Psychiorry Research. 23, IV- I45 Elsevier

137

Dexamethasone Suppression Test and Coping Behavior in Psychosocial Stress

Ede Frecska, Hedvig LukBcs, Mihaly Arat6, L&z16 M6d, Antal Alfiildi, and lstvan Magyar

Received August 26, 1986; revised version received February 12, 1987; accepted February 17. 1987.

Abstract. The dexamethasone suppression test (DST) and the Minnesota Multiphasic Personality Inventory (MMPI) were administered to 144 healthy inductees on day 2 of military service. One hundred and four of them completed a I20-item questionnaire describing their coping responses to this particular challenge. Thirty-six subjects (25%) failed to suppress plasma cortisol adequately. Their mean scores on the MM PI clinical standard scales were within the normal range. High postdexamethasone cortisol levels were associated with denial and passivity, and with low demand for social support. These results suggest that the DST might be more related to coping with a stressor than to a specific diagnosis. The authors speculate that high hypothalamic-pituitary-adrenal activity may have a primary role in psychological defense promoting inattention to the aversive aspects of stressful situations.

Key Words. Dexamethasone suppression test, stress, coping, defense.

The considerable amount of psychiatric research on the dexamethasone suppression test (DST) has focused on its clinical relevance and diagnostic validity (Carroll, 1985). Attempts to establish its psychological parallels and possible etiological associations have been less frequently undertaken (Haier, 1983; Caine et al., 1984; Zimmerman et al., 1984; Reus et al., 1985). A consensus has not yet been reached on the mechanism underlying DST nonsuppression or on its psychopathological correlates (Krishnan et al., 1985; Brown et al., 1986). Generally, a positive DST is attributed to enhanced activity of the hypothalamic-pituitary-adrenal (HPA) axis, a system that is sensitive to psychological stress. An early important discovery in psychoneuroendocrinology was the finding that nonphysical stimuli can stimulate HPA activity, but that great individual differences exist in the susceptibility to such stimulation. These differences appear to be largely determined by the individual’s appraisal of the situation and his coping style (Rose, 1984).

The effect of stress and anxiety on the DST has been relatively little studied. Ceulemans et al. (1985) have reported that approximately 50% of subjects exposed to a stressful situation, which was associated with imminent surgery, failed to suppress plasma cortisol adequately after an overnight dose of I mg dexamethasone.

Ede Frecska, M.D., DipI.-Psychol., and Mihaly Arat6, M.D., Ph.D., are in the National Institute of Nervous and Mental Diseases, Budapest. Hedvig LukHcs, M.D., Laszlo M6d, M.D., Antal Alfiildi, M.D., and Istvan Magyar, M.D., Ph.D., are in the Department of Psychiatry, Semmelweis University School of Medicine, Budapest. (Reprint requests to Dr. E. Frecska, National Institute of Nervous and Mental Diseases, Budapest 27, Pf I, 1281 Hungary.)

Ol65-1781/88/$03.50 @ 1988 Elsevier Scientific Publishers Ireland Ltd.

138

Although the level of state anxiety in nonsuppressors was significantly higher than in controls, their individual levels of anxiety varied considerably. The authors attributed this variability to individual differences in coping with the stressful situation. However, anxiety, per se, is not a good predictor of postdexamethasone cortisol level. In the studies of Faludi et al. (1986) and Schweizer et al. (1986), indices of anxiety did not predict suppressor from nonsuppressor status among patients with panic and anxiety disorders. The vast majority of studies indicate normal DST results in panic disorder (Roy-Byrne and Uhde, 1985). The issue of the psychological correlates of nonsuppression is also confused by the observation of abnormal DST results in mania and not solely in the mixed manic-depressive or dysphoric states (Rihmer and Arato, 1984; Cookson, 1985). Moreover, despite the approximately 50% sensitivity of the DST as a test of depression, dysthymia-dysphoria is not a good predictor of suppressor/ nonsuppressor status either.

To date, it has been unclear which psychological factor or factors are associated with abnormal DST results, and the psychophysiological meaning of the HPA hyperfunction in psychosocial stress also remains obscure. From our point of view, a particular hormonal response is a kind of covert behavior that may be strongly connected to a particular behavioral strategy. In the case of HPA hyperactivity, this hypothetical behavioral strategy could be common in distinct psychopathological states or in normal conditions under stress, and it could be overrepresented in

depression. In the present study, we investigated personality styles and processes of coping

that are associated with high HPA activity in the stressful situation of induction into obligatory military service.

Methods

The DST and the Minnesota Multiphasic Personality Inventory (MMPI) were administered to 144 healthy inductees (median age 20, range 18-22 years) on the second day of their military service. During the first 2 days, they had no physical training. Before induction, all subjects underwent a complete physical and psychiatric examination, and detailed medical and psychiatric histories were also obtained in each case. Subjects with a history of psychiatric illness were not included in this study. For technical reasons only 104 of 144 subjects completed a l20-item questionnaire describing their coping process in response to the stress of military induction. Blood samples for plasma cortisol measurement were drawn at 8 a.m., IO hours after oral administration of I mg dexamethasone, and were assayed by a competitive protein-binding method (Murphy, 1967) with an intra-assay variance below 6% and an interassay variance below 12%. Nonsuppression was defined as > 5 I.rg/dl cortisol in the sample. The coping questionnaire was based on the revised version of Lazarus’ Ways of Coping checklist (Folkman and Lazarus, 1985) plus McCrae’s (1984) additional items, but instead of a 4-point Likert scale, we used the “Yes/No” response format of the original version.

After I month of military training, the DST was repeated in the same way. Student’s r test, ~2 tests, and factor analysis (BMDP4M) with principal component

extraction and varimax rotation were used for statistical calculations.

Results

Within 2 days of starting their military service, 36 (25%) of the 144 subjects failed to suppress plasma cortisol below the 5 pg/dl cutoff point. Using this cutoff criterion in

139

our standard DST method (sampling at 8 a.m. and 3 p.m.), we have found a 66% nonsuppression rate in patients with primary endogenous depression and a 4% false-

positive test result in healthy, unstressed controls (Rihmer and Arat6, 1984). We also reported (Arat et al., 1983) that using only one sample instead of two would have decreased the sensitivity of the DST from 52% to 41% in a total sample of psychiatric inpatients. In view of the fact that only one morning postdexamethasone cortisol measurement was used in this study, an even higher nonsuppression rate than the 25% found here could be expected with the traditional DST method in this stressful situation. Therefore, nonsuppressors are probably underrepresented, and the suppressors may include a significant number of subjects who would have been nonsuppressors at later sampling times (false suppressors).

In an effort to exclude the effect of false suppressors in the statistical analysis, we used extreme quartile comparisons in the MMPI calculations. The upper quartile is considered the DST nonsuppressor group. The lower quartile (subjects with postdexamethasone cortisol level < 0.5 pg/dl) is unlikely to include false suppressors, who probably fall within the interquartile half-range of our morning

sample. Mean (+ SD) T scores on the MMPI clinical scales and calculated Anxiety Indices (Al’s) are presented in Table I.

1 ‘able 1. T scores’ of dexamethasone suppression test (DST) nonsuppressor and DST suppressor inductees on Minnesota Multiphasic Personality Inventory (MMPI) clinical scales and their calculated Anxiety Indices2 (mean f SD)

L F K Hs D Hy Pd

DST nonsuppressors (upper

quartile, n = 36) 54.4 54.8 49.2 63.0 54.6 47.0 53.9

f 9.58 + 9.97 f 7.66 f 8.71 f 6.68 f 8.76 + 7.50

(1) (2) (0) (4) (2) (0) (1)

DST suppressors, (lower

quartile, n = 36) 52.4 57.3 47.6 66.6 55.6 50.8 55.9

+ 9.29 + 9.04 f 9.08 + 9.60 + 7.05 k13.20 + 8.26

(0) (4) (0) (6) (0) (2) (1)

t 0.90 1.11 0.81 1.67 0.62 1.44 1.08

P NS NS NS NS NS NS NS

Nlf Pa Pt SC Ma Si Al

DST nonsuppressors (upper

quartile, n = 36) 49.9 52.3 58.8 57.8 54.3 54.1 57.7 f 5.16 f 6.13 f 4.79 f 7.28 + 8.44 f 9.26 +11.36

(0) (0) (1) (4) (2) (2) (4)

DST suppressors (lower

quartile, n = 36) 51.1 56.0 59.3 59.6 56.9 54.1 55.1 f 4.36 f 6.02 f 5.01 k 6.40 f10.14 f 6.07 k13.83

(0) (0) (0) (1) (2) (0) (3)

t 1.07 2.58 0.43 1.11 1.18 0 0.87

P NS <0.02 NS NS NS NS NS

1. K-corrected values are used. 2. The number of deviant types (T score > 70) iS shown in parentheses.

140

As the data show, nonsuppressors had lower scores on the Pa scale, but we view this difference as unimportant because the nonsuppressors’mean scores were within the normal range and the distribution of deviant code types did not differ in the two groups. These results suggest that psychopathological terms are not appropriate to explain cortisol nonsuppression on the DST, at least in the case of psychosocial stress. On the basis of the Al’s, anxiety is also not explanatory.

In search of an alternative explanation, we screened the 30 MMPI items that differentiated the high and low postdexamethasone cortisol groups and performed a factor analysis. The four factors that emerged are shown in Table 2.

The high cortisol subjects’ responses to those items listed under Factors I and 2, in our interpretation, demonstrate their denial of aversiveness and weakness. It is theoretically important that under these factors they report fewer interoceptive feelings than subjects with low postdexamethasone cortisol levels. Factors 3 and 4 reflect their intolerance and rigidity. One might characterize high cortisol subjects as persons who are less aware of themselves, are unemotional, and do not enjoy socializing. These attitudes are reflected in their responses to the coping question- naire (Table 3). They indicated denial, passive avoidance, and rejection of social support. The analysis of results on the coping questionnaire was based on a ~2 in which a median split was used to divide the 104 subjects according to their postdexamethasone cortisol level. The median split was used instead of the extreme quartile approach in order to meet the conditions of x2 tests for most items. The distinct items were factor-analyzed as previously.

The DST was readministered I month later, and all but one of the nonsuppressors then suppressed cortisol normally. However, four of the original suppressors failed to suppress plasma cortisol adequately. The 3.5% (51144) nonsuppression rate for the total sample in the l-month DST corresponds to the reported 4% “abnormal” response of unstressed individuals (Rihmer and Arato, 1984).

Discussion

In his overview on the endocrinology of stress, Rose (1984) stresses the “crucial importance of the . . . perception of potentially stressful events as the major variable

in determining . . . adrenal cortical responses.” In Lazarus’ well-elaborated stress concept (Folkman and Lazarus, 1985), the “cognitive appraisal”-in other words, the “evaluative perception”-of the situation is an inherent part of the coping process. According to their own reports, the high-cortisol subjects in our study did not view obligatory military induction as a stressful experience. However, this attitude does not stem from their perception of military service as an opportunity offering them a challenge to perform. Items related to this explanation did not emerge in our analysis of their responses on the coping checklist. On the contrary, they tried to avoid the stress of induction in a passive manner: they denied its

significance. Horowitz (1985) regards denial as one phase in the response to serious life events.

According to him, different persons may progress through this phase at different rates. For most people, the period of denial is very brief. For some, however, the period is prolonged, lasting weeks or months before other responses become

141

Table 2. Factor analysis of Minnesota Multiphasic Personality Inventory (MMPI) items discriminating subjects with high and low postdexametha- sone cortisol levels

High-cortisol subjects’

Loading answer

Factor 1 (VP = 3.5)

1405 No trouble swallowing

1133 Never indulged in unusual sexual practices

1187 No clumsy or awkward hands

1300 Never played with dolls

1528 Blush no more often than others

137 Never been in trouble because of sexual behavior

163 No difficulty with bowel movements

1214 Never had breaking out on skin that was worrisome

168 Hardly ever feel pain in neck

Factor 2 (VP = 2.9)

1511 Have a daydream life

1314 Think of things too bad to talk about

1147 Have often lost out on things

1390 Have often felt badly misunderstood

1303 Touchy on some subjects

115 Think of things too bad to talk about

1277 Entertained by the cleverness of a crook

1108 Have a fullness in head or nose

139 Feel like smashing things

1198 Daydream very little

Factor 3 (VP = 2.0)

II 13 Believe in law enforcement

1146 Have wanderlust and never happy unless roaming

149 Better if almost all laws were thrown away

1289 Disgusted with the law when a criminal is freed

139 Feel like smashing things

Factor 4 (VP = 1.8)

1287 Have very few fears compared to friends

1477 In trouble prefer to take the whole blame

1244 Misunderstood by others

1523 Practically never Mush

137 Never been in trouble because of sexual behavior

1229 Like to belong to clubs or lodges

I51 1 Have a daydream life

1265 Safer to trust nobody

1437 RigM to get around the law

1113 Belietie in law enforcement

0.72 True’

0.70 True2

0.68 True2

0.64 True3

0.61 True2

0.59 True2

0.47 True2

0.43 True2

0.33 True’

0.59 False2

0.54 False3

0.53 False2

0.52 False2

0.51 False2

0.50 False3

0.43 False’

0.39 False2

0.38 False2

-0.37 True’

-0.64 True3

0.58 False*

0.56 False’

-0.53 True2

0.45 False2

0.52 True2

0.45 True3

0.36 True*

0.33 True2

0.32 True2

-0.30 False*

-0.28 False’

0.27 True2

-0.27 False3

0.26 True3

l.P<O.l. 2. p < 0.05. 3. p < 0.01.

142

Table 3. Coping methods of subjects with high postdexamethasone cortisol levels

Denial (VP = 2.3)

Made light of the situation

Refused to get serious about it

Put problems out of mind4

Looked for the silver lining

Went along with fate

Refused to think about it

Went on as if nothing happened

Support seeking (VP = 2.1)

Let feelings out

Talked about feelings

Accepted the situation

Asked for advice

Sought suppoti

Passive avoidance (VP = 1.9)

Sought support4

Used food, drink, tobacco, etc.

Felt that time could make difference

Figured out whom to blame’

Refused to think about it

Slept more

Tried to forget Went along with fate

l.P<O.l. 2. p < 0.05. 3. p < 0.01.

Factor loading

0.68

0.60

0.57

0.55

0.47

0.42

0.36

0.72

0.58

-0.56

0.44

0.28

-0.74

0.64

0.55

0.44

0.34

0.33 0.26 0.23

USed

Yes2

Yes2

Yes2

Yes3

Yes3

Yes’

Yes2

No*

No’

Yes’

Nd

No3

No3

Yes3

Yes3

Yes’

Yes’

Yes2 Yes’ Yes3

4. From McCrae’s Coping Questionnaire. All other items from Lazarus’ Ways of Coping (revised).

apparent. We speculate that those people with a prolonged period of denial are identical to those who have persistent elevated HPA activity as measured by the DST.

The question arises: Is denial a result of high HPA activity, or does the HPA activation derive from the ineffectiveness of defensive behavior as asserted in previous publications (Wolff et al., 1964~1, 19646; Katz et al., 1970; Poe et al., 1970).

The classic study of Wolff et al. (1964~) on the parents of children with leukemia demonstrated that individuals could cope by denying their distress but that this defense was easily challenged. Repeatedly, the parents showed evidence of psychological stress and increased 17-hydroxycorticosteroid excretion. Wolff et al. (1984~) emphasized that the presence of denial, although important, did not account for all the interindividual variability in HPA function. With this caveat, it does seem that denial as a personality trait is associated with high levels of HPA activation, as

reported by Fox (1978) and in our own data. Horowitz (1985) proposes that selective inattention, inability to appreciate the

significance of stimuli, and disavowal of meanings of stimuli are major charac-

143

teristics of the denial state. Inattention is a fundamental element in our own interpretation, elaborated below.

The convergence of neuroanatomical, neurochemical, and neurophysiological data indicates that the pituitary-adrenal system is integrally involved in the recognition of and response to environmental stimuli of adaptive significance (Reus et al., 1985). Recently, it has been suggested that brain noradrenergic neurons, particularly those projecting from the locus ceruleus, are an important component of brain information-processing systems acting at a variety of brain areas to enhance the effects of sensory inputs (Foote et al., 1983). There is growing evidence that neurons in the locus ceruleus are activated by stimuli that command attention. It has been suggested that the dorsal noradrenergic bundle is involved in attention to exteroceptive stimuli whereas the ventral noradrenergic bundle facilitates the shifting

of attention toward interoceptive cues (Iversen, 1984). A picture emerges of the ventral noradrenergic bundle as a system that is activated during stress and has an effect on the relative prominence of aversive stimuli (Willner, 1985).

It is well established that noradrenergic stimuli have an inhibitory effect on HPA function (Ganong, 1977). Thus, one may postulate that hypofunction of the ventral noradrenergic bundle forms the basis of the neglect of aversiveness and the concomitant high postdexamethasone cortisol level reported here. However, there are numerous ways in which adrenocorticotropic hormone (ACTH) and cortisol can affect noradrenergic function. Glucocorticoid receptor sites are found in high density in projections of the dorsal noradrenergic bundle, such as the dentate gyrus of the hippocampus, and may be involved in regulating noradrenergic effects on stimulus processing, perhaps through a shift of signal-to-noise ratio (Robbins, 1984). Glucocorticoids can block improvement in selective attention following dorsal noradrenergic bundle stimulation (McEwen and Micco, 1980). Cortisol’s impair- ment of selective attention has been demonstrated (Kopell et al., 1970).

Although these associations are speculative, the convergence of data indicates that high HPA activity affects noradrenergic functions and may be a causal factor in the observed denial of (i.e., inattention to) a stressful situation in subjects with high cortisol levels. Their relative inability to perceive interoceptive signals may result in low self-attention, low emotionality, and thus low empathy, and might explain their rigidity, intolerance, and low sociability. In contrast to previous authors, we presume that high HPA activity is not merely a result of unsuccessful psychological defense but also serves as a defense per se.

In summary, we propose the following: (1) High HPA activity plays a psychophysiological role in perceptual defense to promote selective inattention toward the stressor; (2) denial as a complex tactic has some roots in HPA hyperactivity; (3) this kind of psychobiological defense is common in diverse psycho- pathological states (e.g., mania) and is likely overused and broken down in depression (Sachar, 1967; Powell and Hemsley, 1984).

Arat6, M., Rihmer, Z., BBnki, C.M., and Grof, P. evaluation 170 (1983).

144

Brown, W.A., Arato, M., and Shrivastava, R. Pituitary-adrenocortical hyperfunction and intolerance to fluvoxamine, a selective serotonin uptake inhibitor. American Journal of Psychiatry, 143,88 (I 986).

Caine, E.D., Yerevanian, B.I., and Bamford, K.A. Cognitive function and the dexa- methasone suppression test in depression. American Journal of Psychiatry, 141, I 16 (1984).

Carroll, B.J. Dexamethasone suppression test: A review of contemporary confusion. Journal of Clinical Psychiatry, 46, I3 (I 985).

Ceulemans, D.L.S., Westenberg, H.G.M., and van Praag, H.M. The effect of stress on the dexamethasone suppression test. Psychiatry Research, 14, I89 (1985).

Cookson, J.C. The neuroendocrinology of mania. Journal of Affective Disorders, 8, 233 (1985).

Faludi, G., Kask6, M., Perenyi, A., Arato, M., and Frecska, E. The dexamethasone suppression test in panic disorder and major depressive episodes. Biological Psychiatry, 21, 1008 (1986).

Folkman, S., and Lazarus, R.S. If it changes, it must be a process: Study of emotion and coping during three stages of a college examination. Journal of Personality and Social Psychology. 48, I50 ( 1985).

Foote, S.L., Bloom, F.E., and Aston-Jones, G. Nucleus locus coeruleus: New evidence of anatomical and psychological specificity. Physiological Reviews, 63, 844 (1983).

Fox, B.H. Premorbid psychological factors as related to cancer incidence. Journal of Behavioral Medicine, 1,45 (1978).

Ganong, W.F. Neurotransmitters involved in ACTH secretion: Catecholamines. Annals of the New York Academy of Sciences, 291,509 (1977).

Haier, R.J. Pain sensitivity, evoked potentials, and the dexamethasone suppression test in depressed patients. Psychiatry Research, 10,201 (1983).

Horowitz, M.J. Disasters and psychological responses to stress. Psychiatric Annals, 15, I61 ( 1985).

Iversen, S.D. Cortical monoamines and behavior. In: Descarries, L., Reader, T., and Jasper, H.H., eds. Monoamine Innervation of Cerebral Cortex. Alan R. Liss, Inc., New York ( 1984).

Katz, J.L., Weiner, H., Gallagher, T.F., and Hellman, L. Stress, distress, and ego defenses. Archives of General Psychiatry, 23, I3 I (I 970).

Kopell, B.S., Wittner, W.K., Lunde, D., Warrick, G., and Edwards, D. Cortisol effects on averaged evoked potential, alpha rhythm, time estimation, and two-flash fusion threshold. Psychosomatic Medicine. 32, 39 ( 1970).

Krishnan, K.R.R., France, R.D., Pelton, S., McCann, U.D., Manepalli, A.N., and Davidson, J.R.T. What does the dexamethasone suppression test identify? Biological Psychiatry, 20,957 (1985).

McCrae, R.R. Situational determinants of coping responses: Loss, threat, and challenge. Journal of Personality and Social Psychology, 46,9 I9 ( 1984).

McEwen, B.F., and Micco, B.J. Toward an understanding of the multiplicity of the glucocorticoid action on brain function and behavior. In: De Wied, D., and van Keep, P., eds. Hormones and the Brain. University Park Press, Baltimore (1980).

Murphy, B.E.P. Some studies of the protein-binding of steroids and their application to the routine micro and ultramicro measurement of various steroids in body fluids by competitive protein-binding radioassay. Journal of Clinical Endocrinology and Metabolism, 27, 973 ( 1967).

Poe, R.O., Rose, R.M., and Mason, J.W. Multiple determinants of l7-hydroxy- corticosteroid excretion in recruits during basic training. Psychosomatic Medicine, 32, 369 (I 970).

Powell, M., and Hemsley, D.R. Depression: A breakdown of perceptual defence? British Journal of Psychiatry, 145,358 (I 984).

Req V.I., Peeke, H.V.S., and Miner, C. Habituation and cortisol dysregulation in depression. Biological Psychiatry, 20,980 (1985).

145

Rihmer, Z., and Aratb, M. The DST as a clinical aid and research tool in patients with affective disorders. Psychopharmacology Bulletin, 20, I74 (1984).

Robbins, T. W. Cortical noradrenaline, attention and arousal. Psychosomatic Medicine, 14, 13 (1984).

Rose, R.M. Overview of endocrinology of stress. In: Brown, G.M., Koslow, S.H., and Reichlin, S., eds. Neuroendocrinology and Psychiatric Disorder. Raven Press, New York (I 984).

Roy-Byrne, P.P., and Uhde, T.W. Panic disorder and major depression: Biological relationships. Psychopharmacology Bulletin, 21,55 I (1985).

Sachar, E.J. Corticosteroids in depressive illness: II. A longitudinal psychoendocrine study. Archives of General Psychiatry, 17,554 (1967).

Schweizer, E.E., Swenson, C.M., Winokur, A., Rickels, K., and Maislin, G. The dexamethasone suppression test in generalised anxiety disorder. British Journal of Psychiatry, 149,320 (I 986).

Willner, P. Depression: A Psychobiological Synthesis. John Wiley & Sons, Inc., New York (1985).

Wolff, C.T., Friedman, S.B., Hofer, M.A., and Mason, J.W. Relationship between psychological defenses and mean urinary I7-hydroxycorticosteroid excretion rates: I. A predictive study of parents of fatally ill children. Psychosomatic Medicine, 26, 576 (1964a).

Wolff, C.T., Hofer, M.A., and Mason, J.W. Relationship between psychological defenses and mean urinary I7-hydroxycorticosteroid excretion rates: II. Methodologic and theoretical considerations. Psychosomatic Medicine, 26,592 (19646).

Zimmerman, M., Coryell, W., and Corenthal, C. Attribution style, the dexamethasone suppression test, and the diagnosis of melancholia in depressed inpatients. Journal of Abnormal Psychology, 93,373 (I 984).