The factors we have considered so far—including selectivity, socioeconomic status, health behaviors, prejudice and discrimination, social support, and stress—are typically studied independently, often within the boundaries of separate disciplines. Yet a good deal of research indicates that these factors do not affect health independently but interact among themselves and with biological systems. The study of psychosocial and biological interactions, referred to as biopsychosocial or biobehavioral research, has demonstrated profound effects on biological systems of an array of factors. For instance, psychosocial and behavioral factors have been shown to affect organ systems (e.g., the cardiovascular, neuroendocrine, central nervous, and immune systems) and cellular (e.g., synaptic connections) and molecular (e.g., RNA) factors (Anderson, 1998).
Examination of links to the biological level is critical to understanding health differences in late life because the biological level serves as the pathway to disease and death. A number of studies have found significant racial and ethnic group differences in several biological measures, including the well-known differences in blood pressure, low-density and high-density cholesterol, and obesity, and, more recently, in response to pharmacological agents (Burroughs et al., 2002; James, 1999; Kumanyika, 1987; Whiteley et al., 1999; Zoratti, 1998). Biological factors should not be studied in isolation but as part of a system in which psychosocial, behavioral, and environmental factors lead to potentially pathogenic changes, in ways that may vary by racial or ethnic group.
Researchers in this area have identified candidate biological markers or mechanisms that may be related to psychosocial, behavioral, and environmental factors and may also be associated with risk for disease. These mechanisms include physiological reactivity, allostatic load, psychoneuroimmunology, metabolic syndrome, and neurovisceral integration.
Physiological reactivity is defined as the magnitude of changes in some physiological system in response to acute psychological or behavioral stress. The changes most often studied, typically under laboratory conditions, involve cardiovascular reactivity (e.g., heart rate, blood pressure). The hypothesis is that frequently activated and exaggerated cardiovascular responses to stress raise resting blood pressure levels over time and damage coronary arteries (Matthews et al., 1986; Turner et al., 1992). The strongest evidence for this hypothesis comes from animal experiments that demonstrate an etiological role for stress-induced reactivity in both coronary heart disease (Manuck et al., 1990, 1995) and hypertension (Hallback, 1975; Lovallo and Wilson, 1992; Manuck et al., 1990). Among humans there is evidence that individuals at risk for hypertension—due to a family history or older age, being black or having borderline hypertension—may also exhibit cardiovascular hyperreactivity to stress (Turner et al., 1992). There is also emerging evidence in humans that cardiovascular reactivity among healthy individuals may be prospectively related to later hypertension (Carroll et al., 2001; Falkner et al., 1981; Light et al., 1992; Menkes et al., 1989). Finally, research is emerging on a possible negative correlation between socioeconomic status and reactivity (Gump et al., 1999).
In a new line of research on reactivity, acute psychological stress has been shown to elicit myocardial ischemia, an imbalance between the blood supply to the heart and the demands placed on it to supply the body with oxygenated blood. Myocardial ischemia may result in angina pain, but it is frequently silent, occurring without pain. Psychological stress has been shown to produce silent ischemia in the laboratory; such responses are predictive of ischemia in real life, as measured by ambulatory cardiac recording (Blumenthal et al., 1995). In patients with a history of heart disease, ischemia induced by psychological stress is prospectively related to subsequent cardiac events (Jiang et al., 1996).
A number of studies have explored differences between blacks and whites in stress-induced reactivity as a possible reason for group differences in hypertension. These studies generally find a tendency among black adults and children toward greater peripheral vascular responses to stress in comparison with whites (Anderson, 1989). Blacks show greater cardiovascular responses in laboratory research to stimuli associated with racism (e.g.,
films depicting racist incidents, imagery related to racism) than to nonracist stimuli (Armstead et al., 1989; Guyll et al., 2001). Blacks also show greater blood pressure reactivity than whites to discrimination and unfair treatment (Guyll et al., 2001). However, the contribution of such responses to black-white differences in the prevalence of hypertension has not been determined.
Allostatic load is another concept developed to explain how psychosocial factors might affect biological systems in potentially pathogenic ways (McEwen and Stellar, 1993). It is defined as the overtaxing of several physiological systems (“wear and tear”) in response to stress or other psychosocial or behavioral factors, so that dysregulation and possibly disease may develop. The concept of allostatic load incorporates the notion of physiological reactivity, but it also includes other physiological changes that might be pathogenic. Those changes include having a chronically high level of resting physiological activity (e.g., high levels of stress hormones or high blood pressure); an exaggerated physiological response to psychosocial challenges; the relative inability of the physiological system to recover after acute stress; and, over time, an inadequate physiological response to stress (low reactivity) due to fatigue or dysfunction, which triggers potentially harmful compensatory responses from other systems (McEwen, 1998; Myers and Hwang, 2004). In essence, allostatic load represents a cumulative, multisystems perspective on physiological risk.
Recent research on the Macarthur Successful Aging Cohort indicates that high levels of allostatic load predict morbidity and mortality among older adults. For instance, Seeman et al. (1997) have discovered that higher allostatic load scores are prospectively associated with increased incidence of cardiovascular disease and increased risk for decline in cognitive and physical functioning in older adults. Allostatic load scores have been found to be higher among persons with lower educational attainment and higher levels of hostility than average, suggesting a potential role for allostatic load in the greater disease risk in these populations (Kubzansky et al., 1999). To date, racial and ethnic group differences in allostatic load and the role of such differences in health differences have not been explored.
Psychoneuroimmunology is defined as the examination of the interactions among psychological, behavioral, and social factors with immunological and neuroendocrine outcomes. It is now well established that psychological factors, especially chronic stress, can lead to impairments in immune system
functioning in both the young and older adults (Kiecolt-Glaser and Glaser, 1995). In several studies of older adults, those who are providing care for a relative with dementia report high levels of stress and exhibit significant impairments in immune system functioning when compared with noncaregivers (Kiecolt-Glaser et al., 1987). Stress-induced changes in the immune system may affect a number of outcomes, including slowing the wound healing process and increasing susceptibility to infections (Cohen and Williamson, 1991; Cohen et al., 1991; Kiecolt-Glaser et al., 1995).
To date, there have been few if any published studies on racial and ethnic group differences in stress-induced immunosuppression and into the role it may play in health differences.
THE METABOLIC SYNDROME AND PSYCHOSOCIAL FACTORS
The metabolic syndrome, also known as syndrome X or the insulin resistance syndrome, is characterized by a combination of central adiposity (abdominal fat), insulin resistance, glucose intolerance, and hypertension (Reaven, 1988). The metabolic syndrome has been prospectively linked to type 2 diabetes and cardiovascular disease prevalence and mortality (Baillie et al., 1998; Laaksonen et al., 2002; Lakka et al., 2002; Vague and Raccah, 1992).
Research has uncovered some racial and ethnic differences in metabolic syndrome and its components. In particular, blacks exhibit higher levels of central obesity, insulin resistance, hypertension, and glucose intolerance than whites (Adams-Campbell et al., 1990; Kasim-Karakas, 2000; Lovejoy et al., 1996). Research on the relationship of psychosocial factors and metabolic syndrome is emerging, and there is some evidence of a significant link. For example, Brunner and associates have found an inverse relationship between socioeconomic status and metabolic syndrome, a connection that might be mediated by heightened sympathetic nervous system activity (Brunner et al., 1997, 2002; see also Hjemdahl, 2002). In a prospective analysis of older men, Vitaliano et al. (2002) found that the chronic stress of caregiving for a person with Alzheimer’s was a predictor of metabolic syndrome, which in turn was discovered to mediate the connection between chronic stress and coronary heart disease. Research has not determined whether racial or ethnic differences in metabolic syndrome are mediated by stress or other psychosocial factors or the role of metabolic syndrome in different disease outcomes by group.
Thayer and Friedman (2004) have proposed a model of neurovisceral integration as a new framework that could possibly link research on psycho-
social factors to health outcomes. The model complements and extends previous biopsychosocial models, positing a common element among them: an imbalance between the branches of the autonomic nervous system (ANS), the parasympathetic and sympathetic, that favors the latter. The status of the ANS at any time is a function of the inhibitory influences of the parasympathetic system and the excitatory effects of the sympathetic system. The ANS is important since it controls cardiovascular, neuroendocrine, and immune system functions; it also interacts with key brain regions related to psychological and emotional processes. Thayer and Friedman (2004) hypothesize that the lessening of parasympathetic influence in the ANS (as indexed by heart rate variability), especially under psychosocial challenges such as stress, worry, perseverative thinking, depression, anxiety, racism, and low status, may be a key biological pathway that underlies racial and ethnic differences in health in late life.
Although this model has not been evaluated with respect to health differences, there is evidence that lowered heart rate variability and a higher resting heart rate (as an index of decreased parasympathetic tone) are related to increased all-cause mortality (Thayer and Friedman, 2004).
These biobehavioral models of disease should not be viewed as competing conceptualizations, since there is considerable overlap among them. Cardiovascular reactivity, allostatic load, immune system changes, metabolic syndrome, and heart rate variability are strongly correlated. At this time, it is not clear which of these, or any other biobehavioral model, is most predictive of illness and death. Nevertheless, these models are likely to be related to many of the proposed determinants of health differences, such as stress, social support, racism, health risk behaviors, and socioeconomic status, and may therefore indicate mechanisms through which such risk factors translate into disease susceptibility and health differences.
Much basic research is still needed on biopsychosocial mechanisms, and further work is needed to elucidate the links, if any, with race and ethnicity.
Research Need 13: Clarify how biopsychosocial factors affect health outcomes over time in racial and ethnic groups of middle-aged and older adults.
It has been hypothesized that psychosocial effects on biopsychosocial outcomes, such as cardiovascular reactivity, allostatic load, psychoneuroimmunology, metabolic syndrome, and neurovisceral integration are predictors of morbidity and mortality. It remains an empirical question whether
they indeed affect health among older adults and whether they have different implications for health across racial and ethnic groups.
Basic research is needed on all these potential disease pathways for the effects of acute and chronic stress, as well as on how these mechanisms mediate between such stressors as racism and low status and health outcomes for various racial and ethnic groups. The discovery of differences in such mechanisms, in their linkages to social factors or in their effects on health, would provide evidence of their involvement in differences. Besides longitudinal studies, laboratory experiments and ambulatory studies of physiological activity in the natural environment would be informative. For instance, exploring the biological concomitants of group differences in neighborhood environments and occupational experiences might provide a better understanding of their differential effects across groups.