Psychosomatic Medicine Faster Service from Outside North America
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 Turner-Cobb, J. M.
Right arrow Articles by Spiegel, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Turner-Cobb, J. M.
Right arrow Articles by Spiegel, D.
Related Collections
Right arrow Cancer
Right arrow Social Support
Psychosomatic Medicine 62:337-345 (2000)
© 2000 American Psychosomatic Society

ORIGINAL ARTICLES

Social Support and Salivary Cortisol in Women With Metastatic Breast Cancer

Julie M. Turner-Cobb, PhD, Sandra E. Sephton, PhD, Cheryl Koopman, PhD, Jane Blake-Mortimer, PhD and David Spiegel, MD

From the Department of Psychiatry and Behavioral Sciences (J.M.T.-C., S.E.S., C.K., J.B.-M., D.S.), Stanford University School of Medicine, Stanford, CA; and Department of Psychology (J.B.-M.), University of Adelaide, Adelaide, South Australia, Australia. Current affiliations: Centre for Research in Health Behaviour (J.M.T.-C.), Department of Psychology, University of Kent at Canterbury, Canterbury, Kent, United Kingdom; and Department of Psychiatry (S.E.S.), University of Louisville School of Medicine, Louisville, KY.

Address reprint requests to: Julie M. Turner-Cobb, Centre for Research in Health Behaviour, Department of Psychology, University of Kent at Canterbury, Canterbury, Kent CT2 7LZ, United Kingdom. Email: J.M.Turner-Cobb{at}ukc.ac.uk


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 NEUROENDOCRINE RESPONSES...
 METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 
OBJECTIVE: This study used a cross-sectional design to examine the relationships between social support, both quantity (number of people) and quality (appraisal, belonging, tangible, and self-esteem), and neuroendocrine function (mean and slope of diurnal salivary cortisol) among women with metastatic breast cancer.

METHODS: Participants (

N = 103) were drawn from a study (N = 125) of the effects of group therapy on emotional adjustment and health in women with metastatic breast cancer. They completed the Interpersonal Support Evaluation List and the Yale Social Support Index and provided saliva samples for assessment of diurnal cortisol levels on each of 3 consecutive days. Diurnal mean levels were calculated using log-transformed cortisol concentrations, and the slope of diurnal cortisol variation was calculated by regression of log-transformed cortisol concentrations on sample collection time.

RESULTS: Mean salivary cortisol was negatively related to the Interpersonal Support Evaluation List subscales of appraisal, belonging, and tangible social support. No association was found between quantitative support or the esteem subscale of the Interpersonal Support Evaluation List and mean salivary cortisol. Measures of qualitative and quantitative social support were not associated with the diurnal cortisol slope.

CONCLUSIONS: Results show that greater quality of social support is associated with lower cortisol concentrations in women with metastatic breast cancer, which is indicative of healthier neuroendocrine functioning. These results may have clinical implications in the treatment of breast cancer.

Key Words: social support • cortisol • breast cancer

Abbreviations: HPA = hypothalamic-pituitary-adrenal (axis); ISEL =Interpersonal Support Evaluation List.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
NEUROENDOCRINE RESPONSES...
 METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 
Social support has been described as providing both a generalized beneficial effect and a buffering effect on health (for review, see Ref. 1). As a buffer between stress and illness, social support likely operates to eliminate or reduce the physiological reaction to stress by modulating stress appraisal or enabling reappraisal (1, 2). The importance of supportive social relations in respect to health has been demonstrated by a number of large-scale studies (for reviews, see Refs. 3–5). In a major review, House et al. (4) demonstrated that the magnitude of the relationship between social integration and age-adjusted mortality is comparable to the effect of such standard health behavior risk factors as smoking and serum cholesterol levels. Intervention studies have demonstrated that enhancing social support results in increased length of life as well as quality of life in cancer patients (6–11), although not all studies show this effect (12, 13).

Evidence linking social support to health has indicated the involvement of specific physiological mechanisms (3, 14). In general, greater social support has been associated with the upregulation of various immune parameters, but research investigating the association between social support and neuroendocrine functioning has been less conclusive (for review, see Ref. 5). Effects of both acute and chronic life stress on neuroendocrine function are well established (15–18). However, the current literature does not give a clearcut picture of the effects of social support on neuroendocrine function or activity of the HPA axis (for review, see Ref. 5). Cortisol, the end product of HPA activation, is a reliable physiological stress measure (19, 20). The relationship between different components of social support and cortisol has received comparatively little attention and thus remains largely unresolved (5). So far, this research has been conducted in naturalistic settings, experimental conditions, and intervention trials.

Three studies have examined cortisol under naturalistic settings. Frequency of emotional support and instrumental support has been found to be negatively related to urinary cortisol (in samples collected over a 12-hour period) as well as other neuroendocrine measures in older men but not in older women (21). In a study of pregnant women, both perceived social support and pregnancy support were found to be negatively related to serum cortisol levels (22). Conversely, a naturalistic study of preterm mothers on postpartum day 5 found no association between social support and cortisol sampled from breast milk (23).

Investigators from one experimental study reported finding no association between social support and urinary cortisol in a group of 12-year-old boys (24). However, in a study of healthy adults, gender-specific effects of social support were found to be associated with salivary cortisol responses to a public-speaking stressor (25). Unlike men, women showed no response decrement under "stranger support" conditions and showed a tendency toward increased cortisol responses when accompanied by their boyfriends.

Two studies found no difference in serum cortisol levels between older men and women (26) or between employed and unemployed women (27) receiving a psychosocial intervention. However, two other studies demonstrated an intervention effect of social support on cortisol level. Early stage breast cancer patients receiving group therapy were found to have a significant decrease in plasma cortisol (11). In a study of human immunodeficiency virus–seropositive and –seronegative homosexual men, plasma cortisol levels decreased significantly for those receiving social support from a bereavement intervention group as compared with control subjects (28).

The study of social support and cortisol has lacked clarity because of a number of methodological problems, as noted by Uchino et al. (5). These include not only type of cortisol sampled and time of day that the sample was obtained but also type of support, individual differences, and the need to account for confounding factors, such as medication. Studies that have found a significant association between social support and cortisol are the ones that take at least some of these factors into consideration, particularly sampling time and controlling for medications appropriate to the population under investigation (21, 25, 28).

Furthermore, relationships and social support may deteriorate under stressful conditions. Some may turn away from those under stress because of their own inability to cope with the situation. Even if the amount of support available remains the same, stressful events may negatively influence the amount of perceived support. Prior as well as current stressors may influence the perception of support. Koopman et al. (29) found that social support may shield women with advanced breast cancer from the effects of previous life stress on their emotional adjustment.


   NEUROENDOCRINE RESPONSES SPECIFIC TO CANCER PATIENTS
TOP
 ABSTRACT
 INTRODUCTION
 NEUROENDOCRINE RESPONSES...
METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 
In healthy individuals, cortisol levels are usually highest before awakening and decrease over the course of the day (30). However, in up to 70% of patients with breast cancer, this rhythm seems to be flattened, levels are consistently high, or erratic fluctuations occur (31, 32). Cortisol rhythms are more disrupted in patients with more advanced disease (32, 33). The cause of circadian dysregulation in cancer is undetermined. However, aberrations of cortisol rhythms have been linked both with the physical stress of cancer (34) and with psychological stress (35).

A recent study from our laboratory found that flattened diurnal salivary cortisol rhythms predict early mortality from metastatic breast cancer (36). Representations of "steep" vs. "flat" diurnal cortisol patterns in the current sample are reported separately (36). We found cortisol rhythms to be associated with prognostic indicators as well, including more disseminated metastases, sleep disturbance, and lower number of natural killer cells. Furthermore, flatter diurnal cortisol profiles were marginally associated with marital status, providing some evidence that one severe social stress, disruption of marital ties through separation, divorce, or widowhood, may be associated with loss of normal diurnal cortisol variation (36). Thus, data from our laboratory suggest that the diurnal cortisol profile may be not only a prognostic indicator but also a physiological indicator of accumulated psychosocial stress.

Among other studies that have linked psychosocial stress with cortisol profiles, aberrant diurnal rhythms are evident among those with depression (35), unemployment (37), and posttraumatic stress disorder (38). Furthermore, physically healthy individuals with altered diurnal cortisol profiles have been characterized as chronically stressed (39, 40). Thus, the stresses of having breast cancer (eg, anxiety about diagnosis and prognosis, taxing treatments, and disruption in social, vocational, and family functioning) may disrupt normal circadian cortisol rhythms, an effect potentially buffered by social support.

In the present study, we used a cross-sectional design to examine relationships between both quantity and quality of social support and neuroendocrine function, as measured by diurnal salivary cortisol levels, among women with metastatic breast cancer. We predicted that better quality and quantity of social support would be associated with more normal endocrine function. In particular, we hypothesized that 1) larger social networks and greater quality of support (reflecting appraisal, belonging, tangibility, and self-esteem) would be associated with lower mean salivary cortisol levels and 2) larger social networks and greater quality of support would be associated with a steeper diurnal slope of salivary cortisol.


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 NEUROENDOCRINE RESPONSES...
 METHODS
RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 
Subjects
Subjects were drawn from a sample of 125 women with documented metastatic breast carcinoma who were recruited into a randomized, prospective study of the effects of supportive/expressive group psychotherapy on cancer survival, designed to replicate and extend prior work (6). Women were considered eligible for the study if they had a confirmed diagnoses of metastatic disease, had a physician’s Karnofsky performance rating of 70% or more, had adequate proficiency in English to complete the questionnaires and to participate in an English-speaking therapy group, and lived within the San Francisco Bay area. Characteristics of the sample, including demographics and medical status, are shown in Table 1.


View this table:
[in this window]
[in a new window]
  		Table 1. Sample Characteristics (N = 103)
 
One-hundred four of the 125 subjects provided salivary cortisol samples. Twenty-one subjects were excluded because they were taking hydrocortisone-based medications (N = 10), did not provide saliva samples of adequate volume or number for analysis of diurnal cortisol profiles (N = 8), or reported feeling too ill to collect any saliva samples (N = 3). One subject did not provide data for the ISEL. Thus, analyses were conducted on the 103 subjects who provided full cortisol and ISEL data. The sample of 103 did not differ significantly from the 21 subjects excluded from the analysis on the demographic, medical, and social support variables listed in Table 1.

Psychosocial Measures
Patients completed questionnaires at baseline (before being randomly assigned to the psychosocial treatment or educational control group) to assess demographic and psychosocial variables. Questionnaires were administered as follows.

Social support.
Yale Social Support Index.

For the purposes of this analysis, we used two items from the Yale Social Support Index (41–43) that were specifically designed to assess the size of the individual’s social network: 1) "In general, apart from your children, how many relatives do you have that you feel close to (people you feel at ease with, can talk to about private matters, and can call on for help)?" 2) "In general, how many close friends do you have (people you feel at ease with, can talk to about private matters, and can call on for help)?" Our research focus was assessment of the number of significant people in the women’s lives. Thus, this subscale has excellent face validity. The Yale Social Support Index has been used extensively to assess the quantity, availability, and perceived quality of an individual’s social network and has been found to predict physical performance and mortality in the elderly (21, 43, 44).

Interpersonal Support Evaluation List.
To assess the quality of the respondents’ social support, we used the ISEL (45, 46), which consists of a list of 40 statements regarding the perceived availability and quality of potential social support. In addition to providing an overall score, it has four subscales that measure quality of social support received: 1) appraisal support, which assesses the perceived availability of confidants to talk to about one’s difficulties; 2) belonging support, which examines the availability of people one can do things with; 3) tangible support, which refers to the availability of practical or instrumental help; and 4) self-esteem support, which refers to the presence of others with whom one feels he or she compares favorably. Confirmation of this factor structure and of using both the subscales and overall score has been reported (47). The scale was scored using the four-alternative revised version of the ISEL, which was developed by the authors of the original scale (45, 46) and is viewed by them as a more sensitive instrument than the original version (S. Cohen, personal communication, 1990). This scoring system uses a four-choice response ("definitely true," scored as 3; "probably true," scored as 2; "probably false," scored as 1; or "definitely false," scored as 0) rather than the original two-choice response ("true" or "false"). Thus, a higher number indicates higher support. Although no psychometric data on this revised version have been published, its content is very close to the original, which has good internal consistency for the general population (46). Ranges for the original subscales are 0.70 to 0.82 for appraisal, 0.62 to 0.73 for self-esteem, 0.73 to 0.78 for belonging, and 0.73 to 0.81 for tangible support (46). Test reliability for the original individual subscales ranges from 0.63 to 0.69 for a 6-week interval (46).

Neuroendocrine Measures
Salivary cortisol.
Diurnal salivary cortisol profiles were assessed at baseline. Patients collected their own saliva at home using Salivette devices, which consist of a small cotton swab fitted into a plastic holder that rests inside a centrifuge tube. Patients collected samples at 8 AM, 12 PM, 5 PM, and 9 PM on 3 consecutive days and completed a questionnaire to assess medications, sleep, diet, exercise, and stressors during the sampling period. After delivery to the laboratory, samples were centrifuged, and aliquots of the clear supernatant were frozen at -70°C until they were assayed for cortisol.

Laboratory methods.
Salivary cortisol levels were assessed by means of a 125I radioimmunoassay. Intraassay coefficients of variation of 5.8%, 4.8%, and 5.3% for 10-fold determinations of three different saliva pools (1.34, 0.49, and 0.24 µg/dl, respectively) were obtained. The interassay coefficient of variation was less than 10%. The sensitivity of the assay, calculated from four assays in which 40-fold replications of zero-cortisol tubes were performed, was 0.008 µg/dl.

Data Reduction and Statistical Analyses
Mean cortisol level was calculated by summing the 12 cortisol values and computing a mean. The slope of diurnal change in baseline salivary cortisol levels was calculated as an estimate of diurnal cortisol variability for each patient. As is typical for data representing physiological markers of stress, our distribution of raw cortisol values was skewed. Because of this, and because the normal diurnal cortisol profile may be approximated by an exponential curve, raw cortisol values were log-transformed before analysis of cortisol mean or variability. The diurnal cortisol slope was calculated by regressing the cortisol values on the time of day that each sample was collected. In this calculation, no distinction was made between the 3 separate days of sample collection; rather, data from all 3 days were used in a single regression. ß, or the slope of this line, was used to depict the diurnal cortisol profile and is referred to hereafter as the diurnal cortisol "slope." Computations for cortisol mean and slope are illustrated in Figure 1. Steeper diurnal slopes indicate cortisol declining, as it normally does, more rapidly over the daytime hours and are represented by lower (negative) slope values. Flatter diurnal patterns are represented by higher slope values and indicate more slowly declining levels, abnormal peaks occurring in the afternoon and evening, or levels that actually increased throughout the day (36).



View larger version (13K):
[in this window]
[in a new window]
  		Fig. 1. Illustration of computations of mean salivary cortisol level and slope of diurnal change in cortisol using log-transformed cortisol values. Top, Values for raw salivary cortisol for each time point across the 3 days of sampling. Bottom, Log-transformed cortisol values. Mean cortisol level was assessed by summing across all 12 points and computing a mean; line represents diurnal variability or slope of the diurnal curve, calculated by regressing cortisol values on the time of day collected.

 
Hence, salivary cortisol measures used were the diurnal mean of log-transformed cortisol and the cortisol slope. Because mean cortisol level and area under the curve were correlated at 0.99, cortisol mean was selected for analysis because this variable included a greater sample size. Cortisol slope was selected because it has been shown to have a significant relation to survival time of breast cancer patients (36). Social support network size was assessed using the Yale Social Support Index, and quality of support was assessed using the four subscales of the ISEL (appraisal, belonging, self-esteem, and tangible support).

Possible associations of mean cortisol level and cortisol slope with current medical treatment, including chemotherapy, radiotherapy, and estrogen-related treatment, were examined using Spearman rank correlations. These were used to examine potential associations between the two indices of cortisol and markers of disease status: age at study entry, time from initial diagnosis to diagnosis of metastases or recurrence, dominant site of metastases or recurrence at study entry (chest wall or regional lymph nodes, viscera, or bone), estrogen receptor status, and Karnofsky performance rating. Only site of recurrence in chest wall or regional lymph nodes was significantly correlated with cortisol indices but had no bearing on the results when included as a covariate. Possible associations of mean cortisol and cortisol slope with the health behaviors of sleep and exercise were also examined using Spearman rank correlations. Sleep was measured by total number of hours usually slept per night and total number of hours slept in the last 5 days. Exercise was measured by the separate variables of number of hours per day usually spent engaging in moderate, hard, or very hard activity and number of hours of moderate, hard, or very hard activity in the last 5 days. No associations were observed between the indices of cortisol and sleep or exercise. Hence, these variables were not included as covariates in the main analyses. Pearson correlations were used to analyze the relationships between variables because the data were normally distributed.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 NEUROENDOCRINE RESPONSES...
 METHODS
 RESULTS
DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 
The mean cortisol level and average diurnal slope for the sample are shown in Figure 2. The proportion of the current sample exhibiting aberrations of the diurnal cortisol profile is reported separately (36).



View larger version (11K):
[in this window]
[in a new window]
  		Fig. 2. Mean cortisol level and average diurnal curve for the sample. The average (and SD) diurnal cortisol slope was = -0.084 (0.059).

 
Mean scores on the social support measures used are shown in Table 1. The Yale network size score reveals means that are within the range of that reported for close social networks with a standard deviation twice that reported (21). Although not a perfect comparison, this reveals good variability in number of supportive ties within our group of cancer patients. After accounting for the four-alternative scoring, the overall score on the ISEL scale was consistent with reported norms for the ISEL in healthy populations (46). Within this sample, the ISEL subscale of self-esteem revealed a lower mean than the subscales of belonging, tangible, or appraisal support. No studies including ISEL norms for patient groups could be found in the literature for comparison. Table 2 presents the Pearson product-moment correlation coefficients between the five indices of social support and mean log-transformed salivary cortisol and the log-transformed diurnal slope of salivary cortisol.


View this table:
[in this window]
[in a new window]
  		Table 2. Correlation Coefficients of Social Support Indices With Diurnal Mean and Diurnal Slope of Log-Transformed Salivary Cortisol Values
 
Relationships of Mean Salivary Cortisol With Social Support Indices
Mean log-transformed salivary cortisol had significant and negative relationships with the appraisal, belonging, and tangible support subscales of the ISEL. These relationships are represented in Figure 3. For descriptive purposes, a median split for each of these three significant subscales is used in the figure. Diurnal mean cortisol levels are shown for patients who scored in the higher vs. the lower half of the sample on each subscale. The more positively the women with metastatic breast cancer rated their appraisal, belonging, and tangible dimensions of social support, the lower their mean salivary cortisol value. No significant relationship was found between mean log-transformed salivary cortisol and the self-esteem subscale of the ISEL or size of social network.



View larger version (45K):
[in this window]
[in a new window]
  		Fig. 3. Diurnal mean cortisol levels were higher among patients who reported less appraisal, belonging, and tangible support. Bars show mean (and SEM) raw salivary cortisol levels of patients split at the median of each subscale.

 
Relationships of Diurnal Slope of Salivary Cortisol With Social Support Indices
Only one medical indicator was found to be significantly correlated with either of the primary indices of salivary cortisol: Women whose dominant site of recurrence was in the chest wall or regional lymph nodes had steeper cortisol slopes, rs = -.20 (p < .05), where the negative direction of the coefficient indicates a steeper slope. We therefore controlled for chest wall in examining the relationships between diurnal slope of salivary cortisol and the social support variables. As shown in Table 2, none of the correlation coefficients of social support with diurnal slope of salivary cortisol were statistically significant.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 NEUROENDOCRINE RESPONSES...
 METHODS
 RESULTS
 DISCUSSION
ACKNOWLEDGMENTS
 REFERENCES
 
In our sample of women with metastatic breast cancer, we found mean salivary cortisol to be significantly and negatively associated with the appraisal, belonging, and tangible support subscales of the ISEL but not with the self-esteem subscale or size of social support network. For diurnal slope of cortisol, no significant associations were found with either quantity or quality of social support. Thus, support for our hypothesis regarding quantity of social support and cortisol slope or mean was not found, but confirmation of our hypothesis regarding quality of social support and mean cortisol was found with the belonging, appraisal, and tangible support subscales. These associations are in agreement with previous findings of the effect of quality of social support on cortisol (21, 22). The lack of a relationship between quantity of support and cortisol in this study and other studies further emphasizes the importance of the quality rather than the quantity of social support. Thus, it would seem that it is the psychological appraisal of support quality that may affect or be affected by neuroendocrine activity. These findings are also consistent with results of studies involving other endocrine measures and of those focusing on immune measures, in which social support is associated with improved immune functioning, which is consistent with lower levels of endogenous corticosteroids, which are potently immunosuppressive (eg, Refs. 48–50; for review, see Ref. 5).

Intrinsic to the understanding of the relationship between social support and cortisol is the conceptualization of social support as having a main or buffering effect. To the extent that cortisol levels in this analysis reflect the physiological outcome of an individual’s experience of breast cancer and availability of appropriate social support, the relationships we observed between higher perceived social support and lower mean cortisol levels suggest a stress-buffering effect of social support. However, given that specific data on stress are not included here, it may be premature to conceptualize our findings in terms of the stress-buffering model of social support. Rather, the empirical data presented provide clearer evidence of a main effect relationship between social support and cortisol. Because having breast cancer inevitably carries substantial emotional, psychological, and physical stresses (for reviews, see Refs. 7 and 51), it is plausible to assume that our data reflect variation in stress. Our clinical experience in conducting support groups for cancer patients indicates that the same dilemma experienced in a social context is less stressful (52, 53), an observation that has been confirmed by others (54). The social process of mutual support can contradict rather than reinforce the content of the concern. Such a sense of belonging is possibly an intrinsically universal need and could be particularly relevant to patients with potentially life-threatening illnesses who have been found to benefit from group therapy interventions (11, 28, 55).

Thus, it is plausible that a greater perceived availability of confidants (appraisal) enables an individual to perceive stressful events as less stressful. Furthermore, those who are able to share pleasurable activities with others (belonging) and who have more tangible support may experience more pleasure or less stress, which would be reflected in lower cortisol levels. This interpretation of the current findings fits with established models of stress appraisal (1, 2). It is possible that social support acts as a buffer of psychosocial stress, thereby protecting individuals under stress, helping them to avoid maladaptive physiological responses.

Our finding of higher mean cortisol levels in patients reporting less social support may have clinical relevance. In several animal studies, stress-related elevation of glucocorticoids has been linked with accelerated tumor growth (56–58), and cortisol has been shown to stimulate human tumor cell growth in vitro. For example, Lointier et al. (59) proposed mechanisms of glucocorticoid effects on tumor growth that include metabolic and immune actions. Cortisol inhibits glucose uptake by normal cells, whereas tumor cells may become resistant to this effect and therefore have a metabolic advantage (57, 58, 60). Because glucocorticoids have well-documented immunosuppressive actions (61), it is plausible that tumor-resistant immune functions may be among those that are suppressed by high levels of cortisol (62–64). Thus, tumor progression may be hastened by higher levels of cortisol in subjects with poorer social support. In addition, a recent reanalysis of the group psychotherapy survival finding referred to above (6) found that patients who had undergone an adrenalectomy and who were therefore receiving nonphysiological replacement levels of cortisol had shorter survival times (67). Thus, artificial administration of cortisol prevented the operation of the necessary negative feedback regulation of cortisol. This provides additional human evidence that high cortisol levels may be associated with cancer progression.

The lack of a relationship between the slope of diurnal log-transformed cortisol and social support emphasizes the difference in these two integrative measures of cortisol. Although higher diurnal mean levels are likely to reflect a more activated stress response, the diurnal cortisol slope is probably indicative of dysregulation in the HPA axis (39, 40).

Several methodological limitations must be considered when interpreting these results. Our research design did not allow us to reach causal conclusions. The cross-sectional research design of this study did not permit us to examine possible sequential relationships between social support and cortisol levels. Furthermore, it is interesting to note that the health behaviors of sleep and exercise were not associated with the two cortisol indices assessed here. It is possible that other aspects of health behavior, such as nutrition, may have contributed to the effects found here, although the influence of nutrition is complex. However, previous research (4, 65, 66) has found the magnitude of supportive relations on mortality to be comparable to and independent of the effect of other serious risk factors. Thus, social support variables similar to those related to cortisol levels in the present study have been found to be associated with poorer survival in other studies. Although differences in health status were associated with social support, the relationship between social support and cortisol held even when we controlled for the salient differences in metastatic spread, making it unlikely that the relationship was due to social response to differences in medical condition. Although possible alternative explanations of the relationship between social support and cortisol may point to the role of health behaviors, clearly this is not a full explanation, as demonstrated with this population of metastatic breast cancer patients. Longitudinal research is needed to clarify further the causal relations between social support and neuroendocrine functioning, accounting for detailed health behaviors. Inclusion of specific stress data and assessment of the interaction between stress and social support using a control group without cancer would provide verification and extension of the interpretation of findings presented here.

These findings are important because they provide the first direct evidence linking social support to salivary cortisol levels in metastatic breast cancer patients. Furthermore, this study contributes substantially to the literature because it examined different aspects of social support. The findings suggest that metastatic breast cancer patients could benefit from an assessment of support availability to help determine those at greatest risk for compromised physiological functioning and those for whom intervention would be most beneficial. The interplay between cortisol and other endocrine parameters is of vital importance in understanding these links. Hence, additional research is needed to determine which aspects of social support and HPA activity are related to better patient adjustment and medical outcome.


   ACKNOWLEDGMENTS
TOP
 ABSTRACT
 INTRODUCTION
 NEUROENDOCRINE RESPONSES...
 METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
REFERENCES
 
This research was supported by Grant MH47226 from the National Institute of Mental Health. Robert M. Sapolsky, PhD, provided laboratory space and consultation for cortisol assays; Robert W. Carlson, MD, consulted on the medical aspects of recruiting and assessing patients; and Daniel P. Stites, MD, consulted on the immunological measures. We appreciate the contributions of Elaine A. Miller, project director; Julie Seplaki, Jane Benson, and Karin Calde, project assistants; Sue Dimiceli, data analyst; Greg Schaal and Mike Schaal, laboratory assistants; and all of the women who participated in this research.

Received for publication October 1, 1998.

Revision received September 27, 1999.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 NEUROENDOCRINE RESPONSES...
 METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 

  1. Cohen S, Wills TA. Stress, social support, and the buffering hypothesis. Psychol Bull 1985; 98: 310–57.[Medline]
  2. Levine S, Coe C, Wiener SG. Psychoendocrinology. In: Brush FR, Levine S, editors. Psychoneuroendocrinology of stress: a psychobiological perspective. San Diego (CA): Academic Press; 1989. p. 341–77.
  3. Cohen S. Psychosocial models of the role of social support in the etiology of physical disease. Health Psychol 1988; 7: 269–97.[Medline]
  4. House JS, Landis KR, Umberson D. Social relationships and health. Science 1988; 241: 540–5.[Abstract/Free Full Text]
  5. Uchino BN, Cacioppo JT, Kiecolt-Glaser JK. The relationship between social support and physiological processes: a review with emphasis on underlying mechanisms and implications for health. Psychol Bull 1996; 119: 488–531.[Medline]
  6. Spiegel D, Bloom JR, Kraemer HC, Gottheil E. Effect of psychosocial treatment on survival of patients with metastatic breast cancer. Lancet 1989; 2: 888–91.[Medline]
  7. Spiegel D. Effects of psychosocial support on patients with metastatic breast cancer. J Psychosoc Oncol 1992; 10: 113–20.
  8. Fawzy FI, Fawzy NW, Hyun CS, Elashoff R, Guthrie D, Fahey JL, Morton DL. Malignant melanoma: effects of an early structured psychiatric intervention, coping, and affective state on recurrence and survival 6 years later. Arch Gen Psychiatry 1993; 50: 681–9.[Abstract]
  9. Richardson JL, Zarnegar Z, Bisno B, Levine A. Psychosocial status at initiation of cancer treatment and survival. J Psychosom Res 1990; 34: 189–201.[Medline]
  10. Van der Pompe G, Antoni M, Visser A, Garssen B. Adjustment to breast cancer: the psychobiological effects of psychosocial interventions. Patient Educ Counseling 1996; 28: 209–19.[Medline]
  11. Van der Pompe G, Duivenvoorden HJ, Antoni MH, Visser A, Heijnen CJ. Effectiveness of a short-term group psychotherapy program on endocrine and immune function in breast cancer patients: an exploratory study. J Psychosom Res 1997; 42: 453–66.[Medline]
  12. Ilnyckyj A, Farber J, Cheang M, Weinerman B. A randomized controlled trial of psychotherapeutic intervention in cancer patients. Ann R Coll Physicians Surg Can 1994; 27: 93–6.
  13. Linn MW, Linn BS, Harris R. Effects of counseling for late stage cancer patients. Cancer 1982; 49: 1048–55.[Medline]
  14. Jemmott JB III, Locke SE. Psychosocial factors, immunologic mediation, and human susceptibility to infectious diseases: how much do we know? Psychol Bull 1984; 95: 78–108.[Medline]
  15. Mason JW. A review of psychoendocrine research on the pituitary-adrenal cortical system. Psychosom Med 1968; 30: 576–607.[Abstract/Free Full Text]
  16. Kirschbaum C, Prussner JC, Stone AA, Federenko I, Gaab J, Lintz D, Schommer N, Hellhammer DH. Persistent high cortisol responses to repeated psychological stress in a subpopulation of healthy men. Psychosom Med 1995; 57: 468–74.[Abstract/Free Full Text]
  17. Kiecolt-Glaser JK, Glaser R, Cacioppo JT, MacCallum RC, Snydersmith M, Kim C, Malarkey WB. Marital conflict in older adults: endocrinological and immunological correlates. Psychosom Med 1997; 59: 339–49.[Abstract/Free Full Text]
  18. Turner-Cobb JM, Steptoe A, Perry L, Axford J. Adjustment in patients with rheumatoid arthritis and their children. J Rheumatol 1998; 25: 565–71.[Medline]
  19. Kirschbaum C, Strasburger C, Jammers W, Hellhammer DH. Cortisol and behavior: adaptation of a radioimmunoassay kit for reliable and inexpensive salivary cortisol determination. Pharmacol Biochem Behav 1989; 34: 747–51.[Medline]
  20. Kirschbaum C, Hellhammer DH. Salivary cortisol in psychoneuroendocrine research: recent developments and applications. Psychoneuroendocrinology 1994; 19: 313–33.[Medline]
  21. Seeman TE, Berkman LF, Blazer D, Rowe JW. Social ties and support and neuroendocrine function: the MacArthur studies of successful aging. Ann Behav Med 1994; 16: 95–106.
  22. Wadhwa PD, Dunkel-Schetter C, Chicz-DeMet A, Porto M, Sandman CA. Prenatal psychosocial factors and the neuroendocrine axis in human pregnancy. Psychosom Med 1996; 58: 432–46.[Abstract/Free Full Text]
  23. Groer MW, Humenick S, Hill PD. Characterizations and psychoneuroimmunologic implications of secretory immunoglobulin A and cortisol in preterm and term breast milk. J Perinat Neonatal Nurs 1994; 7: 42–51.[Medline]
  24. Arnetz BB, Edgren B, Levi L, Otto U. Behavioural and endocrine reactions in boys scoring high on Sennton neurotic scale viewing an exciting and partly violent movie and the importance of social support. Soc Sci Med 1985; 20: 731–6.
  25. Kirschbaum C, Klauer T, Filipp S-H, Hellhammer DH. Sex-specific effects of social support on cortisol and subjective responses to acute psychological stress. Psychosom Med 1995; 57: 23–31.[Abstract/Free Full Text]
  26. Arnetz BB, Theorell T, Levi L, Kallner A, Eneroth P. An experimental study of social isolation of elderly people: psychoendocrine and metabolic effects. Psychosom Med 1983; 45: 395–406.[Abstract/Free Full Text]
  27. Arnetz BB, Wasserman J, Petrini B, Brenner SO, Levi L, Eneroth P, Salovaara H, Hjelm R, Salovaara L, Theorell T, Petterson I. Immune function in unemployed women. Psychosom Med 1987; 49: 3–12.[Abstract/Free Full Text]
  28. Goodkin K, Feaster DJ, Asthana D, Blaney NT, Kumar M, Baldewicz T, Tuttle RS, Maher KJ, Baum MK, Shapshak P, Fletcher MA. A bereavement support group intervention is longitudinally associated with salutary effects on the CD4 cell count and number of physician visits. Clin Diagn Lab Immunol 1998; 5: 382–91.[Abstract]
  29. Koopman C, Hermanson K, Diamond S, Angell K, Spiegel D. Social support, life stress, pain and emotional adjustment to advanced breast cancer. Psychooncology 1998; 7: 101–11.[Medline]
  30. Posener JA, Schildkraut JJ, Samson JA, Schatzberg AF. Diurnal variation of plasma cortisol and homovanillic acid in healthy subjects. Psychoneuroendocrinology 1996; 21: 33–8.[Medline]
  31. van der Pompe G, Antoni MH, Heijnen CJ. Elevated basal cortisol levels and attenuated ACTH and cortisol responses to a behavioral challenge in women with metastatic breast cancer. Psychoneuroendocrinology 1996; 21: 361–74.[Medline]
  32. Touitou Y, Bogdan A, Levi F, Benavides M, Auzeby A. Disruption of the circadian patterns of serum cortisol in breast and ovarian cancer patients: relationships with tumour marker antigens. Br J Cancer 1996; 74: 1248–52.[Medline]
  33. Touitou Y, Levi F, Bogdan A, Benavides M, Bailleul F, Misset JL. Rhythm alteration in patients with metastatic breast cancer and poor prognostic factors. J Cancer Res Clin Oncol 1995; 121: 181–8.[Medline]
  34. Mormont MC, Levi F. Circadian-system alterations during cancer processes: a review. Int J Cancer 1997; 70: 241–7.[Medline]
  35. Deuschle M, Schweiger U, Weber B, Gotthardt U, Korner A, Schmider J, Standhardt H, Lammers CH, Heuser I. Diurnal activity and pulsatility of the hypothalamus-pituitary-adrenal system in male depressed patients and healthy controls. J Clin Endocrinol Metab 1997; 82: 234–8.[Abstract/Free Full Text]
  36. Sephton S, Sapolsky R, Kraemer H, Spiegel D. Early mortality in metastatic breast cancer patients with absent or abnormal diurnal cortisol rhythms. J Natl Cancer Inst. In press.
  37. Ockenfels MC, Porter L, Smyth J, Kirschbaum C, Hellhammer DH. 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]
  38. Yehuda R, Teicher MH, Trestman RL, Levengood RA. Cortisol regulation in posttraumatic stress disorder and major depression: a chronobiological analysis. Biol Psychiatry 1996; 40: 79–88.[Medline]
  39. Chrousos GP, Gold P. A healthy body in a healthy mind - and vice versa - the damaging power of "uncontrollable stress" [editorial]. J Clin Endocrinol Metab 1998; 83: 1842–5.[Free Full Text]
  40. Rosmond R, Dallman MF, Bjorntorp P. Stress-related cortisol secretion in men: relationships with abdominal obesity and endocrine, metabolic and hemodynamic abnormalities. J Clin Endocrinol Metab 1998; 83: 1853–9.[Abstract/Free Full Text]
  41. Seeman TE, Berkman LF. Structural characteristics of social networks and their relationship with social support in the elderly: who provides support. Soc Sci Med 1988; 26: 737–49.
  42. Berkman LF, Berkman CS, Kasl S, Freeman DH, Leo L, Ostfeld AM, Coroni-Huntley J, Brody JA. Depressive symptoms in relation to physical health and functioning in the elderly. Am J Epidemiol 1986; 124: 372–88.[Abstract/Free Full Text]
  43. Seeman TE, Berkman LF, Kohout F, Lacroix A, Glynn R, Blazer D. Intercommunity variations in the association between social ties and mortality in the elderly: a comparative analysis of three communities. Ann Epidemiol 1993; 3: 325–35.[Medline]
  44. Seeman TE, Berkman LF, Charpentier PA, Blazer DG. Behavioral and psychosocial predictors of physical performance: MacArthur studies of successful aging. J Gerontol 1995; 50A: M177–83.
  45. Cohen S, Hoberman HM. Positive events and social supports as buffers of life change stress. J Appl Soc Psychol 1983; 13: 99–125.
  46. Cohen S, Mermelstein R, Kamarck T, Hoberman HM. Measuring the functional components of social support. In: Sarason IG, Sarason BR, editors. Social support: theory, research and application. The Hague, Netherlands: Martinus Nijhoff; 1985. p. 73–94.
  47. Brookings JB, Bolton B. Confirmatory factor analysis of the Interpersonal Support Evaluation List. Am J Community Psychol 1988; 16: 137–47.[Medline]
  48. Lekander M, Furst CJ, Rotstein S, Blomgren H, Fredrikson M. Social support and immune status during and after chemotherapy for breast cancer. Acta Oncol 1996; 35: 31–7.
  49. Kirschbaum C, Pirke KM. The ‘ Trier Social Stress Test’—a tool for investigating psychobiological stress responses in a laboratory setting. Neuropsychobiology 1993; 28: 76–81.[Medline]
  50. Armario A, Marti O, Molina T, de Pablo J, Valdes M. Acute stress markers in humans: response of plasma glucose, cortisol and prolactin to two examinations differing in the anxiety they provoke. Psychoneuroendocrinology 1996; 21: 17–24.[Medline]
  51. Spiegel D. Psychosocial aspects of breast cancer treatment. Semin Oncol 1997; 24: S1-36–47.
  52. Spiegel D, Bloom JR, Yalom I. Group support for patients with metastatic cancer: a randomized outcome study. Arch Gen Psychiatry 1981; 38: 527–33.[Abstract]
  53. Spiegel D. Essentials of psychotherapeutic intervention for cancer patients. Support Care Cancer 1995; 3: 252–6.[Medline]
  54. Taylor SE, Lobel M. Social comparison activity under threat: downward evaluation and upward contacts. Psychol Rev 1989; 96: 569–75.[Medline]
  55. Spiegel D, Bloom JR. Group therapy and hypnosis reduce metastatic breast carcinoma pain. Psychosom Med 1983; 45: 333–9.[Abstract/Free Full Text]
  56. Ben-Eliyahu S, Yirmiya R, Liebeskind JC, Taylor AN, Gale RP. Stress increases metastatic spread of a mammary tumor in rats: evidence for mediation by the immune system. Brain Behav Immun 1991; 5: 193–205.[Medline]
  57. Rowse GL, Weinberg J, Bellward GD, Emerman JT. Endocrine mediation of psychosocial stressor effects on mouse mammary tumor growth. Cancer Lett 1992; 65: 85–93.[Medline]
  58. Sapolsky RM, Donnelly TM. Vulnerability to stress-induced tumor growth increases with age in rats: role of glucocorticoids. Endocrinology 1985; 117: 662–6.[Abstract]
  59. Lointier P, Wildrick DM, Boman BM. The effects of steroid hormones on a human colon cancer cell line in vitro. Anticancer Res 1992; 12: 1327–30.[Medline]
  60. Romero L, Raley-Susman K, Redish DM, Brooke SM, Horner HC, Sapolsky RM. Possible mechanism by which stress accelerates growth of virally-derived tumors. Proc Natl Acad Sci USA 1992; 89: 11084–7.[Abstract/Free Full Text]
  61. Munck A, Guyre P. Glucocorticoids and immune function. In: Ader R, Feten D, Cohen N, editors. Psychoneuroimmunology. San Diego (CA): Academic Press; 1991. p. 447–74.
  62. Bergsma J. Illness, the mind, and the body: cancer and immunology: an introduction. Theor Med 1994; 15: 337–47.[Medline]
  63. Landmann RM, Muller FB, Perini C, Wesp M, Erne P, Buhler FR. Changes of immunoregulatory cells induced by psychological and physical stress: relationship to plasma catecholamines. Clin Exp Immunol 1984; 58: 127–35.[Medline]
  64. McDaniel JS, Musselman DL, Porter MR, Reed DA, Nemeroff CB. Depression in patients with cancer: diagnosis, biology, and treatment. Arch Gen Psychiatry 1995; 52: 89–99.[Abstract]
  65. Goodwin JS, Hunt WC, Key CR, Samet JM. The effect of marital status on stage, treatment, and survival of cancer patients. JAMA 1987; 258: 3125–30.[Abstract]
  66. Reynolds P, Kaplan GA. Social connections and risk for cancer: prospective evidence from the Alameda County Study. Behav Med 1990; 16: 101–10.[Medline]
  67. Kogon MM, Biswas A, Pearl D, Carlson RW, Spiegel D. Effects of medical and psychotherapeutic treatment on the survival of women with metastatic breast carcinoma. Cancer 1997; 80: 225–30.[Medline]



This article has been cited by other articles:


Home page
 CLIN PEDIATRHome page
A. Mathur, L. Duda, and D. M. Kamat
Knowledge and Use of Music Therapy Among Pediatric Practitioners in Michigan
Clinical Pediatrics, March 1, 2008; 47(2): 155 - 159.
[Abstract] [PDF]

Home page
 CLIN PEDIATRHome page
L. Avers, A. Mathur, and D. Kamat
Music Therapy in Pediatrics
Clinical Pediatrics, September 1, 2007; 46(7): 575 - 579.
[Abstract] [PDF]

Home page
 Psychosom. Med.Home page
T. W. McDade, L. C. Hawkley, and J. T. Cacioppo
Psychosocial and Behavioral Predictors of Inflammation in Middle-Aged and Older Adults: The Chicago Health, Aging, and Social Relations Study
Psychosom Med, May 1, 2006; 68(3): 376 - 381.
[Abstract] [Full Text] [PDF]

Home page
 Biol Res NursHome page
M. W. Groer
Differences Between Exclusive Breastfeeders, Formula-Feeders, and Controls: A Study of Stress, Mood, and Endocrine Variables
Biol Res Nurs, October 1, 2005; 7(2): 106 - 117.
[Abstract] [PDF]

Home page
 JCOHome page
P. J. Goodwin, M. Ennis, L. J. Bordeleau, K. I. Pritchard, M. E. Trudeau, J. Koo, and N. Hood
Health-Related Quality of Life and Psychosocial Status in Breast Cancer Prognosis: Analysis of Multiple Variables
J. Clin. Oncol., October 15, 2004; 22(20): 4184 - 4192.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
S. K. Lutgendorf, S. Cole, E. Costanzo, S. Bradley, J. Coffin, S. Jabbari, K. Rainwater, J. M. Ritchie, M. Yang, and A. K. Sood
Stress-Related Mediators Stimulate Vascular Endothelial Growth Factor Secretion by Two Ovarian Cancer Cell Lines
Clin. Cancer Res., October 1, 2003; 9(12): 4514 - 4521.
[Abstract] [Full Text] [PDF]

Home page
 Psychosom. Med.Home page
L. S. Porter, M. Mishel, V. Neelon, M. Belyea, E. Pisano, and M. S. Soo
Cortisol Levels and Responses to Mammography Screening in Breast Cancer Survivors: A Pilot Study
Psychosom Med, September 1, 2003; 65(5): 842 - 848.
[Abstract] [Full Text] [PDF]

Home page
 Toxicol SciHome page
E. M. Friedman and D. A. Lawrence
Environmental Stress Mediates Changes in Neuroimmunological Interactions
Toxicol. Sci., May 1, 2002; 67(1): 4 - 10.
[Abstract] [Full Text] [PDF]

Home page
 Psychosom. Med.Home page
L. H. Powell, W. R. Lovallo, K. A. Matthews, P. Meyer, A. R. Midgley, A. Baum, A. A. Stone, L. Underwood, J. J. McCann, K. Janikula Herro, et al.
Physiologic Markers of Chronic Stress in Premenopausal, Middle-Aged Women
Psychosom Med, May 1, 2002; 64(3): 502 - 509.
[Abstract] [Full Text] [PDF]

Home page
 Pers Soc Psychol RevHome page
L. M. Diamond
Contributions of Psychophysiology to Research on Adult Attachment: Review and Recommendations
Personality and Social Psychology Review, November 1, 2001; 5(4): 276 - 295.
[Abstract] [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 Turner-Cobb, J. M.
Right arrow Articles by Spiegel, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Turner-Cobb, J. M.
Right arrow Articles by Spiegel, D.
Related Collections
Right arrow Cancer
Right arrow Social Support

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH