Psychosomatic Medicine 61:469-475 (1999)
© 1999 American Psychosomatic Society
Genetic and Environmental Influences on Coping Styles: A Twin Study
Andreas Busjahn, PhD,
Hans-Dieter Faulhaber, MD,
Kristina Freier, MD and
Friedrich C. Luft, MD
From the Franz Volhard Clinic, Max Delbrück Center for Molecular Medicine, Medical Faculty of the Charité, Humboldt University, Berlin, Germany.
Address reprint requests to: Andreas Busjahn, PhD, Franz Volhard Clinic, Wiltbergstr. 50, 13122 Berlin, Germany. Email: busjahn{at}fvk-berlin.de
 |
ABSTRACT
|
|---|
OBJECTIVE: Coping styles are generally considered to be environmentally driven, primarily by family influences. However, because personality traits are commonly influenced by genetic effects, we hypothesized that heredity is also important for coping.
METHODS: We tested this hypothesis by assessing 19 coping styles, as well as four secondary coping factors, by questionnaire in 212 pairs of monozygotic and dizygotic twins. We then examined heredity by structural equation modeling.
RESULTS: All coping styles showed evidence of genetic influences. The coping styles shared one common genetic factor. In addition, each coping style was also influenced by other separate genetic factors. Shared environment had no significant influence on coping styles. Three of 19 more specific coping styles showed shared environmental effects as well as genetic influences, 14 were solely under genetic influences, and two showed only shared environment effects.
CONCLUSIONS: We suggest that hereditary effects on certain coping style preferences cannot be explained solely by genetic influences on major personality traits and temperament. An analysis of the relationships between coping and personality in twin subjects may elucidate the distinction between genetic and environmental effects.
Key Words: coping • genetics • environment • twins
Abbreviations: DZ = dizygotic; MZ = monozygotic; SD = standarddeviation; SVF = Stressverarbeitungs-Fragebogen (copingquestionnaire).
|
INTRODUCTION
|
|---|
The psychological processes aimed at diminishing or terminating stress are called "coping" processes (1). A working definition of coping might be "the things people do to avoid being harmed by life-strains" (2). Complex factors, such as personality, attitudinal, cognitive, and expectancy elements, are involved (3). Coping is generally assumed to be a learned behavior; however, genetic factors also play a role (4). Kendler et al. (5) studied female twins and used the 14-item "Ways of Coping Checklist." They identified three scales, turning toward others, problem solving, and denial. Their heritability estimates were 30% for turning toward others and problem solving; however, they could find no significant shared environmental influence on these two coping styles. On the other hand, the influence of shared environment for denial was significant at 20%, but no genetic effect was found. Kendler et al. also showed that the equal environment assumption was valid for the coping behaviors measured. Neither the number of social contacts between the twins nor environmental similarities in childhood influenced the test scores. These results agree with those from a study in children aged 9 to 16 years (6). Mellins et al. (6) assessed coping with a structured interview and questionnaire. Four of seven coping scales exhibited genetic influences in their study, two were influenced by shared environmental factors, and one was influenced by both. Another twin study examining genetic influences on coping styles was performed in the context of life situation and self-reported health in Sweden (7). Although heritability was not estimated, the intraclass correlations for MZ and DZ twins in that study allow a rough estimate. The questionnaire used gives a single result for the "sense of coherence," with MZ correlations more than twice those of same-sex DZ correlations. This result indicates heritability due to nonadditive genetic influences.
Additional indirect evidence can be drawn from studies relating coping to personality. There are well-established relationships between major personality traits, like neuroticism or extroversion, and different coping styles, with about 20% of coping variance being explained by personality (see Ref. 8 for review). Because personality itself is partially heritable, such influences may indirectly affect coping as well. However, neither coping nor personality is static. Thus, the correlation between the two is not necessarily unidirectional; coping might influence personality development in adulthood as well as in childhood (9). To gain insight into the genetic influences on coping, we performed a study in adult male and female MZ and DZ twins to test the hypothesis that coping styles are influenced by genetic variability. We then tested the hypothesis that different coping styles share some genetic influences and also have other unique, unshared genetic components.
|
METHODS
|
|---|
Subjects
Our sample consisted of 212 twin pairs, 117 MZ twins (81 female and 36 male) and 95 DZ twins (47 female, 12 male, and 36 different sexes), aged 34 ± 14 years, recruited by calls for volunteers in articles about twin research published in the print media (10–12). MZ and female twins exceeded the expected population mean, a common phenomenon in self-selected twin samples. Because the ages, gender distribution, and geographic origins within Germany were not significantly different between the zygosity groups, and because sex-specific heritability was not estimated, the results of our study are likely to be unbiased. The sample was not selected with respect to any of the phenotypes measured. The protocol was approved by the Ethics Committee on Human Subjects of Humboldt University, and written informed consent was obtained from each subject. Zygosity determinations were made with five microsatellite markers coamplified by polymerase chain reaction (13). The probability of a same-sex DZ twin pair being concordant for all markers tested was .006.
Measures
Coping was assessed with use of the German SVF questionnaire (14). The questionnaire was completed by the twins under the supervision of a trained research nurse. The questionnaire includes 19 scales for different types of reactions to an unspecified range of situations that impair, adversely affect, irritate, or disturb the emotional equilibrium or balance of the subject. The questionnaire is similar to the dispositional form of the COPE questionnaire (15). Each scale consists of six items that are answered using a five-point scale according to the probability of that reaction. Example items of the SVF are listed in Table 1. The reliability of all scales was sufficiently high (median Cronbach’s 
= 0.8; see Table 2 for test–retest reliability reported by the test authors). The validity has been tested by intercorrelations between subscales, by correlations with a variety of questionnaires, and by specification of different stressful situations. Correlations between SVF scales and neuroticism ranged from 0.00 to 0.59, and correlations between SVF scales and extroversion ranged from 0.00 to 0.36, confirming the relationship between coping and personality described by Watson and Hubbard (8). Normal values (T scale; mean = 50 and SD = 10) for the SVF for German volunteers according to gender for the 20- to 64-years-old age range are available. No age effects were found in the standardized sample.
View this table:
[in this window]
[in a new window]
|
Table 2. SVF Questionnaire Scores by Zygosity and Reliability Coefficients (Test–Retest Correlation After 4 weeks, n = 200), given by the test authors for the standardization sample (19)a
|
|
Because the subtests of the SVF questionnaire were not totally independent, secondary factors based on the intercorrelation were determined by the test authors and confirmed in our own sample. In a four-factor solution, 61% of variance is explained; in a six-factor solution, 72% is explained. Two of the six factors are very specific and combine just two scales. We therefore elected to reduce the number of coping factors in relation to sample size and based our analysis on the four-factor solution. The four factors can be characterized as follows: SVF1, defense (play down, compare with others, guilt defense, and ego boost); SVF2, emotional coping (flight, social retreat, rumination, resignation, self-accusation, and aggression); SVF3, substitution (substitutional satisfaction, need for social support, and self medication/alcohol abuse); and SVF4, active coping (situational control, reaction control, positive self-instruction, and avoidance).
Statistical Analyses
Statistical analyses were performed using a computer program (SPSS Inc., Chicago, IL). Heritability estimates were obtained by structural equation modeling using the Mx computer program (16, 17). By comparing MZ twin variances and covariances with those of DZ twins, we estimated the relative influence of additive genetic factors (a2), nonadditive genetic factors (d2), shared environmental factors (c2), and unique environmental factors (e2) contributing to individual differences in coping behavior. Additive genetic variance is the genetic variance associated with the average effect of alleles without allelic interaction or gene–gene interaction. Dominance (allelic interaction within a gene) or epistasis (gene–gene interaction) will lead to nonadditive genetic variance. Shared environmental factors are those operating within families and provide a potential source of family resemblance, as opposed to unique (or random) environmental factors, which contribute to differences between family members. Strong additive genetic effects will lead to significant correlations in MZ and DZ twins, with the MZ correlation twice the DZ correlation. Nonadditive genetic effects or dominance will lead to an increase in the difference between MZ and DZ twin correlations, because the chance of a DZ twin pair sharing two alleles identical by descent is only 25%. Interactions between gene loci (epistasis) will affect the difference in the MZ/DZ twin correlation in the same way but with an even stronger effect when more than two loci are involved. Epistasis is not included in the model but will increase estimates of dominance. Epistasis in complex phenotypes is highly important; however, estimating epistasis is extremely difficult (18). Another source of strong nonadditive genetic variance is the dependency of gene action on the genetic background, as was shown in plant genetics (19). Shared environmental influences will increase MZ and DZ correlations to the same extent, thus decreasing the difference between correlations. The shared environment consists of the family environment and shared influences of school, social class, etc., whereas the nonshared environment includes factors that are unique for each member of a twin pair. Measurement errors will contribute to this source of variation as well; these errors cannot be corrected within sibling pairs. Although nonadditive genetic effects and shared environment might influence a given trait, they are confounded. Nonadditive genetic effects and shared environment have opposite effects on DZ covariance. Thus, the two effects cannot be estimated simultaneously in the same model. We began with a model encompassing additive genetic influences (A), shared influences (C), and nonshared environmental influences (ACE model) and fitted nested models without genetic and/or shared environmental influences. If the MZ correlation of a coping scale was more than twice the DZ correlation, nonadditive genetic effects (D) were tested by fitting ADE and nested models as a post hoc test. The sum of additive and nonadditive genetic influences in an ADE model equals the heredity as estimated in an ACE model. Thus, our analysis does not overestimate heritability. Full and nested models were compared by likelihood ratio 
2 tests, models with the same df by the Akaike’s information criterion, as described elsewhere (17, 20). Testing multiple measures within one sample increases the probability of type 2 errors. Thus, heritability estimates for the questionnaire’s 19 subscales and the four secondary factors are given to illustrate results, to allow comparison with other, more specific coping scales, and to provide starting points for the generation of future hypotheses.
We performed a multivariate path analysis to test for genetic influences on coping. This approach extends the univariate heritability analysis by allowing genetic covariance to be shared by different traits. Thus, the correlation between phenotypes is tested for underlying common genetic factors. An example for the bivariate case with no shared environmental influences is given in Figure 1. Factor Ac is a genetic influence shared by both phenotypes, whereas As is a genetic factor specific to phenotype 2. Corresponding environmental influences are Ec and Es. This bivariate case can be extended to multiple measures. By Cholesky decomposition, a phenotypic variance/covariance matrix is factored as the product of a triangular matrix and its transposition. By comparing nested models, we tested whether all four coping factors shared genetic influences as well as the related hypothesis that all genetic influences on the different coping styles were attributable to a single shared genetic factor. This approach permitted the completion of submodels with a single genetic factor and four unshared genetic factors.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 1. Bivariate path model that decomposes the phenotypic variance and covariance for two given phenotypes, for example, two different coping styles, into additive genetic (Ac) and environmental (Ec) contributions that are common to both phenotypes and factors that are specific to one phenotype only, that is, additive genetic (As) and environmental (Es) contributions.
|
|
|
RESULTS
|
|---|
Scores for the SVF questionnaire subscales are given in Table 2. None of the scales showed a significant difference between MZ and DZ twins nor a significant deviation from the expected population mean of 50. Thus, our MZ and DZ twins seem to be fairly representative of the German population. Because the scores are gender specific, the overrepresentation of female twins will not influence the results. Table 3 shows the twin analysis and provides the significant effects of additive genetic factors (a2), nonadditive genetic factors (d2), shared environmental factors (c2), and nonshared environmental factors (e2). There was no homogeneous pattern of heritability, although numerous parameters showed genetic variability. Fourteen of the 19 scales showed genetic influences without significant shared environmental effects. On the other hand, two scales showed effects of family environment without genetic effects. Finally, three scales showed a combination of genetic and shared environmental effects.
View this table:
[in this window]
[in a new window]
|
Table 3. Percentage of Total Variance Explained by Additive Genetic Factors (a2), Nonadditive Genetic Factors (d2), Shared Environmental Factors (c2), and Nonshared Environmental Factors (e2). Together With the Fit Index (c2/df) for the Selected Modela
|
|
Table 4 gives the correlations within and between twin pairs for secondary coping factors. These relationships form the basis of the subsequent analysis. The results of the multivariate path analysis are given in Table 5, which shows the parameter estimates for the best-fitting model. We were interested in the hypothesis that coping styles are not only influenced by specific genetic factors but also share some genetic influences. We had no specific hypotheses as to which coping scales might share genetic influences. Therefore, we did not further reduce the model. The corresponding fit indices are presented in Table 6 for the fully saturated model (I); the selected model without shared environmental effects, which allowed genetic variance and covariance (II); a reduced model allowing only specific genetic variance for each of the four coping factors (III); and a model with only one genetic factor shared by all coping factors (IV). Dropping the shared environmental influences from the saturated model did not affect model fit, which is in agreement with the univariate analysis of the secondary coping factors. The model allowing only four specific genetic factors without overlap between coping styles (model III) resulted in a significant decrease in fit. When the model was tested with only one common genetic factor without additional specific genetic influences (model IV), the fit decreased significantly as well. Thus, the results support a complex genetic architecture with specific as well as shared genetic influences for each coping behavior.
View this table:
[in this window]
[in a new window]
|
Table 4. Between-Twin and Within-Twin Correlations for Secondary Coping Factors Derived From the SVF Questionnaire for MZ and DZ Twinsa
|
|
|
DISCUSSION
|
|---|
The important finding in our study is that the four coping factors, defense, emotional coping, substitution, and active coping, showed evidence of genetic variance. These factors were derived from a 19-scale coping style questionnaire; 17 of the 19 coping styles showed evidence of heritability. We demonstrated that there is no single genetic factor shared by all different coping styles but rather that there are both specific and shared genetic influences for different coping factors. In terms of these secondary factors, there was no evidence of influences of shared environment on coping. However, this finding does not allow the conclusion that shared environmental influences on coping are negligible. Shared environment could exert effects on a different level, such as on more specific coping strategies. To allow generation of more specific hypotheses on genetic and environmental influences, we performed structural equation modeling for all 19 scales of the coping questionnaire. We understand that the comparison of so many phenotypes with a moderate sample size raises problems of multiple testing and can only illustrate the issue without allowing for statistical testing of hypotheses regarding differences between coping scales.
As in earlier studies, we did not find a homogeneous heritability pattern. Seventeen of the 19 coping scales showed some evidence of genetic variance. Eight showed evidence of additive genetic factors, and 13 were suggestive of nonadditive genetic effects. These nonadditive genetic influences might indicate interactions between genes in the complex genetic background of psychological traits. Such nonadditive genetic influences have been found in other studies (10–12) and may be systematic in twin studies on personality traits, as confirmed in larger samples (21). Shared environmental factors were important for five coping styles, whereas all 19 involved nonshared environmental factors. The moderate 0.5 correlations in MZ twins emphasize the importance of individual environmental influences and possible gene–environment interactions. Given the good reliability estimates for the coping questionnaire, measurement errors should have only a minor influence. Although the preference of coping strategies is influenced by genetic factors, individual coping strategies are developed in cointeraction with the environment.
The first three coping styles (play down, compare with others, and guilt defense) can all be classified as intrapsychic behavior and form an internal dialogue. Nevertheless, we cannot generally conclude that internal coping behavior is heritable and that external coping behavior is learned. Scale 4, distraction from situation, is a cognitive strategy as well but nevertheless showed no genetic influence. The next coping style, substitutional satisfaction, is an acting-out behavior, although the motivation for this kind of behavior in parents would not necessarily be clear to their children. The genetic component of this behavior with the lack of shared environmental influences would argue against learning from parental example in the development of this behavior.
Situational control is accepted as an appropriate and mature way of handling problems and is likely to be consciously included by parents as a strategy in child rearing. Shared environment is likely to complement genetic influences and reflects upbringing as well as learning by example. The reaction control mechanism is an internal coping style defined by the lack of external reactions to stress. This mechanism is less likely to be learned by mimicking the parents, as is reflected by the lack of shared environmental influences in the presence of genetic effects. Scales 11 and 12, avoidance and flight tendency, are both related to the trait of fear; however, they showed differences in family transmission. Although the feeling of fear and thereby the wish to leave threatening situations has a genetic component, the ability to control the related behavior (avoidance) is also influenced by the family environment. A rather surprising result was revealed for the aggression scale (scale 18). We found no genetic influence for this variable, even though in a personality questionnaire we obtained a significant heritability score of 55% for the aggression trait (A. Busjahn, unpublished observations). A possible explanation could be that stress-related reactive aggression is independent of personality and may instead be learned from parental models. The genetic influence found for self-medication/alcohol use is in agreement with studies on alcohol and drug abuse and indicates that this strategy at least partially serves as stress relief.
Our results suggest that heritable and learned coping styles cannot be discriminated by means of a simplistic classification such as inner versus outer behavior. Furthermore, there was no clear pattern in the correlations between coping styles, with and without genetic influences, and either neuroticism or extroversion (19). The genetic factors relevant for coping may be the same factors influencing major personality traits. Given the correlation between major personality traits and coping styles, which ranged from 0.00 to 0.60 (19), and given the complex genetic architecture of coping, only a small part of coping heritability is likely to be due to the inheritance of personality traits. Although this hypothesis is only speculative, multivariate analyses of both coping and personality in extended samples may give better insight into this complex relationship.
Interindividual differences in coping behavior patterns are best explained by simultaneous environmental and genetic interactions. Because the structuring of the questionnaire is only one of many possibilities, different types of classification might result in different heritability patterns. Although our hypothesis of genetic influences on coping was confirmed, the extent of genetic influences can be only roughly estimated from our results. However, our data highlight the necessity of considering genetic influences in this field. We believe the twin subjects were representative of the German population, because the scale results were not significantly different between MZ and DZ twins. Furthermore, the mean population score of 50 was closely approximated by MZ and DZ twins in all scale categories.
We believe our results extend beyond the study of personality traits. Coping strategies are accompanied by simultaneous adjustments in autonomic nervous system tone, heart rate, heart rate variability, baroreceptor function, and blood pressure. These physiological variables may have a bearing on chronic cardiovascular disease. A significant impact of stress responses on cardiovascular risk has been demonstrated (22, 23). To better understand the relationship between stressors and physiological responses, the mediating factors of personality and coping warrant investigation (24–26). Furthermore, improved molecular genetic techniques combined with twin studies will allow assessment of candidate genes influencing physiological (27) and psychological behavior (28). The DZ twins and their parents lend themselves to linkage analysis with molecular genetic markers, after which additional association studies are possible on the same sample. We used this approach in earlier studies (29).
In summary, our data in a large number of normal male and female twins confirm a strong genetic influence on coping strategies. We found that most (17 of 19) coping styles were under the influence of genetic variability. Both additive and nonadditive genetic effects were observed. Our data suggest that a simplistic notion of inner (genetic) and outer (learned) behaviors must be revised. Each individual behavior is a complex interplay of heredity and environment. Given the fact that psychological variables show striking similarities across countries (30), we believe that our results apply to countries outside Germany.
|
ACKNOWLEDGMENTS
|
|---|
We thank Professor Alois Angleitner and his colleagues at the University of Bielefeld for their support in recruiting participants.
Received for publication November 17, 1998.
Revision received March 31, 1999.
|
REFERENCES
|
|---|
-
Costa PT, McCrae RR. Personality, stress, and coping: some lessons from a decade of research. In: Markides KS, Cooper CL, editors. Aging, stress, and health. New York: John Wiley and Sons; 1989. p. 269–85.
-
Pearlin LI, Schooler C. The structure of coping. J Health Soc Behav 1978; 19: 2–21.[Medline]
-
Feifel H, Strack S, Nagy VT. Coping strategies and associated features of medically ill patients. Psychosom Med 1987; 49: 616–25.[Abstract/Free Full Text]
-
Theorell T. Life events and manifestations of ischemic heart disease: epidemiological and psychological aspects. Psychosomatics 1980; 34: 135–48.
-
Kendler KS, Kessler RC, Heath AC, Neale MC, Eaves LJ. Coping: a genetic epidemiological investigation. Psychol Med 1991; 21: 337–46.[Medline]
-
Mellins CA, Gatz M, Baker L. Children’s methods of coping with stress: a twin study of genetic and environmental influences. J Child Psychol Psychiatry 1996; 37: 721–30.[Medline]
-
Lange AL, Fischbein S. Life situation, self-reported health and coping ability of 35-year-old twins and controls: a follow-up of a longitudinal Swedish twin study at adolescence. Acta Genet Med Gemellol (Roma) 1996; 45: 405–16.[Medline]
-
Watson D, Hubbard B. Adaptational style and dispositional structure: coping in the context of the five-factor model. J Pers 1996; 64: 737–74.
-
Aldwin CM, Sutton KJ, Lachmann M. The development of coping in adulthood. J Pers 1996; 64: 837–72.[Medline]
-
Ball D, Hill L, Freeman B, Eley TC, Strelau J, Riemann R, Spinath FM, Angleitner A, Plomin R. The serotonin transporter gene and peer-rated neuroticism. Neuroreport 1997; 8: 1301–4.[Medline]
-
Busjahn A, Faulhaber H-D, Viken RJ, Rose RJ, Luft FC. Genetic influences on blood pressure with the cold pressor test: a twin study. J Hypertens 1996; 14: 1195–9.[Medline]
-
Busjahn A, Knoblauch J, Knoblauch M, Schuster H, Bohlender J, Faulhaber H-D, Luft FC. Angiotensin converting enzyme and angiotensinogen gene polymorphisms, plasma levels, and left ventricular size: a twin study. Hypertension 1997; 29: 165–70.[Abstract/Free Full Text]
-
Becker A, Busjahn A, Faulhaber HD, Baehring S, Robertson J, Schuster H, Luft FC. Twin zygosity: automated determination with microsatellites. J Reprod Med 1997; 42: 260–6.[Medline]
-
Jahnke W, Erdmann G, Kallus W. Der Stressverarbeitungs-Fragebogen [The coping questionnaire]. Göttingen: Verlag für Psychologie; 1985.
-
Carver CS, Scheier MF, Weintraub JK. Assessing coping strategies: a theoretically based approach. J Pers Soc Psychol 1989; 56: 267–83.[Medline]
-
Neale MC. Mx: Statistical modeling, 4th ed. Richmond (VA): Department of Psychiatry, University of Virginia; 1997.
-
Neale MC, Cardon LR. Methodology for genetic studies of twins and families. Dordrecht: Kluwer; 1992.
-
Frankel WN, Schork NJ. Who’s afraid of epistasis? Nat Genet 1996; 14: 371–3.[Medline]
-
Lynch M, Walsh B. Genetics and analysis of quantitative traits. Sunderland, UK: Sinauer; 1997.
-
Akaike H. Factor analysis and AIC. Psychometrika 1987; 52: 317–32.
-
Eaves LJ, Heath AC, Neale MC, Hewitt JK, Martin NG. Sex differences and nonadditivity in the effects of genes on personality. Twin Res 1998; 1: 131–7.[Medline]
-
Alderman MH, Ooi WL, Madhavan S, Cohen H. Blood pressure reactivity predicts myocardial infarction among treated hypertensives. J Clin Epidemiol 1989; 43: 859–66.
-
Rostrup M, Westheim A, Kjeldsen SE, Eide I. Cardiovascular reactivity, coronary risk factors, and sympathetic activity in young men. Hypertension 1993; 22: 891–9.[Abstract/Free Full Text]
-
Müller MM, Elbert T. Ärgerverarbeitung bei kardiovascular Erkrankten [Coping with anger by cardiovascular patients]. Psychother Psychosom Med Psychol 1994; 44: 240–6.[Medline]
-
Perini C, Smith DH, Neutel JM, Smith MA, Henry JP, Buhler FR, Weiner H, Weber MA. A repressive coping style protecting from emotional distress in low-renin essential hypertensives. J Hypertens 1994; 12: 601–7.[Medline]
-
Vogele C, Steptoe A. Emotional coping and tonic blood pressure as determinants of cardiovascular responses to mental stress. J Hypertens 1992; 10: 1079–87.[Medline]
-
Knoblauch H, Busjahn A, Münter S, Nagy Z, Faulhaber H-D, Schuster H, Luft FC. Heritability analysis of lipids and three gene loci in twins link the macrophage scavenger receptor to high-density lipoprotein cholesterol concentrations. Arterioscler Thromb Vasc Biol 1997; 17: 2054–60.[Abstract/Free Full Text]
-
Plomin R, Caspi A. DNA and personality. Eur J Pers 1998; 12: 387–407.
-
Nagy Z, Busjahn A, Bähring S, Faulhaber H-D, Gohlke H-R, Knoblauch H, Schuster H, Luft FC. Quantitative trait loci for blood pressure exist near the IGF-1, the Liddle syndrome, and the angiotensin II-receptor gene loci in man. J Am Soc Nephrol. In press 1999.
-
Weissman MM, Bland RC, Canino GJ, Faravelli C, Greenwald S, Hwu HG, Joyce PR, Karam EG, Lee CK, Lellouch J, Lepine JP, Newman SC, Rubio-Stipec M, Wells JE, Wickramaratne PJ, Wittchen H, Yeh EK. Cross-national epidemiology of major depression and bipolar disorder. JAMA 1996; 276: 293–9.[Abstract]
This article has been cited by other articles:
|
|
|
|
 
X. Wang, R. Trivedi, F. Treiber, and H. Snieder
Genetic and Environmental Influences on Anger Expression, John Henryism, and Stressful Life Events: The Georgia Cardiovascular Twin Study
Psychosom Med,
January 1, 2005;
67(1):
16 - 23.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
E. S. Epel, B. McEwen, T. Seeman, K. Matthews, G. Castellazzo, K. D. Brownell, J. Bell, and J. R. Ickovics
Stress and Body Shape: Stress-Induced Cortisol Secretion Is Consistently Greater Among Women With Central Fat
Psychosom Med,
September 1, 2000;
62(5):
623 - 632.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
