Williams RB , Haney TL , Blumenthal JA. 1985 “Components of type A, hostility, and anger-in: Relationship to angiographic findings” Psychosomatic Medicine 47 : 219-233 .
Dhabhar F , McEwen B. 1996 “Moderate stress enhances, and chronic stress suppresses, cell-mediated immunity in vivo” Society for Neuroscience 22 : 536.3 – p1350
Donovan JE , Jessor R. 1985 “Structure of problem behavior in adolescence and young adulthood” Journal of Consulting and Clinical Psychology 53 : 890-904 .
Dunkel-Schetter C , Gurung RAR , Lobel M , Wadhwa PD. In press “Psychological, biological and social processes in pregnancy: Using a stress framework to study birth outcomes” in Baum A , Revenson T , Singer J (eds.). Handbook of Health Psychology ( Hillsdale NJ : Lawrence Erlbaum ).
Ehlert U , Wagner D , Heinrichs M , Heim C 1999 “Psychobiological aspects of posttraumatic stress disorder” Nervenarzt70/9 : 773-779
Ewart CK. 1991 “Familial transmission of essential hypertension, genes, environments, and chronic anger” Annals of Behavioral Medicine 13 : 40-47 .
Felitti VJ , Anda RF , Nordenberg D , Williamson DF , Spitz AM , Edwards V , Koss MP , Marks JS. 1998 “Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults” American Journal of Preventive Medicine 14 : 245-258 .
Frasure-Smith N , Lesperance F , Talajic M. 1995 “The impact of negative emotions on prognosis following myocardial infarction: Is it more than depression?” Health Psychology14 : 388-398 .
Goldstein IB , Shapiro D , Chicz-DeMet A , Guthrie D. 1999 “Ambulatory blood pressure, heart rate, and neuroendocrine responses in women nurses during work and off work days” Psychosomatic Medicine 61 : 387-396 .
Gross JJ , Levenson RW. 1993 “Emotional suppression: Physiology, self-report, and expressive behavior ” Journal of Personality and Social Psychology 64/6 : 970-986 .
Hagglof B , Bloom L , Dahlquist G , Lonnberg G , Sahlin B. 1991 “The Swedish childhood diabetes study: Indications of severe psychological stress as a risk factor for type I diabetes mellitus in childhood ” Diabetologia34 : 579-583 .
Heim C , Ehlert U , Hellhammer D. 2000a “The potential role of hypocortisolism in the pathophysiology of stress-related bodily disorders” Psychoneuroendocrinology 25/1 : 1-35 .
Heim C , Newport J , Heit S , Graham YP , Wilcox M , Bonsall R , Miller AH , Nemeroff CB . 2000b “Pituitary-adrenal and autonomic responses to stress in women after sexual and physical abuse in childhood” Journal of the American Medical Association 284 : 592-597 .
Henry JP , Stephens P. 1977 Stress, Health and the Social Environment: A Sociobiologic Approach to Medicine ( New York : Springer-Verlag ).
Hobel CJ , Dunkel-Schetter C , Roesch SC , Castro LC , Arora CP. 1999 “Maternal plasmacorticotropin-releasing hormone associated with stress at 20 weeks' gestation in pregnancies ending in preterm delivery” American Journal of Obstetrics and Gynecology180/1 : S257-S263 .
Hrdy, SB. 1999 Mother Nature: A History of Mothers, Infants, and Natural Selection ( New York : Pantheon Books ).
Jaur L , Stoddard S 1999 . Chartbook on Women and Disability in the U.S. ( Washington D.C. : U.S. National Institute on Disability and Rehabilitation Research ).
Johnson JE , Leventhal H. 1974 “Effects of accurate expectations and behavioral instructions on reactions during a noxious medical examination” Journal of Personality and Social Psychology 29 : 710-718 .
Kamen-Siegel L , Rodin J , Seligman MEP , Dwyer J. 1991 “Explanatory style and cell-mediated immunity in elderly men and women ” Health Psychology 10 : 229-235 .
Kandel D , Yamaguchi K. 1993 “From beer to crack: Developmental patterns of drug involvement” American Journal of Public Health 83 : 851-855 .
Kandel DB , Wu P , Davies M. 1994 “Maternal smoking during pregnancy and smoking by adolescent daughters ” American Journal of Public Health 84 : 1407-1413 .
Karasek R , Theorell T. 1990 Healthy Work: Stress, Productivity, and the Reconstruction of Working Life ( New York : Basic Books ).
Kaufman J , Birmaher B , Perel J , Dahl RE , Stull S , Brent D , Trubnick L , al-Shabbout M , Ryan ND. 1998 “Serotonergic functioning in depressed abused children: Clinical and familial correlates” Biological Psychiatry 44 : 973-981 .
Kiecolt-Glaser JK , Dura JR , Speicher CE , Trask OJ , Glaser R. 1991 “Spousal caregivers of dementia victims: Longitudinal changes in immunity and health” Psychosomatic Medicine 53 : 345-362 .
Kubzansky L , Kawachi I , Sparrow D. 1999 “Socioeconomic status, hostility, and risk factorclustering in the normative aging study: Any help from the concept of allostatic load? ” Annals of Behavioral Medicine 21/4 : 330-338 .
Langer EJ , Rodin J. Journal of Personality and Social Psychology 34 : 191-198 .
Lehman CD , Rodin J , McEwen B , Brinton R. 1991 “Impact of environmental stress on the expression of insulin-dependent diabetes mellitus” Behavioral Neuroscience 105 : 241-245 .
Matthews KA , Woodall KL , Kenon K , Jacob T. 1996 “Negative family environment as apredictor of boys' future status on measures of hostile attitudes, interview behavior, and anger expression” Health Psychology 15 : 30-37 .
McEwen BS. 1998 “Protective and damaging effects of stress mediators” New England Journal of Medicine 338 : 171-179 .
McEwen B , Seeman T. 1999 “Protective and damaging effects of mediators of stress: Elaborating and testing the concepts of allostasis and allostatic load” in Adler N , Marmot M ,
McEwen B , and Stewart J , (eds.). “Socioeconomic status and health in industrial nations” Annals of the New York Academy of Sciences 896 : 30-47 .
McEwen BS , Stellar E. 1993 “Stress and the individual: Mechanisms leading to disease” Archives of Internal Medicine 153 : 2093-2101 .
McEwen B , Biron CA , Brunson K , Bulloch K , Chambers W , Dhabar F , Goldfarb R , Kitson R , Miller A , Spencer R , Weiss J. 1997 “Neural-endocrine-immune interactions: The role of adrenocorticoids as modulators of immune function in health and disease” Brain Research Review 23 : 79-133 .
McLoyd VC. 1998 “Socioeconomic disadvantage and child development” American Psychologist 53 : 185-204 .
Ness RB , Grisso JA , Hirschinger N , Markovic N , Shaw LM , Day NL , Kline J. 1999 “Cocaine and tobacco use and the risk of spontaneous abortion” New England Journal of Medicine 340 : 333-339 .
Pavalko EK , Elder GH Jr. , Clipp EC. 1993 “Worklives and longevity: Insights from a life course perspective” Journal of Health and Social Behavior 34 : 363-380 .
Peterson C , Seligman MEP , Vaillant GE. 1988 “Pessimistic explanatory style is a risk factor for physical illness: A thirty-five-year longitudinal study” Journal of Personality and Social Psychology 55 : 23-27 .
Prochaska JO. 1994 “Strong and weak principles from progressing from precontemplation to action on the basis of 12 problem behaviors” Health Psychology 13 : 47-51 .
Prochaska JO , DiClemente CC , Norcross JC. 1992 “In search of how people change: Applications to addictive behaviors ” American Psychologist 47 : 1102-1114 .
Repetti RL , Matthews KA , Waldron I. 1989 “Employment and women's health” American Psychologist 44 : 1394-1401 .
Repetti RL , Taylor SE , Seeman TE. In press “Risky families: Family social environments and the mental and physical health of offspring” Psychological Bulletin. Rosenblum L , Coplan J , Friedman S , Bassoff T , Gorman J , Andrews M. 1994 “Adverse early experiences affect noradrenergic and serotonergic functioning in adult primates” Biological Psychiatry 35 : 221-227 .
Sandman CA , Wadhwa PD , Chicz-DeMet A , Dunkel-Schetter C , Porto M. 1997 “Maternal stress, HPA activity, and fetal/infant outcome” Annals of the New York Academy of Sciences 814 : 266-275 .
Sapolsky R. 1994 Why Zebras Don't Get Ulcers: A Guide to Stress, Stress-Related Diseases, and Coping ( New York : WH Freeman ).
Schmidt SE , Liddle HA , Dakof GA. 1996 “Changes in parenting practices and adolescent drug abuse during multidimensional family therapy” Journal of Family Psychology10 : 12-27 .
Schnall PL , Landisbergis PA , Baker D. 1994 “Job strain and cardiovascular disease” Annual Review of Public Health 15 : 381-411 .
Schulz, R. 1976 “Effects of control and predictability on the physical and psychological well-being of the institutionalized aged” Journal of Personality and Social Psychology 33 : 563-573 .
Seeman TE. 1996 “Social ties and health: The benefits of social integration” Annals of Epidemiology 6 : 442-451 .
Seeman TE , Robbins RJ. 1994 “Aging and hypothalamic-pituitary-adrenal response to challenge in humans” Endocrinology Review 15 : 233-260 .
Seeman TE , Singer B , Rowe JW , Horwitz R , McEwen BS. 1997 “The price of adaptation—Allostatic load and its health consequences: MacArthur studies of successful aging” Archives of Internal Medicine 157 : 2259-2268 .
Seeman TE , Singer B , Rowe JW , McEwen BS. Unpublished manuscript “Allostatic load as a measure of cumulative physiological risk: Is it more than syndrome X?”
Suarez EC , Bates MP , Harralson TL. 1998 “The relation of hostility to lipids and lipoproteins in women: Evidence for the role of antagonistic hostility” Annals of Behavioral Medicine 20 : 59-63 .
Suomi SJ. 1987 “Genetic and maternal contributions to individual differences in rhesus monkey biobehavioral development” in Krasnegor NA , Blass EM , Hofer MA , Smotherman WP (eds.). Perinatal Development: A Psychobiological Perspective ( New York : Academic Press ) 397-420 .
Suomi SJ. 1991 “Up-tight and laid-back monkeys: Individual differences in the response to social challenges” in Brauth S , Hall W , Dooling R (eds.). Plasticity of Development( Cambridge MA : MIT Press ) 27-56 .
Suomi SJ. 1997 “Early determinants of behaviour: Evidence from primate studies” British Medical Bulletin 53 : 170-184 .
Suomi SJ, Levine S. 1998 “Psychobiology of intergenerational effects of trauma: Evidence from animal studies” in Danieli Y (ed.). Intergenerational Handbook of Multigenerational Legacies ( New York : Plenum Press ) 623-637 .
Taylor SE. 1999 Health Psychology 4th Ed. ( Boston MA : McGraw-Hill ).
Thaller, V , Vrkljan M , Hotujac L , Thakore J. 1999 “The potential role of hypocortisolism in the pathophysiology of PTSD and psoriasis.” Collegium Antropologicum 23(2) : 611-619 .
Turner JB , Kessler RC , House JS. 1991 “Factors facilitating adjustment to unemployment:Implications for intervention” American Journal of Community Psychology 19 : 521-542 .
Visintainer MA , Seligman ME , Volpicelli JR. 1983 “Helplessness, chronic stress, and tumor development” Psychosomatic Medicine 45 : 17 (Abstract) .
Wadhwa PD , Sandman CA , Porto M , Dunkel-Schetter C , Garite TJ. 1993 “The association between prenatal stress and infant birth weight and gestational age at birth: A prospective investigation” American Journal of Obstetrics and Gynecology 169/4 : 858-865 .
Walker EA , Gelfand A , Katon WJ , Koss MP , Von Korff M , Bernstein D , Russo J. 1999 “Adult health status of women with histories of childhood abuse and neglect” American Journal of Medicine 107/4 : 332-339 .
Wang S , Mason J. 1999 “Elevations of serum T3 levels and their associations with symptoms in World War II veterans with combat-related posttraumatic stress disorder: Replication of findings in Vietnam combat veterans” Psychosomatic Medicine61 : 131-138 .
Weisz JR , Donenberg GR , Han SS , Weiss B. 1995 “Bridging the gap between laboratory and clinic in child and adolescent psychotherapy” Journal of Consulting and Clinical Psychology 63 : 688-701 .
Wing RR , Epstein LH , Nowalk MP , Lamparski DM. 1986 “Behavioral self-regulation in the treatment of patients with diabetes mellitus” Psychological Bulletin 99 : 78-89 .
Worthman C. 1999 “Epidemiology of human development” in Panter-Brick C , Worthman C (eds.). Hormones, Health, and Behavior. ( Cambridge U.K. : Cambridge University Press ) 47-104 .