STRESS AND HEALTH

Stress may be defined as “environmental demands that tax or exceed the adaptive capacity of an organism, resulting in biological and psychological changes that may be detrimental and place the organism at risk for disease” (Cohen et al., 1998) or disability. Stressors can take many forms, including those associated with economic difficulties, physical deprivation, low status, occupational strain, death of a spouse or loved one, family responsibilities or difficulties, neighborhood instability, and discrimination. Stress is often assessed by using rating scales for stressful life events or it is determined by magnitude of exposure to specific stressors (e.g., job stress) or the presence or absence of a chronic stressor (e.g., caregiving responsibilities).

The connection between stress and morbidity and mortality has been demonstrated in a variety of studies. Stressful life events have been shown to predict mortality in initially healthy populations (Rosengren et al., 1993) and in patients with heart disease (Ruberman et al., 1984). Specific types of stressful events are also linked to mortality and illness. In particular, job strain, bereavement, and providing care for a chronically ill relative have been predictive of all-cause mortality and heart disease (Karasek et al., 1988; Martikainen and Valkonen, 1996; Schnall et al., 1994; Schulz and Beach, 1999).

One type of stressor—caring for a spouse with Alzheimer's disease—is especially relevant for older adults. Although caregiving can have some emotional rewards, caregivers on the whole report extraordinarily high levels of burden and stress (Schulz et al., 1995; Vitaliano et al., 1997), which may increase their mortality risk. A recent study compared 392 older people who were providing care for a spouse with a group of 400 people of the same age without this responsibility. Approximately 50 percent of the caregivers reported emotional strain associated with their caregiver roles. Over the 4-year study, those caregivers who reported mental and emotional strain were 63 percent more likely to die than the noncaregivers. Caregivers not reporting strain were not at increased risk of mortality (Schulz and Beach, 1999). There is therefore at least moderately strong evidence that stress is predictive of morbidity and mortality.

The association between stress and illness may be indirect, in that stress has been shown to influence a number of other risk factors. Stress has been associated with an increase in intravenous drug use, smoking, alcohol use, physical inactivity, and unprotected sex (Dougal and Baum, 2001; Kaplan et al., 1993). Therefore, stress may be looked at as both a potential proximal and distal cause of illness and mortality.

RACE, ETHNICITY, AND STRESS

The Commonwealth Minority Health Survey provides a unique glimpse of racial and ethnic variation in stress (Williams, 2000). On a global measure combining exposure to stressors in five domains (occupation, finances, relationships, racial bias, and violence), blacks, Hispanics, and Asians reported higher levels of stress than whites. Among Hispanics, Puerto Ricans had the highest levels of stress. Interestingly, there was dramatic variation among the Asian subgroups included: Chinese reported higher levels of stress than any other group in the study, Vietnamese were intermediate, and Koreans reported the lowest levels in the study.

Reaction to stressors may also differ across groups. Blacks, especially those at the low end of the economic spectrum, report not only a great number of stressful life events but also stronger responses to them, or greater distress, than whites in a variety of domains (Myers and Hwang, 2004). In some studies, though not all, minorities, especially blacks, react with greater psychological distress than whites to unpleasant events (Mirowsky and Ross, 1990; Myers et al., 2002; Ulbrich et al., 1989; Warheit et al., 1973).

Differences in exposure to stress are partly attributable to group differences in socioeconomic status (Neff, 1984; Warheit et al., 1975). However, race may also interact with socioeconomic status in producing levels of distress. Kessler and Neighbors (1986) found such a race-by-class interaction, such that low-status blacks reported higher levels of distress than high-status blacks or whites of any status. In contrast, after controlling for status, older blacks in the recent Macarthur Study of Successful Aging reported lower levels of distress than older whites. In particular, low-status blacks had lower levels of distress than low-status whites (Kubzansky et al., 2000). Thus, although race and socioeconomic status appear to interact in affecting stress responses, the direction of this interaction is far from clear. In addition, exposure to and the experience of stress for some minority groups may be lower than for whites. In one study (Uppaluri et al., 2001), for instance, Asian Indians and Koreans reported lower levels of stress response over a 2-week period than whites, and immigrants who had lived in the United States less than 1 year reported significantly lower stress response levels than those who had lived in the United States for at least 15 years.

Some stressors may decline in importance for older people; others may become more significant. Retrospective reports of discrimination decline with age. Nevertheless, for current generations of minority older adults, stress due to racism and discrimination may be especially important, given that their life histories extend back to earlier periods when civil rights received less attention. Minority older adults may also have greater exposure than younger cohorts to potential stressors related to acculturative stress, spousal and family caregiving, and raising grandchildren (which can of course also be emotionally rewarding; Myers and Hwang, 2004). However, it is unclear, with regard to family responsibilities, whether minority older adults feel as burdened as their white counterparts. Black and Hispanic grandparents care for grandchildren more often than whites do, but the subjective burden they experience may be less (Myers and Hwang, 2004).

In addition to stress, there may be other psychological factors that are predictive of health outcomes and important for racial and ethnic health differences. For instance, depression, anxiety, and anger or hostility have all been prospectively linked with increased cardiovascular or all-cause mortality (Rozanski et al., 1999). Interestingly, each of these psychological factors is believed to be triggered by stressful life experiences. Few if any studies have examined whether these psychological factors contribute to health differences.

NEEDED RESEARCH

Does greater experience of and exposure to chronic stress account for some racial and ethnic health differences among older adults? This question has yet to be systematically investigated; the evidence is incomplete and so far inconclusive. Better measurement of relevant dimensions of stress would be critical to understanding whether and how it affects health differences. While rating scales try to encompass the variety of sources, how adequately they do so is unclear. Such dimensions of the stressful experience as exposure, appraisal, and response (Lobel et al., 1992) require attention.

Research Need 12: Study populations of different racial and ethnic groups to assess the connection between health and the stresses that accumulate over a lifetime.

Despite the considerable research on stress and health, it is still unclear what role stress plays in the health of older adults of different groups. Some minorities, especially blacks and Hispanics, appear to experience higher levels of chronic stress, but it has not been determined whether this is linked to relatively higher morbidity and mortality as they age. Various stressors may be of concern for different groups—discrimination and racism, as discussed above, but also socioeconomic pressure, acculturative stress for immigrant populations, the stress of family responsibilities, and agingrelated problems. Research should focus on sources of stress that apply particularly to minority populations and on the suggestions in the literature that reactions to stressful events may vary across groups.

The assessment of cumulative stresses over the life course should be given special attention and may be especially relevant. People who have experienced more “lifetime adversities” have been shown to exhibit potentially pathogenic physiological activity (Singer and Ryff, 1999). Older minorities, especially blacks and perhaps other groups, have complex life histories that often involve substantial adversity. Investigation of how this affects health in old age, with attention to cohort and historical effects, is a fruitful direction for investigation.

Several psychological factors are associated with stress, including depression, anxiety, and anger or hostility, and may also play a role in health differences. Although a good deal of research has examined the effects of these factors on morbidity and mortality, little of this research has focused exclusively on older adults, and even less has focused on racial and ethnic differences.