Psychosomatic Medicine Tips for Better Browsing
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Powell, L. H.
Right arrow Articles by Ory, M. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Powell, L. H.
Right arrow Articles by Ory, M. G.
Related Collections
Right arrow Aging
Right arrow Sexual Medicine: Female
Right arrow Stress and Coping
Psychosomatic Medicine 64:502-509 (2002)
© 2002 American Psychosomatic Society

ORIGINAL ARTICLES

Physiologic Markers of Chronic Stress in Premenopausal, Middle-Aged Women

Lynda H. Powell, PhD, William R. Lovallo, PhD, Karen A. Matthews, PhD, Peter Meyer, PhD, A. Rees Midgley, MD, Andrew Baum, PhD, Arthur A. Stone, PhD, Lynn Underwood, PhD, Judith J. McCann, DNSc, RN, Kristi Janikula Herro, MFA and Marcia G. Ory, PhD, MPH

From the Department of Preventive Medicine (L.H.P., P.M.) and the Institute For Healthy Aging (J.J.M., K.J.H.), Rush-Presbyterian-St. Luke’s Medical Center, Chicago, Illinois; the VA Medical Center and Department of Psychiatry and Behavioral Sciences (W.R.L.), University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma; the Department of Psychiatry (K.A.M.) and Psychology (A.B.), University of Pittsburgh, Pittsburgh, Pennsylvania; the Reproductive Sciences Program (A.R.M.), University of Michigan, Ann Arbor, Michigan; the Department of Psychiatry (A.A.S.), State University of New York at Stony Brook; The Fetzer Institute (L.U.), Kalamazoo, Michigan; and Behavioral and Social Science Research Program (M.G.O.), National Institute on Aging, National Institutes of Health, Bethesda, Maryland.

Address reprint requests to: Lynda H. Powell, PhD, Department of Preventive Medicine, Rush-Presbyterian-St. Luke’s Medical Center, 1700 W. Van Buren St., Suite 470, Chicago, IL 60612. Email: lpowell{at}rush.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 
OBJECTIVE: The purpose of this study was to identify physiological markers of chronic stress in middle-aged women that can be assessed simply and are thus feasible for introduction into large-scale, epidemiologic studies of aging.

METHODS: Subjects were 40 nonsmoking, premenopausal women between the ages of 42 and 52 years, 20 of whom were chronically stressed because of undergoing a divorce or separation and 20 of whom were nonstressed because of being in stable marriages. Stressed and nonstressed women were matched for age, ethnicity, and education. Hypotheses focused on morning and evening salivary cortisol, overnight urinary catecholamines, cortisol, and testosterone, and platelet catecholamines.

RESULTS: Relative to the nonstressed control subjects, the stressed women had elevated evening (9 PM) salivary cortisols, a finding that was observed on both days (mixed effects model: effect = 0.44; se = 0.14, p = .003). Support for the importance of the HPA axis was provided by the observation that the stressed women had less suppression of salivary cortisol in response to low-dose dexamethasone. Contrary to our hypothesis that stressed women would have lower overnight urinary testosterone, they had higher testosterone on day 2 (stressed = 0.76 ng/mg, nonstressed = 0.55 ng/mg; p = .04). Post hoc repeated measures analysis revealed a significant group effect over all time periods of observation (F = 5.48, p = .03, df = 1,18). Stressed women had a nonsignificant trend toward elevated platelet catecholamines. No association was found for overnight urinary catecholamines or cortisol.

CONCLUSIONS: Promising markers of marital upheaval in middle-aged women are evening salivary cortisol and urinary testosterone from a first morning void. Replication of these findings with the same and different chronic stressors and with women of older ages is needed. The low cost and minimal burden of these potential markers makes it feasible to introduce them into large-scale epidemiologic studies of health in aging women.

Key Words: cortisol, • catecholamines, • testosterone, • divorce, • chronic stress, • women’s health, • aging.

Abbreviations: HPA = hypothalamic-pituitary-adrenocortical;; HPG = hypothalamic-pituitary-gonadal;; IES = Impact of Events Scale;; PSS = Perceived Stress Scale.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 
Chronic stressors persist over time, challenge one’s ability to cope, and are believed to have physiologic sequelae that translate into chronic disease (1). Large-scale epidemiologic investigations offer great potential to examine physiological links between chronic stress and the development of chronic disease, but most fail to include potential physiologic markers because it is unclear which ones to use and the assessment protocols are too burdensome.

The difficulty in selecting potential stress markers stems from a large literature characterized by diversity in the markers examined (eg, cortisol from the HPA axis and catecholamines from the sympathetic nervous system), the modes of assessing them (eg, plasma, urine, and saliva), the times of sampling (eg, morning, evening, overnight, and 24-hours), and the nature of chronic stressor studied (eg, hostages, race car drivers, air traffic controllers, and unemployment). Given such diversity in design, it is not surprising that results are inconsistent. Among the markers that have been identified are elevated 24-hour cortisol (2, 3), suppression of 24-hour cortisol (4), elevated evening cortisol (2, 5), elevated overnight cortisol (6, 7), elevated morning cortisol (8, 9), elevated 24-hour catecholamines (2), elevated overnight catecholamines (6, 7), and elevated morning catecholamines (2).

Most of these studies were conducted on men. It is difficult to generalize these results to women because women are exposed to more stress (10), report more symptoms of stress (10, 11), have different perceptions of stress (11), and have different patterns of neuroendocrine reactivity to stress (12, 13). Studies that have focused on women have used gender-specific stressors (eg, bereavement, full-time work with a small child at home, caregiving, and job strain) to test the hypothesis that chronic stress is associated with altered diurnal rhythm in cortisol secretion (3, 14–19). Four of these seven studies supported this hypothesis by finding that cortisol, assessed in the urine or saliva over the course of a day, was elevated in stressed women over nonstressed control subjects (3, 15, 16, 19).

Although this supports the importance of stress-induced alterations in diurnal variation of cortisol in women, the burden of repeated collection of saliva or urine throughout the day generally precludes such measures from use in large-scale epidemiologic studies. But it does raise the question of whether there are specific time points during the day at which differences between stressed and nonstressed women are maximized. There is evidence to suggest that the biggest differences are at the beginning (17, 18) or at the end (14) of the day. If this is true, then targeted sampling at those times could serve as a low-cost marker of altered diurnal variation.

Evidence for the value of catecholamines as markers of chronic stress in women is based primarily on data from the Three Mile Island study (6). In a sample that was 66% women, elevations in norepinephrine and, to a lesser extent, epinephrine were observed in a 15-hour overnight urine sample. Catecholamines in the platelets, which reflect sympathetic activity over past days to weeks and are less influenced by acute fluctuations (20), have not been investigated as a potential marker of chronic stress in women.

The androgen testosterone could be a marker of chronic stress in middle-aged women. In animals under acute stress, testosterone is lower than in control subjects and remains lower for as long as 18 hours after the removal of the acute stressor (21). In men who experience such diverse stressors as being a medical resident, driving a heavy vehicle, and having work-related mental strain, lower testosterone levels were observed than in matched nonstressed control subjects (22–24). Alternatively, in a study of male former hostages, testosterone assessed at 11:00 PM was elevated relative to nonstressed control subjects (2). Although there have been no studies of the role of testosterone as a marker of chronic stress in women, it could play an increasingly important role for aging women because it is secreted by the menopausal ovary (25).

To summarize, cortisol, the catecholamines, and testosterone could all serve as physiologic markers of chronic stress in women. Logistically feasible assessment protocols include targeted salivary sampling, overnight urine collection, and/or blood sampling. The purpose of this study was to determine whether one or more of these physiologic measures, assessed in a logistically feasible protocol, was associated with chronic stress in premenopausal, middle-aged women. The specific hypotheses tested were that chronically stressed women, compared with nonstressed controls, would have in the saliva 1) elevated cortisol in the morning, 2) elevated cortisol in the evening; in the overnight urines 3) elevated cortisol, 4) elevated catecholamines, 5) lowered testosterone; and in a blood sample 6) elevated platelet catecholamines.


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 
Subjects
Participants were 40 working women, between the ages of 42 and 52, who were premenopausal by virtue of having regular menstrual cycles. They were recruited through advertisements requesting participation of nonsmoking women who were employed full-time and in either stable, secure marital relationships or anticipating or undergoing a divorce or separation.

The design called for 20 women who were experiencing the marital stressor and 20 matched, nonstressed control subjects who were in stable marital relationships. The marital stressor was defined as anticipating a separation or divorce, or being separated or divorced within the last 2 months, from a partner in a relationship that had lasted at least 2 years. This chronic stressor was chosen because past studies have shown that women, relative to men, have heightened sensitivity to marital-related stressors (26) and that unhappiness in marriage is associated with poor health and adverse physiology (27, 28). Anticipation of a separation or divorce was operationalized as having stated this intention to someone in the social network (eg, partner, sister, and mother) or to a professional (eg, health care provider and lawyer). To ensure that the marital stressor was actually triggering distress, two additional eligibility criteria were used. First, evidence of intrusive thoughts about the stressor had to be documented using the Impact of Events Scale (IES) (29). Second, there had to be evidence of the experience of global stress using the Perceived Stress Scale (PSS) (30). Nonstressed women were defined as women who were in a stable marital relationship for the last 2 years, and who had the lowest possible score on both the IES and the PSS.

The 2 groups were matched for years of education (±1 year), ethnicity, and age (±2 years) to provide control for these potential confounders. Exclusion criteria were 1) current smoking, 2) being neither high nor low on chronic stress, 3) being perimenopausal or postmenopausal, 4) having an acute life event in last 2 weeks that was not related to the marital problems, 5) not having a full-time day job and normal nighttime sleep schedule, 6) a prior history of irregular menstrual cycles or childbearing problems, 7) sickness in the last 2 weeks, 8) any physical condition or medications that would influence blood pressure or cortisol, 9) thyroid disorders, and 10) hospitalization for, or history of, psychiatric disorders.

Protocol
Potential participants called the study center and were screened by telephone to determine eligibility. Those who were eligible were invited to attend a clinic screening.

The clinic screening included a review of the protocol, obtaining informed consent on a consent form approved by the Rush Institutional Review Board, verification of inclusion and exclusion criteria by interview, a brief medical history, assessment of blood pressure, height, and weight, completion of a battery of psychosocial measures, and a 35-ml blood draw. The psychosocial measures included the Center for Epidemiologic Studies-Depression (CES-D) Scale (31), the cynicism subscale of the Cook-Medley Hostility Scale (32), the emotional and instrumental support subscales of the Rand Social Support Scale (33), a life events scale that was specially designed to include events that were germane to the lives of midlife women, and the Ladder of Life measure of quality of life (34). The subject was then asked to call the study center on the first day of her menstrual bleeding.

The protocol began on a Monday or Tuesday and on day 2 to 7 of the beginning of the menstrual bleeding. Women were discouraged from engaging in physical activity and drinking caffeine and alcohol. On day 1, a complete 24-hour urine sample was collected into four separate containers marked by time intervals: wake to noon (discarding first morning void); noon to 6 PM; 6 PM to bedtime; and overnight (after bedtime and at wake-up). Eight saliva samples were collected in response to preprogrammed signals from a watch at wake-up, 20 minutes after wake-up, noon, 2 PM, 4 PM, 6 PM, 9 PM, and bedtime. Data were entered into diaries at the time of saliva collections to provide information about comparability of groups on sleep quality, physical activity, caffeine, alcohol, and acute stressors.

Because of our particular interest in morning and evening samples, the protocol continued on day 2 but focused on the critical evening and morning time periods. Among the collections were two morning saliva samples (wake-up and 20 minutes after wake-up), three evening saliva samples (6 PM, 9 PM, and bedtime), an evening urine sample (6 PM to bedtime), and an overnight urine sample.

On day 3, before bedtime, subjects ingested a 0.5-mg dexamethasone tablet, a drug that suppresses cortisol production in healthy subjects. Day 4 involved three saliva samples that were under the influence of dexamethasone taken the night before. The samples were taken at wake-up, 20 minutes after wake-up, and noon.

Urine was collected in prelabeled bottles, and subjects were provided with a cooler and a block of frozen blue ice within which to store it. Saliva was collected using commercially available salivettes (Sarstedt, Hanover, NJ) and stored at room temperature (35). On completion of the protocol, subjects were paid $400 for their participation.

Hormone Assays
Free cortisol was assayed from saliva centrifuged to obtain a clear specimen of low viscosity. Cortisol levels were then quantified by radioimmunoassay with time-resolved fluorometric detection, as described elsewhere (36). The same assay was used for measuring urinary free cortisol after a 1:50 dilution of the specimen with distilled water. Inter- and intra-assay coefficients of variation were <10% and <8%, respectively.

Urinary catecholamines were assayed after solid phase extraction using a modified commercial kit (Bio-Rad, Hercules, CA). Catecholamines were eluted from the solid phase with ammonium pentaborate and quantified using high-performance liquid chromatography with electrochemical detection (37). Intra-assay coefficients of reliability were 0.98 for both epinephrine and norepinephrine. Platelet catecholamines were measured after isolation of platelets from plasma. Platelets were then sonicated and centrifuged to yield a clear supernatant, and catecholamines were extracted with alumina followed by acid elution. Quantification was carried out by high-performance liquid chromatography and electrochemical detection, using a procedure modified from Strobel et al. (38). Intra-assay coefficients of reliability were 0.89 for epinephrine and norepinephrine.

Urinary total testosterone was assayed using a commercially available radioimmunoassay kit (Coat-A-Count, Diagnostics Products Corporation, Los Angeles, CA). The method shows a high degree of specificity, with less than 1% crossreactivity except for 19-nortestosterone (20%), 11-ketotestosterone (16%), and 5-alpha-dihydrotestosterone (3%). Inter- and intra-assay coefficients of variation were 12% (39).

Analyses
Data analyses began with a comparison of subjects on eligibility characteristics, matching variables, cardiovascular risk factors, and psychosocial risk factors using chi-squared or paired t tests, as appropriate. Statistical comparisons between stressed and nonstressed women on all cortisol and catecholamine dependent variables were made after transformation using the natural logarithm to eliminate the skewedness of the distributions. This made it possible to use statistical tests that rely on normally distributed data.

Comparisons of salivary cortisol levels over time between matched stressed and nonstressed women were tested using a mixed effects model, with the difference in the logged values of the salivary cortisols of the matched pairs as outcome and time of day as the predictor with five data points (wake-up, wake-up + 20 minutes, 6 PM, 9 PM, and bedtime) pooled over 2 days of observation. This model tests whether there is an overall time-related difference between paired observations, pooled over 2 days, and, if so, where that difference lies. Comparisons between groups at single points in time were made using two-tailed paired groups t tests. Stepwise backwards elimination linear regression was used to examine the relative impact of chronic stress group vs. other potential confounders on evening salivary cortisol.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 
Baseline Characteristics
Table 1 presents the baseline characteristics of the 20 stressed and the 20 matched nonstressed women. The eligibility criteria and matching strategy resulted in differences between the groups on the marital stressor and perceived stress measures and comparability between them on age, ethnicity, and education. The groups did not differ on body mass index or systolic blood pressure, but the stressed group had a 5 mm Hg higher diastolic blood pressure than did the nonstressed group (p < .05). The psychosocial profile of the stressed group was worse than that of the nonstressed group. The stressed women were more depressed (p < .001) and had less instrumental support (p < .001), more major life events (p < .001), and a lower quality of life (p < .01).


View this table:
[in this window]
[in a new window]
  		Table 1. Baseline Characteristics
 
Salivary Cortisol Hypotheses
Table 2 presents the results of the cortisol hypotheses. To test the hypotheses that stressed women have an elevated morning or evening salivary cortisol, mixed effects models were fit using unstructured variance/covariance matrices. Though the time effect was significant overall (p = .002), only the 9 PM differences were significantly different from 0 (9 PM effect = 0.44, se = 0.14, p = .003). There was no time effect for morning salivary cortisols. Table 2 reveals that there was considerably more variability in the morning samples than that observed in the evening samples. The test-retest reliability of salivary cortisols assessed between day 1 and day 2 were significant (9 PM rxy = 0.56, wake-up rxy = 0.63, both p < .01). Figure 1 presents a graph of the salivary cortisols after transformation at all time points of collection, accompanied by paired-groups t tests for individual comparisons. On day 1, the most frequent collection period, the stressed women had slightly elevated cortisols throughout the day, but the differences were maximized in the evening.


View this table:
[in this window]
[in a new window]
  		Table 2. Cortisol Assessments
 


View larger version (16K):
[in this window]
[in a new window]
  		Fig. 1. Salivary cortisol over 2 days and in response to a 0.5-mg dexamethasone suppression test in stressed and nonstressed women. w = Wake-up time; w + 20 = 20 minutes after wake-up; 12 = noon; 2 = 2 PM; 4 = 4 PM; 6 = 6 PM; 9 = 9 PM; b = bedtime. Data were transformed with the natural logarithm.

 
To confirm the importance of cortisol as a stress marker, we examined salivary cortisol response to low-dose dexamethasone. Table 2 and Figure 1 show that at 20 minutes after wake-up and at noon, stressed women were significantly less suppressed than nonstressed women. Comparing the groups with themselves before (day 1) and after dexamethasone at the noon assessment, stressed women had a 31% reduction in salivary cortisol compared with the nonstressed women, who had a 56% reduction.

There were no differences between the stressed and nonstressed women in urinary free cortisol assessed in an overnight collection on day 1 or day 2 (Table 2). There were no correlations between 9 PM salivary cortisol and overnight urinary cortisol on day 1 (rxy = .12, NS) or on day 2 (rxy = .05, NS).

Catecholamine Hypotheses
Table 3 presents the results of the catecholamine hypotheses. The hypothesis that chronic stress would be associated with elevated overnight urinary catecholamines was not supported either for epinephrine or for norepinephrine on day 1 or day 2. The hypothesis that chronic stress would be associated with elevated platelet catecholamines was not supported by conventional standards, but marginally significant associations were evident. Platelet epinephrine and norepinephrine were significantly correlated with 9 PM salivary cortisols on day 1 (platelet epinephrine rxy = .463, p < .01; platelet norepinephrine rxy = .329, p < .05) and, for norepinephrine only, on day 2 (platelet epinephrine rxy = .244, NS; platelet norepinephrine rxy = .342, p < .05).


View this table:
[in this window]
[in a new window]
  		Table 3. Catecholamine Assessments
 
Testosterone Hypothesis
Table 4 presents the testosterone results. The hypothesis that chronically stressed women would have a lower overnight urinary testosterone than nonstressed women was not supported. Instead, stressed women had higher overnight testosterone on day 2 only. Although the day 1 data show the same mean difference, the increased variability on day 1 precluded significance. The test-retest reliability of overnight testosterone assessed on day 1 and day 2 was 0.71 (p < .01). The overnight testosterone levels were significantly correlated with the 9 PM salivary cortisols (day 1: rxy = .34; day 2: rxy = .36; both p < .05).


View this table:
[in this window]
[in a new window]
  		Table 4. Overnight Urinary Free Testosterone Assessments (ng/mg)
 
Because of the unexpected direction of the association between testosterone and stress, we examined the data for each time point at which urine was collected (Figure 2). Repeated measures analysis revealed a significant group effect (F = 5.48, p = .03, df = 1,18) and time effect (F = 4.83, p = .03) and a nonsignificant group by time effect (F = 0.92, p = .42). Comparisons at individual time points showed significant differences for all except the day 1 evening and overnight samples (Figure 2).



View larger version (17K):
[in this window]
[in a new window]
  		Fig. 2. Urinary testosterone over 2 days in stressed and nonstressed women.

 
Multivariate Analysis
Baseline comparisons of the stressed and nonstressed groups revealed that the chronically stressed women were more depressed. Because depression has been associated with elevated cortisol (2, 40), it is possible that it was the depression, rather than the chronic marital stressor, that accounted for the results. Moreover, differences between the groups on the experience of acute stress on the protocol day could confound observed associations for chronic stress. We conducted a stepwise backward elimination multivariate analysis in which chronic stress group, depression, and, from the day 1 diary, overnight sleep quality, rating of anxiety/tension, and rating of anger/irritability were candidates for exclusion in a model predicting 9PM salivary cortisol. The only significant variables to remain after backward elimination were chronic stress group (standardized beta = 0.42, p = .009) and overnight sleep quality (standardized beta = 0.40, p = .01). This suggests that it was the chronic stressor, rather than depression or acute stress, that accounted for the elevation of 9 PM salivary cortisol. A second stepwise model was run that accounted for the above variables but, in addition, for physical activity and caffeine intake from the diary during the protocol day. In this model, all variables were eliminated except for the chronic stress group, which was marginally significant (standardized beta = 0.28, p = .07). The first model better accounted for the variability in 9 PM salivary cortisol than the second model (first model R2 = 0.22; second model R2 = 0.08). Thus, chronic stress group and overnight sleep quality were the best set of predictors of 9 PM salivary cortisol.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
ACKNOWLEDGMENTS
 REFERENCES
 
Because the hypothesis that chronic stress alters the diurnal variation in cortisol secretion is generally supported by the literature on physiological markers of chronic stress in women (3, 15, 11, 19), an aim of this investigation was to determine the easiest and least burdensome way to capture this diurnal variation. Although salivary cortisol was mildly elevated in stressed women throughout the day, it was only in the evening that these differences were significant. This finding was observed both on day 1 and day 2 and was supported by the results of the dexamethasone suppression test. Moreover, the effect size was substantial: over one-half of a standard deviation difference between groups on both day 1 and day 2. Samples of salivary cortisol are simple to obtain and store and are inexpensive to assay. This overcomes some of the most common barriers against introduction of such measures into large-scale epidemiological studies.

Studies of the physiological concomitants of unemployment (18) and job strain (17) found the biggest differences in salivary cortisol in the morning. It is possible that the salience of the stressor influences the timing of maximum discrimination. In the case of job strain and unemployment, the stressor may be most salient in the morning when women face going to the stressful job or the absence of a job. In the case of divorce or separation, the stressor may be most salient for working women when they face an empty home at the end of the day where once there had been a spouse.

It is not clear why the stress group difference in evening salivary cortisol was not also observed in the overnight urinary cortisol. Failure to find correspondence between salivary and urinary cortisols has been observed by others (2, 3) and may be related to the characteristics of saliva-free cortisol and urinary-free cortisol and/or to the disproportionate influence of the preawakening surge in cortisol secretion that would be reflected in the overnight collection. This surge is heavily determined by diurnal influences and may have been minimally affected by the chronic stressor (41).

An unexpected but potentially seminal finding was the robust elevation in urinary testosterone in the stressed women. This implicates the hypothalamic-pituitary-gonadal axis as a marker of chronic stress in middle-aged women. We hypothesized that testosterone would be lowered in chronically stressed women because we expected that divorce or separation would be perceived by the women as a failure and because past studies have linked depressed testosterone to the failure side of a "mastery-failure" dimension (42–44). In finding the opposite effect, our first inclination was to suspect some type of a coding error. We checked for this by reanalyzing a small number of selected samples and got the same result, thus ruling out this explanation.

It is possible that the women in our sample who were undergoing marital upheaval were not perceiving themselves to be failures. Among the psychological characteristics that have been associated with elevations in testosterone are elation at being freed in newly-released hostages (2), dominance, persistence, combativeness, and focused attention (45), repressed anger (46), and arousal and tension (47). In absence of a direct assessment of the women’s feelings about their divorce, we can only speculate that they may have been experiencing any of these. In men, the years surrounding divorce have been associated with elevations in testosterone (48, 49). Unfortunately, because most studies of chronic stress and testosterone have been conducted in men, and sex differences seem to be evident (47), our understanding of how this mechanism works in women is limited and deserves continued study.

Although we did not find support for any of the catecholamine hypotheses, we believe that continued research on platelet catecholamines should be conducted. In our study, both platelet epinephrine and norepinephrine were marginally significant (p <= .10), they differed between the stressed and nonstressed groups by about one-third of a standard deviation, and they were significantly correlated with 9 PM salivary cortisol. Because catecholamines in the platelets reflect activity over the last few weeks (20), they may be a more reasonable marker of the effects of chronic stress on the sympathetic nervous system than plasma or urinary assessments, which reflect more acute reactivity. Moreover, they are easy to assess using a simple, nonfasting blood draw.

There are several limitations of this study. It was conducted on a very small sample of highly selected women who were homogeneous for age, ethnicity, and menopausal status and who were undergoing a very specific chronic stressor. Thus, the ability of these markers to discriminate in more heterogeneous groups of women who are undergoing a variety of types of chronic stress is not known. The testosterone findings were a post hoc observation that was the opposite of what was expected. Thus, they must be replicated before placing weight on them. The robustness of the effect should encourage such replication.

It would be of considerable value to introduce promising physiological markers of chronic stress into large-scale epidemiologic studies. By so doing, we could create models in which the self-reported experience of chronic stress, its potential physiological pathways, and the development of clinical outcomes can be studied simultaneously. Toward that end, more focused studies such as this one can identify other promising markers and determine whether these results can be replicated. Of particular interest is the hypothesis that the chronic stressor of divorce or separation in middle-aged women is associated with a correlated response involving both the HPA and the HPG axes and, possibly, the chronic stimulation of the sympathetic nervous system. If such a correlated response exists, it is of further interest to determine whether it occurs in older women in response to such common chronic stressors as bereavement and caregiving.


   ACKNOWLEDGMENTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
REFERENCES
 
This study was supported by a grant from the John E. Fetzer Institute.

Received for publication September 7, 2000.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 

  1. Cohen S, Kessler RC, Gordon LU. Strategies for measuring stress in studies of psychiatric and physical disorders. In: Cohen S, Kessler RC, Gordon LU, editors. Measuring stress: a guide for health and social scientists. New York: Oxford University Press; 1995. p. 3–26.
  2. Rahe RH, Karson S, Howard NS, Rubin RT, Poland RE. Psychological and physiological assessments on American hostages freed from captivity in Iran. Psychosom Med 1990; 52: 1–16.[Abstract/Free Full Text]
  3. Luecken LJ, Suarez EC, Kuhn CM, Barefoot JC, Blumenthal JA, Siegler IC, Williams RB. Stress in employed women: impact of marital status and children at home on neurohormone output and home strain. Psychosom Med 1997; 59: 352–9.[Abstract/Free Full Text]
  4. Yehuda R, Southwick SM, Nussbaum G, Wahby V, Giller EL, Mason JW. Low urinary cortisol excretion in patients with posttraumatic stress disorder. J Nerv Ment Dis 1990; 178: 366–9.[Medline]
  5. Lundberg U, Granqvist M, Hansson T, Magnusson M, Wallin L. Psychological and physiological stress responses during repetitive work at an assembly line. Work Stress 1989; 3: 143–53.
  6. Baum A, Schaeffer MA, Lake CR, Fleming R, Collins DL. Psychological and endocrinological correlates of chronic stress at Three Mile Island. In: Williams RB Jr, editor. Perspectives on behavioral medicine. Vol 2: Neuroendocrine control and behavior. Orlando (FL): Academic Press; 1985. p. 201–17.
  7. Hawk LW, Dougall AL, Ursano RJ, Baum A. Urinary catecholamines and cortisol in recent-onset posttraumatic stress disorder after motor vehicle accidents. Psychosom Med 2000; 62: 423–34.[Abstract/Free Full Text]
  8. Coeck C, Jorens PG, Vandevivere J, Mahler C. ACTH and cortisol levels during residency training. N Engl J Med 1991; 325: 738.[Medline]
  9. Theorell T, Perski A, Akerstedt T, Sigala F, Ahlberg-Hulten G, Svensson J, Eneroth P. Changes in job strain in relation to changes in physiological state. Scand J Work Environ Heath 1988; 14: 189–96.
  10. Davis MC, Matthews KA, Twamley EW. Is life more difficult on Mars or Venus? A meta-analytic review of sex differences in major and minor life events. Ann Behav Med 1999; 21: 83–97.
  11. Lundberg U, Mardberg B, Frankenhaeuser M. The total workload of male and female white collar workers as related to age, occupational level, and number of children. Scand J Psychol 1994; 35: 315–27.[Medline]
  12. Stoney CM, Matthews KA, McDonald RH, Johnson CA. Sex differences in lipid, lipoprotein, cardiovascular, and neuroendocrine responses to acute stress. Psychophysiology 1988; 25: 83–7.
  13. Kirschbaum C, Wust S, Hellhammer D. Consistent sex differences in cortisol responses to psychological stress. Psychosom Med 1992; 54: 648–57.[Abstract/Free Full Text]
  14. Adam EK, Gunnar MR. Relationship functioning and home and work demands predict individual differences in diurnal cortisol patterns in women. Psychoneuroendocrinology 2001; 26: 189–208.[Medline]
  15. Turner-Cobb J, Sephton SE, Koopman C, Blake-Mortimer J, Spiegel D. Social support and salivary cortisol in women with metastatic breast cancer. Psychosom Med 2000; 62: 337–45.[Abstract/Free Full Text]
  16. Vedhara K, Cox NKM, Wilcox GK, Perks P, Hunt M, Anderson S, Lightman SL, Shanks NM. Chronic stress in elderly carers of dementia patients and antibody response to influenza vaccination. Lancet 1999; 353: 627–31.[Medline]
  17. Steptoe A, Cropley M, Griffith J, Kirschbaum C. Job strain and anger expression predict early morning elevations in salivary cortisol. Psychosom Med 2000; 62: 286–92.[Abstract/Free Full Text]
  18. Ockenfels MC, Porter L, Smyth J, Kirschbaum C, Hellhammer DH, Stone AA. Effect of chronic stress associated with unemployment on salivary cortisol: overall cortisol levels, diurnal rhythm, and acute stress reactivity. Psychosom Med 1995; 57: 460–7.[Abstract]
  19. Jacobs SC, Mason J, Kosten TR, Kasl SV, Ostfeld AM, Wahby V. Urinary free cortisol and separation anxiety early in the course of bereavement and threatened loss. Biol Psychiatry 1987; 22: 148–52.[Medline]
  20. Chamberlain KG, Pestell RG, Best JD. Platelet catecholamine contents are cumulative indexes of sympathoadrenal activity. Am J Physiol 1990; 259 (2 Pt 1): E141–7.[Abstract/Free Full Text]
  21. Tsuchiya T, Horii I. Different effects of acute and chronic immobilization stress on plasma testosterone levels in male Syrian hamsters. Psychoneuroendocrinology 1995; 20: 95–102.[Medline]
  22. Singer F, Zumoff B. Subnormal serum testosterone levels in male internal medicine residents. Steroids 1992; 57: 86–9.[Medline]
  23. Leo MAJ, Kreuz E, Rose RM, Jennings JR. Suppression of plasma testosterone levels and psychological stress. Arch Gen Psychiatry 1972; 26: 479–82.[Medline]
  24. Cullen J, Fuller R, Dolphin C. Endocrine stress responses of drivers in a ‘real-life’ heavy-goods vehicle driving task. Psychoneuroendocrinology 1979; 4: 107–15.[Medline]
  25. Judd HL, Judd GE, Lucas WE, Yen SSC. Endocrine function of the postmenopausal ovary: concentration of androgens and estrogens in ovarian and peripheral vein blood. J Clin Endocrinol Metab 1974; 39: 1020–3.[Medline]
  26. Ross M, Holmberg D. Recounting the past: gender differences in the recall of events in the history of close relationships. In: Olson JM, Zanna MP, editors. Self-influence processes. Hillsdale (NJ): Erlbaum; 1990. p. 135–52.
  27. Carels RA, Szczepanski R, Blumenthal JA, Sherwood A. Blood pressure reactivity and marital distress in employed women. Psychosom Med 1998; 60: 639–43.[Abstract/Free Full Text]
  28. Burnan B, Margolin G. Analysis of the association between marital relationships and health problems: an interactional perspective. Psychol Bull 1992; 112: 39–63.[Medline]
  29. Horowitz M, Wilner N, Alvarez W. Impact of event scale: a measure of subjective stress. Psychosom Med 1979; 41: 209–18.[Abstract/Free Full Text]
  30. Cohen S, Kamarck T, Mermelstein R. A global measure of perceived stress. J Health Soc Behav 1983; 24: 385–96.[Medline]
  31. Radloff LS. The CES-D scale. A self-report depression scale for research in the general population. Appl Psychol Meas 1977; 1: 385–401.
  32. Barefoot JC, Dodge KA, Peterson BL, Dahlstrom WG, Williams RB Jr. The Cook-Medley hostility scale: item content and ability to predict survival. Psychosom Med 1989; 51: 46–57.[Abstract/Free Full Text]
  33. Sherbourne CD, Stewart AL. The MOS social support survey. Soc Sci Med 1991; 32: 705–14.
  34. Andrews FM, Withey SB. Social indicators of well-being: American’s perceptions of life quality. New York: Plenum; 1976.
  35. Kirschbaum C, Hellhammer D. Salivary cortisol in psychobiological research: an overview. Neuropsychobiology 1989; 22: 150–69.[Medline]
  36. Dressendoerfer RA, Kirschbaum C, Rohde W, Stahl F, Strasburger CJ. Synthesis of a cortisol-biotin conjugate and evaluation as tracer in an immunoassay for salivary cortisol measurement. J Steroid Biochem Molec Biol 1992; 43: 683–92.[Medline]
  37. DeBellis MD, Baum AS, Birmaher B, Keshavan MS, Eccard CH, Boring AM, Jenkins FJ, Ryan ND. A. E. Bennett Research Award. Developmental traumatology. Part I: Biological stress systems. Biol Psychiatry. 1999; 45: 1259–70.[Medline]
  38. Strobel G, Werle E, Helfinger H, Griebel D, Weicker H. Syntheses of the sulfoconjugated isomers of norepinephrine and dopamine, controlled by HPLC with ultraviolet detection. Europ J Biochem 1988; 176: 397–402.[Medline]
  39. Coat-A-Count total testosterone. Los Angeles, CA: Diagnostics Products Corporation; 1999.
  40. vanEck M, Berkhof H, Nicolson N, Sulon J. The effects of perceived stress, traits, mood states, and stressful daily events on salivary cortisol. Psychosom Med. 1996; 58: 447–58.[Abstract/Free Full Text]
  41. Lovallo WR, Thomas TL. Stress hormones in psychophysiological research: emotional, behavioral, and cognitive implications. In: Cacioppo JT, Tassinary LG, Berntson G, editors. Handbook of psychophysiology. 2nd ed. New York: Cambridge University Press; 2000. p. 342–67.
  42. Bernhardt PC, Dabbs JM Jr, Fielden JA, Lutter CD. Testosterone changes during vicarious experiences of winning and losing among fans at sporting events. Physiol Behav 1998; 65: 59–62.[Medline]
  43. Francis KT. The relationships between high and low trait psychological stress, serum testosterone, and serum cortisol. Experientia 1981; 37: 1296–7.[Medline]
  44. Kreuz LE, Rose RM, Jennings JR. Suppression of plasma testosterone levels and psychological stress. Arch Gen Psychiatry 1972; 26: 479–82.
  45. Dabbs JM Jr, Carriere E, Fielden JA. Trial lawyers and testosterone: blue-collar talent in a white-collar world. J Appl Soc Psychol 1998; 28: 84–94.
  46. Mason JW, Giller EL, Ostroff RB. Relationships between psychological mechanisms and the pituitary-gonadal system. Curr Clin Pract Series 1984; 26: 215–28.
  47. Dabbs JM Jr, Strong R, Milun R. Exploring the mind of testosterone: a beeper study. J Res Per 1997; 31: 577–87.
  48. Mazur A, Michalek J. Marriage, divorce, and male testosterone. Social Forces 1998; 77: 315–30.
  49. Booth A, Dabbs JM. Testosterone and men’s marriages. Social Forces 1993; 72: 463–77.



This article has been cited by other articles:


Home page
 J. Clin. Endocrinol. Metab.Home page
M. J. H. J. Dekker, J. W. Koper, M. O. van Aken, H. A. P. Pols, A. Hofman, F. H. de Jong, C. Kirschbaum, J. C. M. Witteman, S. W. J. Lamberts, and H. Tiemeier
Salivary Cortisol Is Related to Atherosclerosis of Carotid Arteries
J. Clin. Endocrinol. Metab., October 1, 2008; 93(10): 3741 - 3747.
[Abstract] [Full Text] [PDF]

Home page
 Pers Soc Psychol RevHome page
D. A. Sbarra and C. Hazan
Coregulation, Dysregulation, Self-Regulation: An Integrative Analysis and Empirical Agenda for Understanding Adult Attachment, Separation, Loss, and Recovery
Personality and Social Psychology Review, May 1, 2008; 12(2): 141 - 167.
[Abstract] [PDF]

Home page
 Biol. Reprod.Home page
V.J. Grant, R.J. Irwin, N.T. Standley, A.N. Shelling, and L.W. Chamley
Sex of Bovine Embryos May Be Related to Mothers' Preovulatory Follicular Testosterone
Biol Reprod, May 1, 2008; 78(5): 812 - 815.
[Abstract] [Full Text] [PDF]

Home page
 Psychosom. Med.Home page
D. A. Glei, N. Goldman, Y.-L. Chuang, and M. Weinstein
Do Chronic Stressors Lead to Physiological Dysregulation? Testing the Theory of Allostatic Load
Psychosom Med, October 1, 2007; 69(8): 769 - 776.
[Abstract] [Full Text] [PDF]

Home page
 Hispanic Journal of Behavioral SciencesHome page
D. L. Mangold, R. Veraza, L. Kinkler, and N. A. Kinney
Neuroticism Predicts Acculturative Stress in Mexican American College Students
Hispanic Journal of Behavioral Sciences, August 1, 2007; 29(3): 366 - 383.
[Abstract] [PDF]

Home page
 Psychosom. Med.Home page
K. Matthews, J. Schwartz, S. Cohen, and T. Seeman
Diurnal Cortisol Decline is Related to Coronary Calcification: CARDIA Study.
Psychosom Med, September 1, 2006; 68(5): 657 - 661.
[Abstract] [Full Text] [PDF]

Home page
 Psychosom. Med.Home page
A. Steptoe, L. Brydon, and S. Kunz-Ebrecht
Changes in Financial Strain Over Three Years, Ambulatory Blood Pressure, and Cortisol Responses to Awakening
Psychosom Med, March 1, 2005; 67(2): 281 - 287.
[Abstract] [Full Text] [PDF]

Home page
 Psychosom. Med.Home page
M. G. Ory and M. Chesney
Aging and the Life-Course: Advancing Psychosomatic Medicine Research
Psychosom Med, May 1, 2002; 64(3): 367 - 369.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Powell, L. H.
Right arrow Articles by Ory, M. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Powell, L. H.
Right arrow Articles by Ory, M. G.
Related Collections
Right arrow Aging
Right arrow Sexual Medicine: Female
Right arrow Stress and Coping

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS