This Article Abstract Full Text (PDF) Services Download to citation manager PubMed PubMed Citation

Ethnic Variation in the Impact of Negative Affect and Emotion Inhibition on the Health of Older Adults

^a Center for Studies of Ethnicity and Human Development, Long Island University, Brooklyn, New York
^b Department of Geriatric Psychiatry, State University of New York Health Science Center, Brooklyn, New York
^c Department of Sociology, University of Alberta, Edmonton, Canada

Carol Magai, Long Island University, 1 University Plaza, Brooklyn, NY 11201 E-mail: cmagai{at}liu.edu.

Decision Editor: Margie E. Lachman, PhD

	Abstract

TOP Abstract Methods Results Discussion References

 
The relations between patterns of emotional experience, emotion inhibition, and physical health have been little studied in older adults or ethnically diverse samples. Testing hypotheses derived from work on younger adults, the authors examined the relations between negative affect and emotion inhibition and that of illness (hypertension, respiratory disease, arthritis, and sleep disorder) in a sample (N = 1,118) of community-dwelling older adults from four ethnic groups: U.S.-born African Americans, African Caribbeans, U.S.-born European Americans, and Eastern European immigrants. Participants completed measures of stress, lifestyle risk factors, health, social support, trait negative emotion, and emotion inhibition. As expected, the interaction of ethnicity with emotion inhibition, and, to a lesser extent, negative affect, was significantly related to illness, even when other known risk factors were controlled for. However, the relations among these variables were complex, and the patterns did not hold for all types of illness or operate in the same direction across ethnic groups. Implications for emotion–health relationships in ethnically diverse samples are discussed.

It has long been suggested that differential patterns of emotion and emotion expression have differential consequences with respect to health. Many researchers have maintained that the open expression of emotion is vital to optimal physical as well as mental health (Gross and Levenson 1997; Leventhal and Patrick-Miller 2000; Pennebaker and Seagal 1999; Polivy 1998). In what is probably the most well-known and best developed of these theories, Pennebaker and colleagues (e.g., Pennebaker 1995; Pennebaker and Seagal 1999; Pennebaker and Susman 1988) have argued that the inhibition of emotion has negative health consequences, particularly when it occurs over time. In contrast, the active expression of emotion is said to have salutary consequences. Pennebaker has been able to demonstrate a wide range of positive health benefits deriving from the written expression of emotional events, including greater positive affect, fewer doctor's visits, and improved immune functioning (see Pennebaker and Seagal 1999, and Smyth 1998, for recent reviews).

Although Pennebaker (Spera, Buhrfeind, and Pennebaker 1994) acknowledged that the causal mechanisms in expression–health relations are unclear, the inhibition of expression would appear to be an active process that requires a degree of psychological or physiological "work" (see also Leventhal and Patrick-Miller 2000; Polivy 1998). According to Pennebaker, negative emotion that is not expressed is sustained and is experienced as stress, resulting in negative health outcomes when inhibition occurs over longer periods of time. Prolonged inhibition of expression places a cumulative burden on the body and results in an increased vulnerability to stress-related diseases.

There is some further support for Pennebaker and Susman 1988 ideas in recent experimental work on the effects of inhibiting emotional expressions and emotional experience. In a series of studies, Gross and colleagues have demonstrated how inhibiting the expressions of disgust (Gross 1998; Gross and Levenson 1993), amusement, and sadness (Gross and Levenson 1997) is associated with a range of changes in autonomic indices. They have found, for example, that suppressing the visible expression of disgust causes an increase in sympathetic arousal, as indexed by changes in skin conductance, finger pulse amplitude, and heart rate (Gross and Levenson 1993).

Although Pennebaker's early views (Pennebaker and Beall 1986; Pennebaker and Susman 1988) are widely accepted, a small group of researchers have argued that it is the chronic experience and expression of negative affect that may have negative health consequences (e.g., Bonanno, Keltner, Holen, and Horowitz 1995; Bonanno, Znoj, Siddique, and Horowitz 1999; Kennedy-Moore and Watson 1999; Tice and Baumeister 1993). Tice and Baumeister, for example, suggested that the expression of anger intensifies rather than alleviates experience, and it has been demonstrated that the expression of anger is predictive of increased cardiovascular activity and increased risk for coronary heart disease (see T. W. Smith 1992, for a review). A series of studies of bereavement by Bonanno and colleagues (Bonanno and Keltner 1997; Bonanno et al. 1995, Bonanno et al. 1999; Keltner and Bonanno 1997) have demonstrated that verbal-autonomic dissociation (a discrepancy between expressed affect and physiological responsiveness) was not associated with negative health consequences, as indexed by the number of doctor's visits. Fewer facial expressions of grief predicted less grief and better health over time (Bonanno and Keltner 1997), and overt verbal disclosure during bereavement predicted increased distress, grief, and poorer health.

Although this theory and research underscores the likely complexity of relations between inhibition and health, the preponderance of the empirical data is nonetheless consistent with Pennebaker and Susman 1988 basic theory. Data contradictory to the notion that inhibition is negatively associated with health can only be gleaned for the specific emotions of anger (e.g., T. W. Smith 1992) and sadness/grief (Bonanno et al. 1995, Bonanno et al. 1999), and further research and well-informed theory are clearly needed on this issue. It may be, for example, that inhibition is particularly unhealthy where it is chronic, inflexible, and insensitive to environmental demand (Gross and Levenson 1997) or only where expression is central to the function of the emotion in question (Consedine, Magai, and Bonanno 2002).

In this vein, it is worth noting that certain personality types are known to differ systematically in the frequency of specific emotional experiences and intensity (Larsen and Diener 1987) as well as in both regulatory and expressive styles; moreover, there is generally robust continuity in styles of expression over time over the adult years (see review by Magai 2001). In terms of continuity from childhood to adulthood, the evidence is somewhat more indirect, as there are no longitudinal studies that have directly measured expressive behavior. However, there exist a number of longitudinal studies of attachment styles from infancy to early adulthood, these studies being germane because attachment styles are intimately related to expressive patterns, with avoidant children and adults showing restricted expressivity and ambivalent children and adults showing heightened or extreme expressivity (Cassidy 1994, Cassidy 2000; Fraley and Shaver 2000; Fuendeling 1998). In general, longitudinal studies of attachment have shown continuity from infancy or early childhood through adolescence and into early adulthood (Allen and Land 1999), although continuities are linked to several important contextual factors (Thompson 1999). There is also evidence of continuity of attachment styles over the adult years, as evidenced in a study that tracked a cohort of women from age 21 to age 52 (Klohnen and Bera 1998). Taken together, this body of research is suggestive of considerable continuity in emotion styles and emotion regulation over the life span.

As is implicit in a discussion of attachment, the ability to regulate emotional arousal is acquired relatively early in development, with pronounced individual differences emerging rather rapidly (Cicchetti 1996; Fox 1994). Most children learn to regulate their emotions in accord with the social and cultural conventions that are optimal for their given environment. Some, however, inhibit their emotions unduly or, conversely, show an exaggerated or elevated level of emotionality (Cassidy 1994). Although there is a marked dearth of research documenting the consequences of discrepancies between the environments in which regulatory styles were developed and those in which they are currently relevant, the literature has certainly documented the difficulties faced by immigrant groups (e.g., E. Myers 1993). However, given the early development and longitudinal stability of regulatory styles noted above, it seems reasonable to suppose that at least part of these difficulties relate to discrepancies between what may have been culturally appropriate regulatory strategies and their utility in new social and cultural environments.

More generally, a growing literature continues to demonstrate the robust relations between the experience of negative affect and health (see, e.g., Mayne 1999, for a recent review). Although many studies have focused on expression as the way to index levels of negative affect, data have nonetheless shown that the experience of negative emotion, particularly where it is traitlike or chronic, is associated with both the provocation and the increased incidence of health problems (Cooper and Faragher 1993; Eysenck 1988). In their meta-analysis of the "disease-prone" personality, for example, H. S. Friedman and Booth-Kewley 1987 have shown that trait anger/hostility, depression, and anxiety levels are systematically related to disorders ranging from asthma to arthritis. Strengthening a causal interpretation between traitlike emotional characteristics and health, these authors also found that the effect size from prospective studies was only slightly smaller than that for concurrent studies. Despite such findings (see also Grossarth-Maticek, Bastiaans, and Kanazir 1985; Grossarth-Maticek, Kanazir, Schmidt, and Vetter 1985), it remains difficult to evaluate the relation between personality variables and health because most of the data are correlational (see Pennebaker 1993) and disease processes are patently multifactorial (Eysenck 1994). In addition to the need for large samples, personality variables like trait negative emotion must be examined in concert with both social and health risk factors to help clarify the relations (H. S. Friedman and Booth-Kewley 1987).

The relations among emotion, physiological activation, and health outcomes become increasingly important when considering the role of emotion and emotion regulation in older populations. H. Leventhal and colleagues 1998, for example, have recently argued that whether the physiologically arousing effects of emotions are functional depends on the ability of the somatic system to deal with the arousal. These authors have suggested that aging processes are associated with a reduction in both the overall amount of energy available and the system's ability to effectively regulate its use. Older bodily systems appear to regulate energy use less effectively, and arousal may well last longer (see also Panksepp and Miller 1995). This is an important point, for rather than assist the organism in its interactions with the environment, prolonged physiological activation may well be maladaptive and lead to negative health consequences (Gross et al. 1997; H. Leventhal and Patrick-Miller 2000). Although the literature is beginning to systematically document improvements, or at least normative changes, in the strategies by which older adults regulate emotion experience and expression (e.g., Carstensen, Gottman, and Levenson 1995; Gross et al. 1997; Labouvie-Vief and Diehl 2000), normal aging is nonetheless associated with both physical and physiological decline. The consequence of these changes is that emotional arousal may last longer for older adults, with more deleterious effects accruing. If emotion is closely linked to physiology in the manner suggested by our review of the literature, and adults manifest traitlike patterns with respect to emotion (Gest 1997; Magai 2001), then the manner in which older adults regulate and express their emotions may be of the utmost importance to their physical health and well-being. Furthermore, the study of older adults provides an excellent opportunity to examine how the effects of emotion and regulatory styles on health accumulate across the life span.

Despite the recent emergence of a more complex view of the relation between emotion and health, the foregoing literature has given scant attention to cultural variation. This omission stands despite the likelihood that culture plays a key role in determining how emotion regulation and health relate. Although there is a distinct literature on the anthropology and sociology of emotion (e.g., Hochschild 1979; Lutz 1987; Lutz and White 1986), there is little in the way of a systematic corpus of data on how different ethnic or cultural groups vary in emotional experiences, the tendency to become emotionally aroused, and the tendency to express or inhibit when emotionally distressed or, perhaps most important, on how such differences might relate to health outcomes. Indeed, although the concept of emotion regulation is embedded within a framework of learned "cultural display rules" (Saarni 1989, Saarni 1992), there has been remarkably little research on how particular cultures influence expressive styles and what impact this may have on the ability to withstand stress and moderate emotion–health relations. However, most of the factors involved in emotion processes are known to show some degree of cultural variation (Mesquita and Frijda 1992; Mesquita, Frijda, and Scherer 1997), and cultural specificity in emotion socialization (e.g., Brice 1992; Deater-Deckard, Dodge, Bates, and Pettit 1996; Gopaul-McNicol 1993; Payne 1989; Pinderhughes, Dodge, Bates, Pettit, and Zelli 2000; Young 1974, Young 1979) should be made manifest in differential patterns of emotional expression and experience, with individuals from different cultures learning to use emotion expressions in different ways, with different intentions, and with different outcomes.

The health literature has long documented the health disparities and differential mortality rates that exist between two broad ethnic groups, African Americans and European Americans, but there has been little examination of within-group differences among African Americans, or among European Americans, for that matter. Indeed, researchers have tended to examine individuals within very broad ethnic categories such as Asian, Caucasian, Hispanic, and African American (e.g., Bach, Cramer, Warren, and Begg 1999; H. F. Myers, Kagawa-Singer, Kumanyika, Lex, and Markides 1995; Sanders-Phillips 1996). In addition, there has been little examination of psychosocial determinants of health and illness across ethnic groups, over and above income and educational disparities or health behaviors. A recent review of the literature (Lillie-Blanton and Laveist 1996) has indicated that although economic conditions exert a powerful influence on health status, they do not account for all the difference. Apropos, the current study was concerned with documenting the relation between health and emotion in older adults, testing the thesis that negative affect and emotion inhibition are two factors that have negative health consequences over and above factors that have traditionally been considered sources of adverse health consequences, and examining ethnic differences therein. In our analytic model we take into account factors that have already been established as health risks, including demographic variables such as income and education (Lillie-Blanton and Laveist 1996), stressful life events (Cobb and Steptoe 1996, Cobb and Steptoe 1998), low social support (Bonanno et al. 1995; Holahan, Moos, Holahan, and Cronkite 1999), and lifestyle risk factors such as smoking (Johnson, Anderson, Bastida, Kramer, Williams, and Wong 1995), alcohol consumption, and obesity (H. F. Myers, et al. 1995; Najjer, Rowland, and Rowland 1988).

In the present study, we focus on four major stress-responsive health syndromes: hypertension, respiratory disease, arthritis, and sleep disorder. Stress has consistently been related to hypertension in most racial groups (Barnes, Schneider, Alexander, and Staggers 1997), and stress management is a key element of successful psychoeducational interventions for the disease (Dusseldorp, van Elderen, Maes, Meulman, and Kraaij 1999). Stress is a known trigger for asthma (Sarafino, Paterson, and Murphy 1998); the attacks themselves are correlated with more general respiratory difficulties such as the frequency of infections (Sarafino and Dillon 1998), and respiratory infections are among the disorders listed in Hicks and Hyler 1998 Stress-Related Problems Scale. Stress has consistently been related to both rheumatoid arthritis (Koehler 1985; see Huyser and Parker 1998, for a recent review), and osteoarthritis (e.g., Creamer and Hochberg 1998; Weinberger, Hiner, and Tierney 1987), although there has been some suggestion that the two arthritis subtypes may differ somewhat in the variables relevant to their prediction (Zautra, Burleson, Matt, Roth, and Burrows 1994). Finally, stress has been related to the incidence of sleep disorders (Partinen 1994; Sadeh 1996), although the data may be somewhat more complex in older samples (L. Friedman, Brooks, Bliwise, Yesavage, and Wicks 1995).

On the basis of the literature reviewed above, we made the following predictions:

HYPOTHESIS 1: Background variables, stress, social support, and lifestyle risk factors (alcohol consumption, smoking, obesity) would be significant predictors of health status in this population of older adults.

HYPOTHESIS 2: Negative affect would contribute significant variance, even where demographic variables, stress, social support, and health risk factors are controlled, with higher levels of trait negative emotion being predictive of greater illness.

HYPOTHESIS 3: Emotion inhibition would contribute significant variance, even where sociodemographic variables and negative emotion are controlled, with greater levels of inhibition being predictive of greater illness.

A further purpose of the current research was to examine ethnic differences in the variables predicting health. Although the absence of an empirical literature considering how the predictors of illness may vary across ethnic groups prevents the generation of precise hypotheses, we believe the literature describing ethnic differences in emotion and emotion regulation described above provides general grounds for the expectation of ethnic differences in the variables predicting health. The discovery of such differences has the potential to be of considerable benefit, providing some preliminary guidelines for the development of more ethnically differentiated research models.

	Methods

TOP Abstract Methods Results Discussion References

 
Participants
The data for this investigation of emotion and illness were taken from a larger population-based study on stress and coping in older Americans. Participants were 1,118 community-dwelling older European Americans and African Americans living in a large urban city and recruited on the basis of a stratified cluster sampling plan. At the initial stage, data on census blocks were gathered from the Household Income and Race Summary Tape File 3A of the 1990 census files. Blocks were then stratified by ethnic group (African American, European American) and on the basis of income (high, medium, or low). Random selection without replacement was used to choose samples of block groups from each stratum. Trained interviewers were sent to conduct interviews with respondents who lived within the selected blocks and who received $20 for their participation. For further details of the sampling design, see Magai and colleagues 2001. A further distinction was made within the African American and European American groups. In the African American sample, we differentiated between U.S.-born African Americans and those from the English-speaking Caribbean, restricting the sample to the largest local immigrant groups deriving from the islands of Jamaica, Trinidad, and Barbados. Because immigration status, rather than differences in culture of origin, might account for intragroup differences, the European American sample was broken down into U.S.-born European Americans and immigrant European Americans. The majority of the immigrant Europeans were originally from Eastern Europe, predominantly from Ukraine (n = 57) and Russia (n = 41), with smaller groups from Poland and other Eastern-bloc groups of the former Soviet Union. We originally considered testing the Eastern Slavic groups (Ukrainians, Russians) separately, but they share a sufficiently similar heritage (Wierzbicka 1998, Wierzbicka 1999) and did not differ on the variables of interest. The remaining groups were too small to allow feasible comparisons. Here and throughout we refer to the larger composite group as Eastern Europeans. Table 1 presents the background characteristics for the sample as a whole as well as by ethnicity.

Procedures
Data were collected during face-to-face interviews by interviewers of the same race as that of the respondents. The sessions lasted approximately 1.5 hr and were conducted in the respondent's home or another location of their choice such as a senior center or a church. The measures were administered in a standard order for all respondents.

Measures
Demographic and Health Risk Questionnaire
This first questionnaire elicited information on age, sex, race, ethnicity, educational attainment, household income, smoking, alcohol consumption, height, and weight. The latter two variables yielded a body mass index (BMI), which is a measure of obesity, computed by dividing weight in kilograms by height in meters squared (Deurenberg, Yap, and van Staveren 1998; Dressler 1996). This variable had 14 outliers, which were recoded back into the distribution slightly more than 3 standard deviations from the mean. Alcohol consumption was measured by the Frequency-Quantity Index (Knupper, Fink, Clark, and Goffman 1963). Information on smoking history was elicited by asking respondents to indicate the number of cigarettes or packs of cigarettes they smoked a day and the number of years they had been smoking. Because the distributions for alcohol consumption and smoking were highly skewed, with about half the sample reporting no use of alcohol or cigarettes, we converted these scores into dichotomous variables (present/absent). The income variable was positively skewed; its distribution was improved with a log transformation.

Physical health status
Physical health was measured using the physical health scales of the Comprehensive Assessment and Referral Evaluation (CARE) instrument (Golden, Teresi, and Gurland 1984; Teresi, Golden, and Gurland 1984). The CARE was originally developed to assess health and social status of community-dwelling older adults in a cross-national study. The full scale assesses the presence and severity of specific psychiatric, physical, and social/environmental problems, as well as their impact on day-to-day functioning. The CARE is administered in a semistructured interview format; has been used extensively with geriatric populations, including indigent and minority populations; and has demonstrated good construct validity (Teresi, Golden, Gurland, Wilder, and Bennett 1984), as well as good concurrent and predictive validity (Teresi, Golden, and Gurland 1984). For example, Teresi, Golden, and Gurland 1984 found that sleep disorder, somatic conditions, and heart disease were significantly associated with subsequent mortality at 1 year after the initial assessment. Here we report data on the four stress-responsive syndromes that were the focus of the present investigation: hypertension (four items: Do you have a blood pressure problem? Does your doctor say you have a blood pressure problem? Do you have blood pressure/heart problem symptoms? Are you taking a drug prescribed for hypertension?), respiratory disease (six items covering symptoms of breathlessness, coughing, and voice hoarseness), arthritis (nine items, including symptoms of backaches; neck aches; swelling in the joints, muscles, tendons, and/or feet; pain in legs, joints, muscles, and/or arms/hands; having arthritis or rheumatism), and sleep disorder (eight items including difficulty falling asleep, intermittent sleep, feeling tired despite sleep, taking medication to sleep). The alpha coefficients of these scales were .89, .64, .86, and .72, respectively.

Stress
Stress was measured by the stress scale used in the National Survey of Black Americans (Chatters 1993), which had respondents indicate the degree of stress experienced in 10 target areas: health, money, job, problems with family or marriage, problems with people outside the family, children, crime, police, love life, and racial conflict. Summing across the different domains yields a total stress score. The Cronbach's alpha for the scale was .71. The distribution of the variable, which was positively skewed (.76), was improved by a standard square-root transformation (Tabachinick and Fidell 2001).

Social support
Because social support is known to be related to health (e.g., S. Cohen and Herbert 1996), the strength and quality of social networks was measured by the Network Analysis Profile (C. I. Cohen and Sokolovsky 1979; Sokolovsky and Cohen 1981). This is a semistructured interview that assesses the quality of the respondent's social network. Respondents indicate the total number of persons in their family (broadly defined as including the extended family) with whom they have had at least a 15-min conversation within the past 3 months or with whom they have engaged in other activities or material exchanges. The scale contains an affective/affiliative dimension, which includes ratings on whether the respondent can share intimate thoughts, can count on them, and can understand them (each scored 1 for yes, 0 for no). A total social support score derives from summing the positive items across the individuals in the social network. In the present sample, the distribution was positively skewed (.36) and was improved by a square-root transformation (Tabachinick and Fidell 2001).

Emotional experience
Disposition for emotional arousal was measured by a trait version of the Differential Emotions Scale (DES), Version III (Izard 1972). This 30-item scale was developed for use with individuals who may have limited education. There are three items for each of 10 fundamental emotions: joy, surprise, interest, fear, sadness, anger, contempt, disgust, shame, and guilt. Respondents rate, on a scale of 1–5, how much each emotion characterizes their day-to-day experience. The scale has been used in numerous investigations on emotion and enjoys strong psychometric properties (Izard 1972). The alpha coefficients for all scales are .84 or greater. The average 1-week test–retest reliability for the scales is .77. In the present sample, we used the subscales linked to clinically significant emotions—anger, sadness, fear, and shame. Their alpha coefficients were .80, .78, .85, and .74, respectively. These were combined to constitute an aggregate measure of negative emotion.

Emotion inhibition
As we were interested in emotion regulation patterns that are established in childhood and played out over the course of the life span, as well as the possibility of examining the accrual of damage across a life span, the tendency to express or inhibit emotion was assessed with the Emotions as a Child Questionnaire (EAC), a 48-item inventory that asks respondents to indicate how likely it was that items reflected their own style of responding when they were afraid, angry, sad, and ashamed as a child, on a scale ranging from 1 (not at all like me) to 5 (very much like me). Sample items include "go to my mother," "try to solve the problem on my own," "read," "withdraw," "keep the problem to myself," and so forth. Items for the scale were developed from responses of research participants in another study to questions in an adapted version of the adult attachment interview (Magai et al. 2001). Scores for the inhibition items of the Anger, Sadness, Fear, and Shame subscales were combined to form an aggregate inhibition scale. The alpha in the present sample was .71.

The 2-week test–retest reliability for the aggregate inhibition score was .72 in an independent sample (n = 60) of adults, and the alpha was .77. In another independent sample (n = 288, mean age = 27 years, SD = 10, 75% female), scores on the EAC emotion inhibition variable were compared with scores on an adult analogue of the EAC—the Present Personality Questionnaire (PPQ), a 24-item scale that measures the tendency to inhibit emotion in the respondent's present life. Sample items include "I have difficulty expressing my anger," "I try not to let my anxieties show," and "I call on my friends when I feel sad." The alpha for the PPQ inhibition scale was .82, and it correlated with the EAC inhibition subscale .58 (p < .0001).

Data Analysis
Analysis of variance and chi-square tests were used to examine ethnic differences with respect to health risk variables, including negative affect and emotion inhibition. All post hoc tests were Games-Howell tests. In testing the specific contribution of negative affect and emotion inhibition to health, we conducted regression analysis. Background demographic (age, sex, education, income, marital status) and health risk variables (stress, social networks, smoking and drinking history, and BMI) were entered as control variables, along with the two key variables of interest, negative emotion and emotion inhibition. Ethnicity was also treated in the model by creating ethnic dummy variables. Following conventional practice, we treated the U.S.-born European Americans as the reference group. For this purpose, European Americans were coded 0 and each of the other three other groups—U.S.-born African Americans, African Caribbeans, and Eastern Europeans—was coded 1. However, because this regression did not permit a comparison between U.S.-born Blacks and Caribbean Blacks, and because the beta coefficients for these two groups—approximately 2 standard errors—indicated that U.S. Blacks scored significantly higher than immigrant Blacks on the illness measures, we conducted an additional regression specifically comparing U.S.-born and Caribbean-born Blacks. The data were screened for departures from linearity, normality, homoscedasticity, and multivariate outliers.

	Results

TOP Abstract Methods Results Discussion References

 
Table 2 presents the mean scores on the health risk variables (traditional health risk variables as well as negative affect and emotion inhibition). As indicated, there were significant ethnic differences on all the variables.

Table 3 presents the intercorrelations among the predictor variables for the regression models. Table 4 contains the raw and standardized coefficients from the regressions of the four dependent variables on ethnicity, negative affect, emotion inhibition, their product terms, and the control variables. As indicated, only stress had a consistent (positive) effect across all four sets of symptoms. Age was related to increased hypertension, female gender was associated with increased levels of arthritis and hypertension, and lower income was associated with arthritis and sleep disturbance. Neither marital status nor social networks was significantly related to symptomatology.

Negative emotion made a significant contribution to sleeping disturbance, and the interaction between negative emotion and ethnicity indicated that the effect was greater among Eastern European immigrants in the prediction of arthritis. The interaction effects between emotion inhibition and ethnicity were significant for three illness conditions, indicating that the effect of emotion inhibition on arthritis, hypertension, and respiratory distress is reduced among Eastern Europeans. Finally, the interaction between ethnicity and emotion inhibition in the case of sleep disturbance indicates that the impact of emotion inhibition on sleep disturbance is reduced, relative to its impact in the U.S.-born European sample, among both U.S.-born African Americans and African Caribbeans.

In summary, when the three ethnic groups are compared with U.S.-born European Americans, there are clear differences in symptomatology. The effects of negative emotion and emotion inhibition on symptomatology are also closely interwoven with those of ethnicity, the most obvious pattern being that negative emotion is associated with increased symptomatology among Eastern European immigrants and emotion inhibition with reduced symptomatology.

The general pattern of ethnic effects in this model, coupled with our exploratory interest in within-group differences, led us to conduct a second regression specifically comparing U.S.-born African Americans with African Caribbeans (see Table 5 ). This analysis indicated that BMI was associated with arthritis and hypertension and that smoking was associated with respiratory disorder and sleep disturbance. Caribbeans reported significantly less arthritis, respiratory disorder, and sleep disturbance than U.S.-born African Americans and marginally less hypertension (p = .06). Negative emotion was associated with greater respiratory disorder and sleep disturbance; an Ethnicity x Negative Emotion interaction indicated that the effect of negative emotion on sleep disturbance was reduced among Caribbeans. Inhibited emotion was associated with hypertension and sleep disturbance. The impact of emotion inhibition on sleep disturbance was reduced among Caribbeans.

	Discussion

TOP Abstract Methods Results Discussion References

 
In this study, we evaluated the role of negative affect and emotion inhibition in the stress-responsive illnesses of hypertension, respiratory disease, arthritis, and sleeping disorder. We predicted that negative affect would account for unique variance even when background variables and a range of other known risk factors were controlled. In line with this expectation, trait negative emotion was positively associated with reports of sleep symptoms in the overall model. In the model comparing the two groups of African descent, negative emotion again predicted sleep symptoms as well as respiratory disorder. Consistent with our general expectation, there were interactions between negative emotion and ethnicity. Specifically, the data indicated that the effect of negative affect on arthritis was enhanced among Eastern Europeans (when compared with European Americans) and the effect of negative emotion on sleep disorder was reduced among Caribbeans (when compared with African Americans). It is important to point out that although these relations are modest by experimental standards, the magnitude of the contribution of negative affect to health is on par with that of other risk variables (e.g., smoking and BMI), both as measured in this sample and as reported in other studies (e.g., Brandon, Copeland, and Saper 1995; H. S. Friedman and Booth-Kewley 1987). Thus, modest effects can be of great significance clinically.

Overall, however, trait negative emotion was less consistently related to health outcomes than previous theory led us to expect. Although the effect sizes in the relations that were obtained were on par with the effects of traditional risk variables, trait negative emotion was not a powerful predictor across disease indices. Here we speculate that this pattern of results relates, in part, to characteristics of our sample, the relations between negative emotion and stress, and the particularly stringent hypothesis tested here. Older cohorts are noteworthy in that they often report less negative emotion (Magai 2001; Mroczek and Kolarz 1998), perhaps as a result of early socialization experiences resulting in the tendency to route negative affect from consciousness (Magai et al. 2001). It may be that negative affect is underreported or even that our older adults were more comfortable reporting "stress" than they are negative emotions. In addition, the correlation among health risk factors, particularly that between stress and negative emotion, may lead to situations in which the importance of personality variables in disease processes, such as negative emotion or emotion inhibition, is underestimated (see T. W. Smith 1992). Indeed, stress and negative emotion were correlated at .35 in our sample (see Table 3 ); the fact that cultural and personality factors, including negative affect, were entered in the regression models at the same time creates the possibility that the stress measure was absorbing some of the variance we had expected to be associated with negative emotion.

Although we feel these data favor a direct causal role for negative affect in illness processes, it could nonetheless be argued that negative emotion is a consequence, rather than a cause, of illness inasmuch as the results of the present study are correlational in nature. As H. Leventhal and Patrick-Miller 2000 noted, the impact of disease and illness on emotion is conspicuous, and the possibility that illness is causing negative emotion cannot be denied. Alternatively, it is also possible that negative emotion acts as an indicator of systemic vulnerability rather than exerting a direct influence on health in and of itself (S. Cohen et al. 1995; H. Leventhal et al. 1998; Mayne 1999). Lack of quality sleep, difficulty breathing, or pain from arthritis might all cause either stress or negative emotions.

Such a consideration notwithstanding, a causal interpretation linking the characteristic experience of negative emotion to respiratory disorder and sleep disturbance remains viable for a number of reasons. First, if negative emotion is merely an indicator of somatic resources (cf. H. Leventhal et al. 1998), we would expect a general increase in negative affectivity with age as bodily system resources decline. However, there is considerable evidence that the frequency and intensity of positive and negative affective experiences do not undergo substantial change over the life course (Carstensen et al. 1995; Magai 2001). One of the largest population-based studies of later life found that negative emotion appears to remain the same with age even into advanced old age, despite the increase in chronic illness (Baltes and Mayer 1999). Some researchers, in fact, have suggested that negative affect may decrease until the very end of life (Carstensen et al. 1995; Gross et al. 1997), although positive affect may also decrease among the oldest old (e.g., Mroczek and Kolarz 1998; J. Smith, Fleeson, Geiselmann, Settersten, and Kunzmann 1999). Furthermore, a number of prospective longitudinal studies have demonstrated that levels of negative affect measured at an earlier time are predictive of health-related variables at a later time (Barefoot, Dahlstrom, and Williams 1983; see T. W. Smith 1992, for a recent review). Although many writers have assumed that premorbid ratings of negative affect affect later health indirectly through health-related behaviors (e.g., Fisher and Feldman 1998; H. Leventhal et al. 1998), longitudinal research has indicated that ratings of negative emotion are more than just a consequence of illness or disease, and our data show that trait negative emotion is associated with illness in addition to and independent of its relation to health risk behaviors.

However, perhaps the most important consideration in the present context is the fact that the measure of negative affect used here—the trait version of the DES—is a characterological measure of emotion designed to index stable aspects of personality (Izard 1972). Trait emotionality shows good stability over periods of up to 7 years (Magai 2001; Watson and McKee 1996), and the DES measure in particular has likewise demonstrated reliability over time (Izard, Libero, Putnam, and Haynes 1993). Although trait and state emotion measures are correlated to a certain extent (e.g., E. A. Leventhal, Hansell, Diefenbach, Leventhal, and Glass 1996), recent research has stressed the importance of trait over state measures in the prediction of health (e.g., Bonanno and Keltner 1997). Raikkoenen and colleagues (Raikkoenen, Matthews, Flory, and Owens 1999; Raikkoenen, Matthews, Flory, Owens, and Gump 1999), for example, have shown that levels of trait anxiety and trait hostility are directly related to physiological arousal and health status. More important, their results have indicated that for individuals who are high in trait anxiety or hostility, state mood fluctuations bear no relation to autonomic activity. These findings, coupled with the nature of the emotion measure used here, suggest that an emotion-induced symptom-reporting bias is unlikely to be entirely responsible for the relation between trait emotion and health.

We also predicted that the extent to which emotions were inhibited would account for unique variance even after all other factors and negative affect had been taken into account. In theory, emotion inhibition requires psychophysiological work and should thus be associated with negative health consequences (Greenberg and Stone 1992; Pennebaker and Seagal 1999). We reasoned that these consequences should be particularly evident in an older sample because somatic resources decline with age (H. Leventhal et al. 1998) and in light of the fact that behavioral inhibition tendencies are likely stable across the life span (Gest 1997; Magai 2001) and are expected to have a cumulative impact. Our data show that inhibiting emotion was associated with symptomatology in a complex manner. Although the within-race regression (African Americans vs African Caribbeans) showed greater inhibition to be associated with increased hypertension and sleep disturbance, inhibition was not related to health outcome in the general model, although it did interact with ethnicity such that the effects of inhibition on arthritis, hypertension, and respiratory disorder were significantly reduced among Eastern Europeans. Ethnicity also interacted with inhibition in the prediction of sleep disturbance such that the effect of inhibition was less among Africans of Caribbean descent.

As with the results for trait negative emotion, the fact that these effects are complicated and of modest size should not deter researchers from continuing to probe the finer points of how emotion expression and inhibition moderate the course of aging and vulnerability to disease. To take but one example, we know that African Americans sustain high rates of hypertension, validated once again in the present sample (see Table 4 ), and that the known risk factors of smoking, obesity, and drinking are typically higher for persons of this descent (Johnson et al. 1995; H. F. Myers et al. 1995; Najjer et al. 1988; see also Table 2 ). The fact that emotion inhibition was a better predictor than either smoking or alcohol consumption in this sample, and was as useful a predictor as BMI (see Table 5 ), suggests its continued relevance to the health of African Americans and ethnically sensitive models of personality and health. We consider the effects of ethnicity in our data more fully below.

The present data on emotion inhibition might also be taken as indicating that emotion inhibition, per se, does not necessarily have negative health consequences in older adults. Theory regarding emotion expression and health has been developed in ethnically homogenous college-age student samples, and older persons constitute a population that is not frequently studied in expression–health research (Consedine et al. 2002; Smyth 1998). Strictly speaking, research demonstrating that expression of previously undisclosed trauma is associated with improved health has not proven that inhibition is necessarily deleterious to health (Consedine et al. 2002). In addition, research in the area of adult bereavement has found little evidence for a negative health cost of emotional avoidance in older groups. In fact, the 25-month follow-up study of bereavement in middle-aged and older adults by Bonanno and colleagues (see Bonanno et al. 1999; Keltner and Bonanno 1997) found that the expression of negative emotion at baseline predicted both more severe grief and poorer perceived health at a later date. These authors argued that some forms of emotion inhibition may actually be adaptive, in that not expressing negative emotions may facilitate (rather than weaken) social relationships and avoid prolonging or intensifying experiences that may be physiologically toxic to older individuals.

Further assistance in seeking to determine why emotion inhibition was less consistently related to health than was expected may be found in a consideration of ethnicity as a factor in our models. Although the absence of clear empirical literature prevented a priori hypothesis generation, we expected that ethnicity would emerge as an important predictor of illness and that there would be ethnic variation in the variables that predicted illness (interaction effects). Consequently, we use the opportunity afforded us by the unique populations under consideration here to generate a basis for further research.

Initially, the overall model showed that the three dummy-coded ethnicity variables had a significant main effect in the prediction of three of the four health indices. African Americans, African Caribbeans, and immigrant Europeans all reported greater levels of hypertension than U.S.-born Europeans. In contrast, and more favorably for minority group health, African Americans reported less respiratory difficulty or sleep disturbance than U.S.-born Europeans. The regression comparing the two groups of African descent also showed a consistent ethnic effect: African Caribbeans reported significantly lower scores on all four of the health indices than the comparison African American group. This model also showed that the effects of both negative affect and emotion inhibition on sleep disorder were reduced among Caribbeans. In addition, the impact of negative emotion on arthritis was more pronounced in Eastern Europeans than in U.S.-born European Americans, and the effect of emotion inhibition on arthritis, hypertension, and respiratory scores was reduced among Eastern Europeans compared with the U.S.-born majority.

In beginning to understand the complexity evident here, recent views of the impact of emotion inhibition on health have suggested that the question of whether inhibition has negative health consequences may well depend on the reasons for inhibition and the meaning it has for the individual (Gross 1998; Kennedy-Moore and Watson 1999). Although the current study did not assess the reasons underlying particular tendencies to inhibit or express (and to our knowledge there have been few studies of this issue, much less in an ethnic context), it seems plausible that variables related to ethnic background, such as the chronicity of suppressive tendencies or individuals' control over their inhibition, may influence the manner in which regulatory styles relate to health (Consedine et al. 2002). A small literature on emotion socialization has suggested that both African American (e.g., Deater-Deckard et al. 1996; Pinderhughes et al. 2000; Young 1974, Young 1979) and Caribbean parents (Brice 1992; Gopaul-McNicol 1993; Payne 1989) tend to punish uncontrolled displays of emotion in their children, although perhaps for different reasons. Differential socialization of emotion between African American and European American children (Deater-Deckard et al. 1996) may help explain the marked baseline differences in negative emotion evident in Table 2 , in that punitive socialization of emotion may promote regulatory styles whereby negative affect is routed from consciousness (Magai et al. 2001). In terms of distinguishing U.S.-born African Americans from others of African descent, Dilworth-Anderson 1998 has suggested that as a result of their minority status and experiences with racism and discrimination, African Americans are exposed to more anger and distress-provoking conditions than other Americans and experience greater external pressures to conceal their emotions, at least in some situations. In contrast, although cultural values may encourage emotion inhibition, persons of Caribbean descent grew up in lands where they constituted the majority rather than the minority culture and where slavery was abolished in advance of the United States. As such, Caribbeans may inhibit their emotions more voluntarily, or at least for different reasons, than is the case for U.S.-born African Americans. Although the reasons for the difference remain unclear, the fact that the effect of inhibition on sleep scores was less among Caribbeans nonetheless strengthens a view of emotion inhibition in which the manner in which it influences health is related to the meaning of emotion inhibition. Along with age, the meaning that individuals ascribe to emotion inhibition may help determine how difficult the inhibition is, as well as its effects on psychological and physiological systems, and thus health. This possibility would clearly benefit from well-informed empirical research.

Less readily interpretable within a "meaning" framework is the finding that the effects of inhibition on arthritis, hypertension, and respiratory disorder scores were lower among Eastern Europeans than among European Americans (Table 4 ). Historically, Eastern Europeans such as Russians and Ukrainians have been described as being highly emotional as compared with Western Europeans (Consedine and Magai 2002; Wierzbicka 1998, Wierzbicka 1999). Because of the cultural values associated with free expression, we might expect these individuals to suffer greater negative effects from emotion inhibition, but they clearly do not. Instead, ethnicity and emotion inhibition interacted such that the effects of emotion inhibition on health were reduced among Eastern Europeans in comparison with U.S.-born European Americans.

In considering why this somewhat counterintuitive finding may have arisen, however, it is also worth remembering that the effects of negative emotion were significantly greater in the Eastern European group, at least for arthritis and sleep disorder. Taking these two findings together illustrates the intriguing possibility recently raised by Consedine and colleagues 2002 that the typically negative effects of emotion inhibition may be less evident in populations characterized by high levels of negative emotion. In the view offered by these authors, emotion inhibition is a double-edged sword, particularly among older individuals, insofar as it may have both negative and positive implications for health. Because the Eastern European sample had such high levels of trait negative emotion (Table 2 ), and because negative affect is more closely related to arthritis and sleep difficulty, it may be that a degree of inhibition has positive social and physical consequences in addition to the typically negative impact of suppression. Depending on topic and context (Roloff and Ifert 2000), too high a level of emotional disclosure may produce adverse personal and social consequences (Bonanno and Kaltman 1999; Keltner, Ellsworth, and Edwards 1993; Lemerise and Dodge 1993; Levenson and Gottman 1983; Tavris 1984, Tavris 1989) or produce further negative affect (Afifi & Guerrero, 2000).

Although the post hoc nature of the theory offered above must be borne in mind, the totality of findings presented here nonetheless reinforces the notion that health care providers can no longer ignore ethnic factors in health and illness. Although the absence of both theoretical and empirical literatures makes our jobs more difficult in the short run, disambiguating the role of ethnicity in psychosocial models of health may result in more effective treatments and interventions in the long run, as well as assist in fleshing out existing models (Betancourt and Lopez 1993).

In summary, the present study was able to demonstrate that negative affect, emotion inhibition, and their interactions with ethnicity are predictive of several stress-responsive syndromes, even after other well-known sources of risk have been statistically controlled. Although the nature of the association between negative affect and illness in older adults requires validation in longitudinal work, the relation between emotion and illness was nontrivial in magnitude. The levels of variance explained by negative affect, alone or in interaction with ethnicity, were equivalent with other known health risk factors (H. S. Friedman and Booth-Kewley 1987), a particularly noteworthy finding in light of the fact that negative emotion scores exert this effect even where known risk factors are controlled. Our data also suggest that the role of emotion inhibition may need to be rethought when diverse populations of older adults are of interest. Contrary to much theory and previous research in younger participants, emotion inhibition scores were inconsistently related to health indices, suggesting that emotion inhibition is associated with mixed consequences in this group.

The data presented here likewise underscore the importance of an ethnic context in the consideration of aging and health. The four ethnic groups we surveyed varied widely with respect to the predictors of health. Such findings suggest that we cannot assume the models developed among particular populations or ethnic groups can simply be reworked and applied to the study of new samples. However, rather than seek to create completely new models, we must begin seeking data to describe and explain particular ethnic differences and use the understanding thus acquired to inform and detail more general models.

	Acknowledgments

 
The research and work on this article was supported by joint funding from National Institute of General Medical Sciences Grant SO6 GM54650 and Career Leadership Award to Carol Magai from the National Institute on Aging Grant KO7 AG00921.

Received for publication December 19, 2000. Accepted for publication December 5, 2001.

	References

TOP Abstract Methods Results Discussion References

 

• Afifi W. A., Guererro L. K., 2000. Motivations underlying topic avoidance in close relationships. Petronio S., , ed.Balancing the secrets of private disclosures 165-179. Erlbaum, Mahwah, NJ.
• Allen J. P., Land D., 1999. Attachment in adolescence. Cassidy J., Shaver P.., , ed.Handbook of attachment 319-335. Guilford Press, New York.
• Bach P. B., Cramer L. D., Warren J. L., Begg C. B., 1999. Racial differences in the treatment of early-stage lung cancer. New England Journal of Medicine 341:1198-1205. [Abstract/Free Full Text]
• Baltes P. B., Mayer K. U., , ed.The Berlin Aging Study: Aging from 70 to 100 1999Cambridge University Press, Cambridge, England.
• Barefoot J. C., Dahlstrom W. G., Williams R. B., Jr. 1983. Hostility, CHD, incidence, and total mortality: A 25-year follow-up study of 255 physicians. Psychosomatic Medicine 45:59-63. [Abstract/Free Full Text]
• Barnes V., Schneider R., Alexander C., Staggers F., 1997. Stress, stress reduction and hypertension in African Americans: An updated review. Journal of the National Medical Association 89:464-476. [Medline]
• Betancourt H., Lopez S. R., 1993. The study of culture, ethnicity, and race in American psychology. American Psychologist 48:629-637.
• Bonanno G. A., Kaltman S., 1999. Toward an integrative perspective on bereavement. Psychological Bulletin 125:760-776. [Medline]
• Bonanno G. A., Keltner D., 1997. Expressions of emotion and the course of conjugal bereavement. Journal of Abnormal Psychology 106:126-137. [Medline]
• Bonanno G. A., Keltner D., Holen A., Horowitz M. J., 1995. When avoiding unpleasant emotions might not be such a bad thing: Verbal-autonomic response dissociation and midlife conjugal bereavement. Journal of Personality and Social Psychology 69:975-989. [Medline]
• Bonanno G. A., Znoj H., Siddique H. I., Horowitz M. J., 1999. Verbal-autonomic dissociation and adaptation to midlife conjugal loss: A follow-up at 25 months. Cognitive Therapy and Research 23:605-624.
• Brandon T. H., Copeland A. L., Saper Z. L., 1995. Programmed therapeutic messages as a smoking treatment adjunct: Reducing the impact of negative affect. Health Psychology 14:41-47. [Medline]
• Brice J., 1992. West-Indian families. McGoldrick M., Pearce J. K., Giordano J., , ed.Ethnicity and family therapy 123-133. Guilford Press, New York.
• Carstensen L. L., Gottman J. M., Levenson R. W., 1995. Emotional behavior in long-term marriage. Psychology and Aging 10:140-149. [Medline]
• Cassidy J., 1994240).Serial No. Emotion regulation: Influences of attachment relationships. Monographs of the Society for Research in Child Development 59(2–3:240).Serial No.
• Cassidy J., 2000. Adult romantic attachments: A developmental perspective on individual differences. Review of General Psychology 4:111-131.
• Chatters L. M., 1993. Health disability and its consequences for subjective stress. Jackson J. S., Chatters L. M., Taylor R. J., , ed.Aging in black America 167-183. Sage, Newbury Park, CA.
• Cicchetti, D. (Ed.). (1996). Regulatory processes [Special issue]. Development and Psychopathology, Regulatory Processes, 8.
• Cobb J. M. T., Steptoe A., 1996. Psychosocial stress and susceptibility to upper respiratory tract illness in an adult population. Psychosomatic Medicine 58:404-412. [Abstract/Free Full Text]
• Cobb J. M. T., Steptoe A., 1998. Psychosocial influences on upper respiratory infectious illness in children. Journal of Psychosomatic Research 45:319-330. [Medline]
• Cohen C. I., Sokolovsky J., 1979. Clinical use of network analysis for psychiatric and aged populations. Community Mental Health Journal 15:203-213. [Medline]
• Cohen S., Doyle W. J., Skoner D. P., Fireman P., Gwaltney J. M., Newsom J. T., 1995. State and trait affect as predictors of objective and subjective symptoms of respiratory viral infections. Journal of Personality and Social Psychology 68:159-169. [Medline]
• Cohen S., Herbert T. B., 1996. Health psychology: Psychological factors and physical disease from the perspective of human psychneuroimmunology. Annual Review of Psychology 47:113-142. [Medline]
• Consedine N. S., Magai C., 2002. The uncharted waters of emotion: Ethnicity, trait emotion and emotion expression in older adults. Journal of Cross-Cultural Gerontology 17:71-100. [Medline]
• Consedine N. S., Magai C., Bonanno G. A., 2002. Moderators of the emotion inhibition–health relationship: A review and research agenda. Review of General Psychology 6:204-228.
• Cooper C. L., Faragher E. B., 1993. Psychosocial stress and breast cancer: The inter-relationship between stress events, coping strategies and personality. Psychological Medicine 23:653-662. [Medline]
• Creamer P., Hochberg M. C., 1998. The relationship between psychosocial variables and pain reporting in osteoarthritis of the knee. Arthritis Care and Research 11:60-65.
• Deater-Deckard K., Dodge K. A., Bates J. E., Pettit G. S., 1996. Physical discipline among African American and European American mothers: Links to children's externalizing behaviors. Developmental Psychology 32:1065-1072.
• Deurenberg P., Yap M., van Staveren W. A., 1998. Body mass index and percent body fat: A meta analysis among different ethnic groups. International Journal of Obesity 22:1164-1171. [Medline]
• Dilworth-Anderson P., 1998. Emotional well-being in adult and later life among African Americans: A cultural and sociocultural perspective. Schaie K. W., Lawton M. P., , ed.Annual review of gerontology and geriatrics: Vol. 17. Focus on emotion and adult development 282-303. Springer, New York.
• Dressler W. W., 1996. Social identity and arterial blood pressure in the African-American community. Ethnicity & Disease 6:176-189.
• Dusseldorp E., van Elderen T., Maes S., Meulman J., Kraaij V., 1999. A meta-analysis of psychoeducational programs for coronary heart disease patients. Health Psychology 18:506-519. [Medline]
• Eysenck H. J., 1988. The respective importance of personality, cigarette smoking and interaction effects for the genesis of cancer and coronary heart disease. Personality & Individual Differences 9:453-464.
• Eysenck H. J., 1994. Cancer, personality and stress: Prediction and prevention. Advances in Behavior Research and Therapy 16:167-215.
• Fisher L., Feldman S. S., 1998. Familial antecedents of young adult health risk behavior: A longitudinal study. Journal of Family Psychology 12:66-88.
• Fox, N. A. (Ed.). (1994). The development of emotion regulation: Biological and behavioral considerations. Monographs of the Society for Research in Child Development, 59(2–3, Serial No. 240).
• Fraley R. C., Shaver P. R., 2000. Adult romantic attachment: Theoretical developments, emerging controversies, and unanswered questions. Review of General Psychology 4:132-154.
• Friedman H. S., Booth-Kewley S., 1987. The "disease-prone personality": A meta-analytic view of the construct. American Psychologist 42:539-555. [Medline]
• Friedman L., Brooks J. O., Bliwise D. L., Yesavage J. A., Wicks D. S., 1995. Perceptions of life stress and chronic insomnia in older adults. Psychology and Aging 10:352-357. [Medline]
• Fuendeling J. M., 1998. Affect regulation as a stylistic process within adult attachment. Journal of Social and Personal Relationships 15:291-322. [Abstract]
• Gest S. D., 1997. Behavioral inhibition: Stability and associations with adaptation from childhood to early adulthood. Journal of Personality and Social Psychology 72:467-475. [Medline]
• Golden R. R., Teresi J. A., Gurland B. J., 1984. Development of indicator scales for the Comprehensive Assessment and Referral Evaluation (CARE) interview schedule. Journal of Gerontology 39:138-146.
• Gopaul-McNicol S., 1993. Working with West Indian families Guilford Press, New York.
• Greenberg M. A., Stone A. A., 1992. Emotional disclosure about trauma and its relation to health: Effects of previous disclosure and trauma severity. Journal of Personality and Social Psychology 63:74-85.
• Gross J. J., 1998. Antecedent- and response-focused emotion regulation: Divergent consequences for experience, expression, and physiology. Journal of Personality and Social Psychology 74:224-237. [Medline]
• Gross J. J., Carstensen L. L., Pasupathi M., Tsai J., Goetestam Skorpen C., Tsu A. Y. C., 1997. Emotion and aging: Experience, expression, and control. Psychology and Aging 12:590-599. [Medline]
• Gross J. J., Levenson R. W., 1993. Emotional suppression: Physiology, self-report, and expressive behavior. Journal of Personality and Social Psychology 64:970-986. [Medline]
• Gross J. J., Levenson R. W., 1997. Hiding feelings: The acute effects of inhibiting negative and positive emotion. Journal of Abnormal Psychology 106:95-103. [Medline]
• Grossarth-Maticek R., Bastiaans J., Kanazir D. T., 1985. Psychosocial factors as strong predictors of mortality from cancer, ischaemic heart disease and stroke: The Yugoslav prospective study. Journal of Psychosomatic Research 29:167-176. [Medline]
• Grossarth-Maticek R., Kanazir D. T., Schmidt P., Vetter H., 1985. Psychosocial and organic variables as predictors of lung cancer, cardiac infarct and apoplexy: Some differential predictors. Personality & Individual Differences 6:313-321.
• Hicks R. A., Hyler B., 1998. Handedness and stress-related health problems. Perceptual and Motor Skills 87:1042[Medline]
• Hochschild A. S., 1979. Emotion work, feeling rules, and social structure. American Journal of Sociology 85:551-575.
• Holahan C. J., Moos R. H., Holahan C. K., Cronkite R. C., 1999. Resource loss, resource gain, and depressive symptoms: A 10-year model. Journal of Personality and Social Psychology 77:620-629. [Medline]
• Huyser B., Parker J. C., 1998. Stress and rheumatoid arthritis: An integrative review. Arthritis Care and Research 11:135-145.
• Izard C. E., 1972. Patterns of emotions: A new analysis of anxiety and depression Academic Press, New York.
• Izard C. E., Libero D. Z., Putnam P., Haynes O. M., 1993. Stability of emotion experiences and their relations to traits of personality. Journal of Personality and Social Psychology 64:847-860. [Medline]
• Johnson K. W., Anderson N. B., Bastida E., Kramer B. J., Williams D., Wong M., 1995. Panel II: Macrosocial and environmental influences on minority health. Health Psychology 14:601-612. [Medline]
• Keltner D., Bonanno G. A., 1997. A study of laughter and dissociation: Distinct correlates of laughter and smiling during bereavement. Journal of Personality and Social Psychology 73:687-702. [Medline]
• Keltner D., Ellsworth P. C., Edwards K., 1993. Beyond simple pessimism: Effects of sadness and anger on social perception. Journal of Personality and Social Psychology 64:740-752. [Medline]
• Kennedy-Moore E., Watson J. C., 1999. Expressing emotion: Myths, realities, and therapeutic strategies Guilford Press, New York.
• Klohnen E. C., Bera S., 1998. Behavioral and experiential patterns of avoidantly and securely attached women across adulthood: A 31-year longitudinal perspective. Journal of Personality and Social Psychology 74:211-223. [Medline]
• Knupper, G., Fink, R., Clark, W. B., & Goffman, A. S. (1963). Factors related to amount of drinking in an urban community (California Drinking Practices Study in Report No. 6). Berkeley, CA: State Department of Public Health Division of Alcoholic Rehabilitation.
• Koehler T., 1985. Stress and rheumatoid arthritis: A survey of empirical evidence in human and animal studies. Journal of Psychosomatic Studies 29:655-663.
• Labouvie-Vief G., Diehl M., 2000. Cognitive complexity and cognitive-affective integration: Related or separate domains of adult development. Psychology and Aging 15:490-504. [Medline]
• Larsen R. J., Diener E., 1987. Affect intensity as an individual difference characteristic: A review. Journal of Research in Personality 21:1-39.
• Lemerise E. A., Dodge K. A., 1993. The development of anger and hostile interactions. Lewis M., Haviland J. M., , ed.Handbook of emotions 537-546. Guilford Press, New York.
• Levenson R. W., Gottman J. M., 1983. Marital interaction: Physiological linkage and affective exchange. Journal of Personality and Social Psychology 45:587-597. [Medline]
• Leventhal E. A., Hansell S., Diefenbach M., Leventhal H., Glass D. C., 1996. Negative affect and self report of physical symptoms: Two longitudinal studies of older adults. Health Psychology 15:193-199. [Medline]
• Leventhal H., Patrick-Miller L., 2000. Emotions and physical illness: Causes and indicators of vulnerability. Haviland J., Lewis M., , ed.The handbook of emotion 2nd ed. 523-537. Guilford Press, New York.
• Leventhal H., Patrick-Miller L., Leventhal E. A., Burns E. A., 1998. Does stress-emotion cause illness in elderly people?. Schaie K. W., Lawton M.P., , ed.Annual review of gerontology and geriatrics: Vol. 17. Focus on emotion and adult development 138-184. Springer, New York.
• Lillie-Blanton M., Laveist T., 1996. Race/ethnicity, the social environment and health. Social Science & Medicine 43:83-91.
• Lutz C., 1987. Goals, events, and understanding in Ifalaluk emotion theory. Quinn N., Holland D., , ed.Cultural models in language and thought 290-312. Cambridge University Press, Cambridge, MA.
• Lutz C., White G., 1986. The anthropology of emotion. Annual Review of Anthropology 15:405-436.
• Magai C., 2001. Emotion over the lifecourse. Birren J., Schaie K.W., , ed.Handbook of the psychology of aging 299-426. Academic Press, San Diego, CA.
• Magai C., Cohen C., Milburn N., Thorpe B., McPherson R., Peralta D., 2001. Attachment styles in older European American and African American adults. Journal of Gerontology: Social Sciences 56B:S28-S35. [Abstract/Free Full Text]
• Mayne T. J., 1999. Negative affect and health: The importance of being earnest. Cognition and Emotion 13:601-635.
• Mesquita B., Frijda N. H., 1992. Cultural variations in emotions: A review. Psychological Bulletin 112:179-204. [Medline]
• Mesquita B., Frijda N. H., Scherer K. R., 1997. Culture and emotion. Berry J. W., Dasen P. R., , ed.Handbook of cross-cultural psychology: Vol. 2. Basic processes and human development 2nd ed. 255-297. Allyn & Bacon, Needham Heights, MA.
• Mroczek D. K., Kolarz C. M., 1998. The effect of age on positive and negative affect: A developmental perspective on happiness. Journal of Personality and Social Psychology 75:1333-1349. [Medline]
• Myers E., , ed.Challenges of a changing America: Perspectives on immigration and multiculturalism in the United States 1993Austin & Winfield, San Francisco, CA.
• Myers H. F., Kagawa-Singer M., Kumanyika S. K., Lex B., Markides K. S., 1995. Behavioral risk factors related to chronic diseases in ethnic minorities. Health Psychology 14:613-621. [Medline]
• Najjer M. F., Rowland M., Rowland M., 1988. Anthropometric reference data and prevalence of overweight, United States, 1976–1980 (DHHS Publication No. PHS. 87-1688) National Center for Health Statistics, Washington, DC.
• Panksepp J., Miller A., 1995. Emotions and the aging brain: Regrets and remedies. Magai C., McFadden S. H., , ed.Handbook of emotion, adult development and aging 3-26. Academic Press, San Diego, CA.
• Partinen M., 1994. Sleep disorders and stress. Journal of Psychosomatic Research 38:89-91.
• Payne M. A., 1989. Use and abuse of corporal punishment: A Caribbean view. Child Abuse and Neglect 13:389-401. [Medline]
• Pennebaker J. W., 1993. Overcoming inhibition: Rethinking the roles of personality, cognition and social behavior. Traue H. C., Pennebaker J. W., , ed.Emotion, inhibition and health 100-115. Hogrefe & Huber, Seattle.
• Pennebaker J. W., , ed.Emotion, disclosure and health 1995American Psychological Association, Washington, DC.
• Pennebaker J. W., Beall S. K., 1986. Confronting a traumatic event: Toward an understanding of inhibition and disease. Journal of Abnormal Psychology 95:274-281. [Medline]
• Pennebaker J. W., Seagal J. D., 1999. Forming a story: The health benefits of narrative. Journal of Clinical Psychology 55:1243-1254. [Medline]
• Pennebaker J. W., Susman J. R., 1988. Disclosure of traumas and psychosomatic processes. Social Science & Medicine 26:327-332.
• Pinderhughes E. E., Dodge K. A., Bates J. E., Pettit G. S., Zelli A., 2000. Discipline responses: Influences of parents' socioeconomic status, ethnicity, beliefs about parenting, stress, and cognitive–emotional processes. Journal of Family Psychology 14:380-400.
• Polivy J., 1998. The effects of behavioral inhibition: Integration internal cues, cognition, behavior and affect. Psychological Inquiry 9:181-204.
• Raikkoenen K., Matthews K. A., Flory J. D., Owens J. F., 1999. Effects of hostility on ambulatory blood pressure and mood during daily living in healthy adults. Health Psychology 18:44-53. [Medline]
• Raikkoenen K., Matthews K. A., Flory J. D., Owens J. F., Gump B. B., 1999. Effects of pessimism and trait anxiety on ambulatory blood pressure and mood during everyday life. Journal of Personality and Social Psychology 76:104-113. [Medline]
• Roloff M. E., Ifert D. E., 2000. Conflict management through avoidance: Witholding complaints, suppressing arguments, and declaring topics taboo. Petronio S., , ed.Balancing the secrets of private disclosures 151-163. Erlbaum, Mahwah, NJ.
• Saarni C., 1989. Children's understanding of strategic control of emotional expression in social transaction. Saarni C., Harris P., , ed.Children's understanding of emotion 181-208. Cambridge University Press, New York.
• Saarni C., 1992. Children's emotional-expressive behaviors as regulators of others' happy and sad emotional states. Eisenberg N., Fabes R. A., , ed.Emotion and its regulation in early development 91-106. Jossey-Bass, San Francisco.
• Sadeh A., 1996. Stress, trauma, and sleep in children. Child and Adolescent Psychiatric Clinics of North America 5:685-700.
• Sanders-Phillips K., 1996. Correlates of health promotion behaviors in low-income black women and Latinas. American Journal of Preventative Medicine 12:450-458.
• Sarafino E. P., Dillon J. M., 1998. Relationships among respiratory infections, triggers of attacks, and asthma severity in children. Journal of Asthma 35:497-504. [Medline]
• Sarafino E. P., Paterson M. E., Murphy E. L., 1998. Age and the impacts of triggers in childhood asthma. Journal of Asthma 35:213-217. [Medline]
• Smith J., Fleeson W., Geiselmann B., Settersten R. A., Jr. Kunzmann U., 1999. Sources of well-being in very old age. Baltes P. B., Mayer K. U., , ed.The Berlin Aging Study: Aging from 70 to 100 450-471. Cambridge University Press, New York.
• Smith T. W., 1992. Hostility and health: Current status of a psychosomatic hypothesis. Health Psychology 11:139-150. [Medline]
• Smyth J. M., 1998. Written emotional expression: Effect sizes, outcome types, and moderating variables. Journal of Consulting and Clinical Psychology 66:174-184. [Medline]
• Sokolovsky J., Cohen C. I., 1981. Toward a resolution of methodological dilemmas in network mapping. Schizophrenia Bulletin 7:109-116.
• Spera S. P., Buhrfeind E. D., Pennebaker J. W., 1994. Expressive writing and coping with job loss. Academy of Management Journal 37:722-733.
• Tabachinick B. G., Fidell I. S., 2001. Using multivariate statistics Allyn & Bacon, Boston.
• Tavris C., 1984. On the wisdom of counting to ten: Personal and social dangers of anger expression. Review of Personality and Social Psychology 5:170-191.
• Tavris C., 1989. Anger: The misunderstood emotion (Rev. ed.). New York , ed.Touchstone Books
• Teresi J. A., Golden R. R., Gurland B. J., 1984. Concurrent and predictive validity of indicator scales developed for the Comprehensive Assessment and Referral Evaluation interview schedule. Journal of Gerontology 39:158-165.
• Teresi J. A., Golden R. R., Gurland B. J., Wilder D. E., Bennett R. G., 1984. Construct validity of indicator-scales developed from the Comprehensive Assessment and Referral Evaluation interview schedule. Journal of Gerontology 39:147-157.
• Thompson R., 1999. Early attachment and later development. Cassidy J., Shaver P., , ed.Handbook of attachment 265-286. Guilford, New York.
• Tice D. M., Baumeister R. F., 1993. Controlling anger: Self-induced emotion change. Wegner D. M., Pennebaker J. W., , ed.Handbook of mental control 393-409. Prentice Hall, Englewood Cliffs, NJ.
• Watson D., McKee L., 1996. The long-term stability and predictive validity of trait measures of affect. Journal of Personality and Social Psychology 70:567-577. [Medline]
• Weinberger M., Hiner S. L., Tierney W. M., 1987. In support of hassles as a measure of stress in predicting health outcomes. Journal of Behavioral Medicine 10:19-31. [Medline]
• Wierzbicka A., 1998. Russian emotional expression. Ethos 26:456-486.
• Wierzbicka A., 1999. Emotions across languages and cultures: Diversity and universals Cambridge University Press, Paris.
• Young V. H., 1974. A Black American socialization pattern. American Ethnologist 1:415-431.
• Young V. H., 1979. Family and childhood in a southern Negro community. American Anthropologist 72:269-288.
• Zautra A. J., Burleson M., Matt K., Roth S., Burrows L., 1994. Interpersonal stress, depression, and disease activity in rheumatoid arthritis and osteoarthritis patients. Health Psychology 13:139-148. [Medline]

This Article Abstract Full Text (PDF) Services Download to citation manager PubMed PubMed Citation