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
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Alpers, G. W.
Right arrow Articles by Roth, W. T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Alpers, G. W.
Right arrow Articles by Roth, W. T.
Related Collections
Right arrow Neuroendocrine
Right arrow Neuropsychology
Right arrow Other Psychiatric Disorders
Right arrow Stress and Coping
Psychosomatic Medicine 65:679-687 (2003)
© 2003 American Psychosomatic Society

ORIGINAL ARTICLES

Salivary Cortisol Response During Exposure Treatment in Driving Phobics

Georg W. Alpers, PhD, James L. Abelson, MD, PhD, Frank H. Wilhelm, PhD and Walton T. Roth, MD

From Stanford University School of Medicine, and VAPA Health Care System (G.W.A., F.H.W., W.T.R.), Palo Alto, California; Julius-Maximilians Universität (G.W.A.), Würzburg, Germany; and the University of Michigan (J.L.A.), Ann Arbor, Michigan.

Address reprint requests to: Georg W. Alpers, PhD, Biological and Clinical Psychology, Universität Würzburg, Marcusstrasse 9-11, D-97070 Würzburg, Germany. Email: alpers{at}psychologie.uni-wuerzburg.de

Received for publication January 4, 2002; revision received September 10, 2002.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 
OBJECTIVE: Extensive research on the hypothalamic-pituitary-adrenal (HPA) axis response to stress has not clarified whether that axis is activated by phobic anxiety. We addressed this issue by measuring cortisol in situational phobics during exposure treatment.

METHODS: Salivary cortisol was measured in 11 driving phobics before and during three exposure sessions involving driving on crowded limited-access highways and compared with levels measured in 13 healthy controls before and during two sessions of driving on the same highways. For each subject, data collected in the same time period on a comparison nondriving day served as an individual baseline from which cortisol response scores were calculated.

RESULTS: Cortisol levels of driving phobics and controls did not differ on the comparison day. Phobics also had normal cortisol response scores on awakening on the mornings of the exposures but these were already increased 1 hour before coming to the treatment sessions. Phobics had significantly greater cortisol response scores during driving exposure and during quiet sitting periods before and afterward. These greater responses generally paralleled increases in self-reported anxiety. At the first exposure session, effect sizes for differences in cortisol response scores between the two groups were large. Initial exposure to driving in the first session evoked the largest responses.

CONCLUSION: The data demonstrate that the HPA axis can be strongly activated by exposure to, and anticipation of, a phobic situation.

Key Words: salivary cortisol, • driving phobia, • in vivo exposure, • anxiety disorders, • hormones, • behavior therapy.

Abbreviations: ANOVA = analysis of variance;; BDI = Beck Depression Inventory;; BMI = body mass index (kg/m2);; DP = driving phobia;; GG = Greenhouse-Geisser correction;; HPA = hypothalamic-pituitary-adrenal;; PTSD = posttraumatic stress disorder;; SCL-90-R = Symptom Check List, revised version;; STAI = State Trait Anxiety Inventory.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 
There is general agreement that acute psychological stress in humans leads to a cascade of hormonal changes regulated by the hypothalamic-pituitary-adrenal (HPA) axis, an increase in cortisol being the most frequently measured. However, determining exactly what triggers the human HPA axis has been difficult. Recent work suggests that stressors elicit cortisol response only if emotionally laden (1) and that "negative affect" may be the prime activator of this system in humans (2). However, although phobic fear is perhaps one of the most intensely negative of human affects, the literature is not at all clear on whether the emotion elicited when a phobic person is exposed to the object of his or her fear is consistently associated with cortisol release. This research, begun decades ago (3) and focusing on laboratory studies of phobias to small animal, spiders, and insects, has given inconsistent results. In the earliest study, exposing patients to their feared object produced high levels of self-reported anxiety but no cortisol release (3). In a follow-up study, the same group did find some evidence of an increase in plasma cortisol in response to such exposures (4, 5) but the increases were small to moderate, and sometimes absent, even in the presence of intense fear ratings, and there was no consistent link between the intensity of fear and cortisol levels (4, 5). The effect size for a cortisol response was low compared with that of other psychophysiological variables such as heart rate (5). Two studies utilizing photographs rather than direct exposure to phobic cues have been reported. In one, projected pictures of phobic objects elevated urinary cortisol relative to neutral pictures but anxiety and cortisol responses correlated poorly (6). In another, in which pictures of phobic cues were combined with startle probes, no urinary cortisol elevations appeared despite clear increases in subjective stress ratings (7). However, these studies have had several limitations. First, none collected information about cortisol levels before entry into the laboratory situation itself although anticipatory elevations could reduce phobic exposure responses due to feedback inhibition (8). Second, nonspecific responses to the laboratory setting may have raised cortisol levels in comparison conditions. Third, the type of phobias studied was restricted by the fact that most clinically common situational phobias cannot be reproduced in the laboratory. Fourth, the duration of exposure has been fairly limited.

HPA studies of anxiety disorders other than specific phobias have also produced mixed results. One study of social phobia found an exaggerated cortisol response to public speaking in some patients but reductions of cortisol in others, although self-reported anxiety increased in all patients (9). Panic disorder patients have shown elevated cortisol during panic in some studies (10, 11) but not in others (12). Sustained elevations of cortisol levels in panic disorder were sometimes found (13–15) but not always (16). On the other hand, the results from studies of people without clinical phobias under challenge situations have been more consistent. For example, bungee jumping increases both anxiety and salivary cortisol (17). Parachute jumping in inexperienced jumpers elicits a robust cortisol response (18–20). Cognitive challenge under intense social scrutiny is fairly consistent in activating the HPA axis (21–23). These findings suggest that, in nonphobics, situational fear is reliably associated with cortisol elevations. The differences between the clinical and nonclinical samples could be due to biological differences between them but are also consistent with the idea that fearful situations in natural contexts are more potent cortisol releasers than feared objects in laboratory contexts.

If cortisol release is driven by fear, parallel decreases in both might be expected after prolonged or repeated exposure as part of therapy. In early phobic exposure work, the moderate cortisol elevations observed in some patients tended to diminish when the exposure was repeated in a second session (4, 5). On the other hand, extended and successful exposure therapy of two height phobics did not lead to a decline in cortisol response to heights (24). Nonphobic subjects also show reductions in cortisol responses on repeated exposure to a frightening situation. For example, cortisol response in novice parachutists decreased over the course of three jumps (18). In another study of 11 consecutive jumps from a tower, cortisol was elevated only in the first two jumps (19, 20). Repeated exposure to a speech stressor led to diminishing cortisol responses (22, 25). These results suggest that cortisol responses diminish as fear habituates, as is also the case in animals (26–28). One limitation of these studies is a difficulty in distinguishing whether the cortisol reductions with repetition are due to a decrease in the novelty or in the fearfulness of the stimulus, because novelty can also cause increased cortisol responses.

The present study was designed to assess the HPA response to in vivo phobic cue exposure in patients afraid to drive on crowded limited-access highways. We measured free cortisol from saliva, which is less intrusive (29) and easier to obtain under naturalistic conditions (30) than previous methods. This method has successfully detected the cortisol release elicited by a variety of anticipated and current daily stressors (31). We applied it to driving phobia, a specific situational phobia in the natural environment. This phobia has particular advantages for endocrine research because driving does not involve much physical exertion or postural changes, which can influence cortisol (30, 32, 33). In addition, fear of driving has serious implications for people’s lives in developed countries (G. W. Alpers, F. H. Wilhelm, and W. T. Roth, unpublished data, 2003, 34). Exposure therapy is the treatment of choice (35, 36) which, when properly administered, initially generates substantial fear that wanes with repetition of the exposure. Exposure therapy thus affords an ethical way to study a distressing experience in an ecologically valid context that provides direct benefit to the experimental subject and allows endocrine responses to be tracked over time.

In the study described here, we exposed a group of women with driving phobia to fear-provoking driving situations and observed the cortisol response during three repetitions of this task. We hypothesized that driving on limited-access highways would elevate cortisol levels in phobics, relative to their normal circadian levels (37, 38) on a nonexposure day and relative to healthy controls exposed to the same driving situations. In addition, we expected a decrease of this cortisol response in phobics with repeated exposure to the same situation.


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 
Eleven driving phobia (DP) participants collected saliva samples for cortisol assay before, during, and after driving exposure and on a nondriving, comparison day. They were a subgroup of the participants in a multichannel assessment of psychophysiological activation within and across three repeated sessions of exposure in patients with fear of driving (34). Saliva collection was begun when the initial project was already well under way. Collection with cotton swabs is free from pain or stress and provides an accurate estimate of plasma cortisol and, thus, adrenocortical activity (29, 39). Only women were enrolled because they were easier than men to recruit, although ideally we would have tested both sexes, which sometimes differ in endocrine responses to stress (40). The driving situations were chosen to correspond to each patient’s specific fears. All three exposure sessions of individual patients were in the same situation. Comparison day data were collected at the same time of day as the driving exposure data and were used to correct for diurnal variation in baseline hormone levels and for nonfear-related individual variation in baseline cortisol secretion. A comparison group of individually matched nonphobic women was used to control for the effects of aspects of the driving experience not specific to anxiety (eg, heightened vigilance and muscular activity associated with steering and braking). This was important because driving has been found to produce an HPA response in professional bus drivers (41, 42) and normal automobile drivers (as indexed by 11-hydroxycorticosteroid) (43). To control for habituation of environmental or task novelty effects (44) that might occur with repeated monitoring sessions in both fearful and nonfearful drivers (45), we had the 13 healthy controls come back for a second driving session.

Participants
Potential subjects were recruited by newspaper advertisements and were offered free evaluation and limited free treatment. Controls were paid $10 per hour for filling out questionnaires and for the interview session, $15 per hour for participation in the recording sessions, and $15 for saliva sampling on a comparison day. Subjects had to have a driver’s license and access to a registered and insured automobile that they could drive during the recording sessions. Eleven volunteers meeting criteria for a DSM-IV (46) diagnosis of specific phobia established by a telephone screening, questionnaire information (34), and a SCID-interview (47) and 13 controls participated in this study. Controls were selected not to have fear of driving and to match a specific phobic participant for sex and age. Exclusion criteria for all subjects were past or current psychotic disorder; current major depression or dysthymia; current taking of psychoactive or cardiovascularly active medication; current cardiac, neurological, or respiratory disease; hypertension; or poor visual acuity.

The groups did not differ in age (phobics: 48.4 ± 9.4 years; controls: 48.6 ± 8.6 years) or body mass index (phobics: 22.67 ± 3.4; controls: 23.39 ± 2.6) or in a number of descriptive variables listed in Table 1. Severity of general pathology was determined by the SCL-90-R general symptom index (48), which was not very high in patients (0.76 ± 0.46) but significantly higher than in controls (0.17 ± 0.22; F(1,21) = 16.48, p = .001), and by the BDI score (49) which was at the cutoff for mild depression in phobics (10.3 ± 8.0) but significantly higher than in controls (1.9 ± 2.5; F(1,21) = 12.87, p = .002). Patients’ trait anxiety, as measured with the STAI (50), was moderately high (42.3 ± 12.9) and significantly higher than in controls (26.9 ± 5.4; F(1,21) = 15.23, p = .001). Patients’ main target phobia as assessed in the Fear Questionnaire (51) was high in phobics (5.8 ± 2.6) and significantly higher than in controls (0.5 ± 0.6; F(1,21) = 57.95, p < .001). Three of the 10 patients also met diagnostic criteria for panic disorder, and two patients had experienced panic attacks in the past but did not meet full criteria for panic disorder.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Descriptive Variables Related to Cortisol
 
Procedure
At the initial interview patients were told that, although the experience of anxiety can be very unpleasant, it is nonetheless harmless. Their symptoms were explained as being part of a natural "fight-or-flight" reaction. They were given information about the documented efficacy of exposure therapy and the necessary components of successful exposure (no avoidance, no distraction, and prolonged and repeated exposure). No further cognitive interventions were given. Controls were told that their data were needed for a comparison with phobics’ reactions in the same situations. Written informed consent was obtained after the procedures were fully explained.

On a questionnaire, subjects rated their likelihood of being able to drive in a variety of situations on a scale from 0 to 100% (Driving Self-Efficacy Inventory) (52). For the exposure we generally chose a driving situation from the lowest range of self-efficacy ratings and which was located near the hospital. Exposure was generally carried out on 55-km stretches of one of two multilane limited-access highways (a relatively uncrowded freeway through unpopulated areas or a crowded freeway through heavily populated areas). The most difficult driving exposure situation, to which an individual patient would agree, was chosen. Exposure was repeated in the same situation. One patient who participated in this study only avoided bridges, and a long bridge over water (3 km long, 25 m high) was used.

Control subjects followed the same routes as the patients, with whom they were individually matched. Driving was scheduled for matched pairs at the same time of day to control for diurnal fluctuations in traffic. Controls were instructed to drive as they usually did but to be sure to remain under the speed limit (110 km/h) and not to listen to the radio or engage in conversation.

Patients had driving sessions on 3 separate days and control subjects on 2 separate days. The interval from driving session one to two ranged from 2 to 5 (median 3) days in patients and 2 to 9 (median 4) days in controls. The interval ranged from 2 to 10 (median 4) days from session two to three in patients. Patients were told to wait with self-administered exposure until after the last session. On arriving at the laboratory on the days of driving sessions, subjects filled out a questionnaire inquiring about medicines recently taken and the STAI-State questionnaire. Electrodes and transducers were attached, and the subject sat quietly for 7 minutes in a comfortable chair while physiological activity was recorded (quiet sitting). Then the sensors were calibrated and participants drove to their exposure situation. After driving on the freeway for about 30 minutes or driving across a bridge (Leg 1), drivers returned on the same route (Leg 2). Then they drove back to the hospital for another period of quiet sitting and a repeat calibration of the sensors. The experimenter accompanied all participants during the entire driving period. Patients were asked to collect saliva samples on a comparison day between sessions 2 and 3 and controls between sessions 1 and 2.

Saliva Collection
Sampling was done with Salivette cotton swabs in tubes (Sarstedt, Rommelsdorf, Germany). Participants were instructed in the proper use of the swab at the initial interview and allowed to practice sample collection with one tube. They were told to avoid eating or drinking during the 30 minutes before collection (53). Detailed instructions and a log to record sample times and self-report data were handed out. Participants were instructed to collect the first sample on the day of the driving sessions immediately after awakening but before brushing their teeth (the "Awake" sample). The second sample was collected 1 hour before the scheduled driving session (Pre). The next sample was collected after the first quiet sitting period, which lasted 7 minutes (SitA). Exposure samples were collected immediately after each leg of the drive, that is just before turning around (Drive1) and just before returning (Drive2) to the hospital. The last sample was collected after 7 minutes of postexperimental quiet sitting (SitB).

Although salivary cortisol is quite stable at room temperature (39), participants were instructed to keep the collected samples refrigerated until they brought them to the sessions. At the hospital, samples were frozen (-20° C) until they were shipped on dry ice to the assaying laboratory. Anxiety ratings were made immediately after saliva collection for the pre-exposure and quiet sitting samples. During driving ratings were made after 3 minutes of exposure and every 5 minutes thereafter. The average of all ratings of one leg was used to capture anxiety levels during this exposure task. Sample collection on the comparison day followed the same time schedule as on the second driving day. The accuracy of the timing of collection was confirmed by review of subject logs. Self-report data were also recorded in these logs, in conjunction with saliva collection. Participants were instructed to follow their normal daily routine and to take the sampling tubes with them if they were away from home.

Cortisol Analysis
Frozen samples were thawed and centrifuged at 2800 rpm for 3 minutes to provide a yield of about 0.7 ml of saliva. Cortisol was assayed using a direct, nonextraction assay kit (Coat-a-Count kit; Diagnostic Products Corporation, Los Angeles, CA). This is a simple, rapid radioimmunoassay with intra-assay variability of less than 5%.

Anxiety ratings and cortisol levels for a few samples that were forgotten (1% of all data) and levels from samples containing too little saliva to be analyzed (5%) were estimated by linear regression based on preceding data points and coefficients derived from the corresponding group. One phobic driving participant was excluded from the analysis because several of her samples contained too little saliva.

Statistical Analysis
Our principal analysis for cortisol levels and self-reported anxiety was analysis of variance (ANOVA) with repeated measures. To assess if there were time of day or group differences in cortisol levels independent of driving, we compared samples collected during the comparison day in an ANOVA with the factors Group (phobics/controls) and Time (Awake, Pre, SitA, Drive1, Drive2, SitB – these labels refer to the situation from the driving day that corresponded in time to the sample being collected on the comparison day). For the driving days, response scores were calculated as Driving sample minus Comparison Day sample for each situation or time point to reduce the influence of intra-individual differences and to control for circadian variability (38). Response scores of the control subjects were first tested for differences from zero (intercept in ANOVA with the factors, Situations (six levels) and Sessions (two levels)) in order to determine whether driving itself induced a cortisol response in nonphobics.

To compare groups on cortisol response scores, a global ANOVA was done with the factors Group (phobics/controls), Session (1, 2), and Situation (all six repeated measures: Awake, Pre, SitA, Drive1, Drive2, SitB). This analysis included all data available for both phobics and controls. Significant effects were followed up with similar, more specific analyses, separately examining the quiet sitting periods (SitA, SitB), the predriving period (Awake, Pre), and the driving period (Drive1, Drive2) across the two groups and two driving sessions. Finally we did a phobic-session ANOVA to determine whether cortisol and anxiety scores changed in patients over their three exposure sessions. This ANOVA had the factors Session (1–3) and Situation (Awake, Pre, SitA, Drive1, Drive2, SitB). We used the Greenhouse-Geisser correction (GG) for repeated measures analyses where appropriate but always report the original degrees of freedom.

In order to quantify the magnitude of the difference between groups during driving, effect sizes (54) were calculated as d = [phobic mean - control mean]/SD. SD was calculated as the square root of the pooled estimate of population variance [SD2 = N1 * SD12 + N2 * SD22)/(N1 + N2 – 2)].

To obtain a measure of association between intensity of anxiety and size of cortisol response, rank-order correlations were calculated between the cortisol response scores and anxiety response scores of Session 1, where we expected anxiety to be highest in phobics.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 
Cortisol
Comparison Day
On the nondriving day, there were no group differences (F(1,21) = 0.34, p = .567) or significant interactions involving Group (10 phobics and 13 controls), but time points did differ (Time F(5,105) = 23.1; GG p = .001). The highly significant linear (F(1,21) = 55.6, p < .001) and quadratic (F(1,21) = 22.5, p < .001) trends reflected the usual decelerating downward course of cortisol levels over the day (see comparison day panel in Figure 1, lower left). The lack of any patient-control differences on the comparison day supports our use of these data to control for individual and circadian differences in basal activity, by subtracting the comparison day cortisol level from the corresponding time point from the level on the driving day. Subsequent analyses are based on these response scores for the 10 phobics and 13 controls on whom complete data were available. Group means over all times and all sessions are presented in Figure 1.



View larger version (26K):
[in this window]
[in a new window]
  		Fig. 1. Response scores (means and standard errors) of salivary cortisol during experimental days minus comparison day data (raw values depicted for the comparison day). Driving phobics (DP) and healthy controls (HC). Awake = awakening; Pre = 1 hour before; SitA = quiet sitting; Drive1 = driving outbound leg; Drive2 = driving inbound leg; SitB = quiet sitting. Comparison data (nondriving) was collected at the same times as these epochs on the driving days.

 
Driving Days
Before comparing the groups, we confirmed in a separate ANOVA for controls that their cortisol levels during the driving days did not differ significantly from comparison day levels. (Intercept of response scores is not different from zero: F(1,12) = 1.33, p = .27). The global ANOVA showed that the two groups’ cortisol response to driving differed during the first two exposure sessions. There was a significant main effect of Group (F(1,21) = 12.01, p = .002) and a significant Group by Situation interaction (F(5,105) = 3.41, GG p = .018). Components contributing to the significant interaction effect were dissected in follow-up analyses.

We looked first at the predriving cortisol responses (awakening and before leaving home for the hospital) across the two groups and two driving sessions. Main effects of Group and Time were not significant. A significant Group by Time interaction (F(1,21) = 7.74, p = .011) was due to an increase in response scores in phobics from time of awakening to the sample collected 1 hour before arrival at the hospital, relative to controls. Further dissection showed that the groups did not differ in cortisol response on awakening (F(1,21) = 0.87, p = .363). Paired t tests confirmed that response scores were higher for patients 1 hour before the exposure session than on awakening (Session 1: t(9) = -2.54, p = .032; Session 2: t(9) = -2.44, p = .038; and Session 3: t = -4.13, p = .003). This difference was not significant for controls for either of their two sessions. We then examined cortisol responses during quiet sitting (SitA and SitB). The ANOVA showed that patients had higher quiet sitting cortisol response scores, pre- and postexposure, (main effect of Group, F(1,21) = 11.88; p = .002). There was no Time effect – cortisol responses were not different before and after exposure – and there were no significant interactions. Quiet sitting cortisol responses were higher during Session 1 than Session 2 (Figure 1), but this difference did not reach significance (F(1,21) = 3.24, p = .086).

We next examined cortisol response during driving (looking only at samples taken during the two legs of the driving exposure – Drive1 and Drive2). This ANOVA showed that phobics had higher cortisol response scores than controls (Group F(1,21) = 11.91, p = .002), indicating that patients secreted more cortisol during driving, relative to comparison day, than did controls. Effect sizes for group differences in response scores were d = 1.29 for cortisol at Drive1 and d = 1.14 at Drive2 of Session 1. The Session main effect was not significant, but there was a significant Session by Group interaction (F(1,21) = 4.41, p = .048); a Session by Leg effect (F(1,21) = 4.98, p = .037); and a Session by Leg by Group interaction (F(1,21) = 7.97, p = .010). Examination of Figure 1 suggests that all three interactions involve an elevated cortisol response score in patients during Leg 1 (Drive1) of Session 1. This was confirmed by selective follow-up analyses of Leg 1 and Leg 2. At Leg 1 there was a Session by Group interaction (F(1,21) = 5.79, p = .025), which was not present at Leg 2 (Drive2) (F(1,21) = 2.11, p = .162). In patients, response scores decreased from Leg 1 to Leg 2 in Session 1 (t(9) = 2.08, p = .034; one-tailed), a change which was absent in Session 2 and in controls. Also in patients, cortisol declined significantly at Leg 1 from Session 1 to Session 2 (t(9) = 1.92, p = .044; one-tailed), which was not the case for controls.

To assess changes across all three exposure sessions in phobics, a phobic-session ANOVA was conducted. This confirmed that cortisol levels were significantly higher on driving days than on the comparison day in patients (intercept different from zero; F(1,9) = 8.16, p = .019). There were no significant changes in cortisol response scores over the three sessions (Session F(2,18) = 2.37; p = .14). The effect of Situations was significant (F(5,45) = 5.81; GG p = .002), as expected, but the Situation by Session interaction was not, indicating that the significant impacts of anticipation and exposure on cortisol response (seen in Figure 1 as higher presession, quiet sitting, and higher driving responses, relative to awakening responses) were similar in the three sessions.

Because there was a differential anticipation effect (phobics had greater cortisol response scores 1 hour before coming to the exposure session) in the predriving ANOVA described above, we examined the cortisol responses at this point more closely. A separate ANOVA with two factors – Sessions and Time – examined whether anticipation (Pre) and exposure (Drive1) had significantly different effects on cortisol release and if this changed across the three sessions. There were no significant Session, Time, or Interaction effects (all p > .3), suggesting that anticipation and exposure did not have different effects on cortisol. Cortisol response scores to anticipation of driving (mean ± SD, 0.15 ± 0.03) were in fact nearly identical to the cortisol response to driving itself (mean ± SD, 0.14 ± 0.05).

Anxiety
Comparison Day
On the nondriving, comparison day, anxiety ratings were slightly higher in phobics (N = 10) than in controls (N = 13) (F(1,21) = 4.46, p = .047). When Greenhouse-Geisser corrected, the Group by Time interaction missed significance (F(5,105) = 2.48, p = .071). The trend is explained by a marginally significant linear trend for the Group by Time interaction (F(1,21) = 4.11, p = .055) (see comparison day panel in Figure 2, bottom left), with phobics showing a slight rise in anxiety over the day whereas controls showed no change. Subtracting the comparison day anxiety ratings from the corresponding time point from the level on the driving day gives us driving response ratings corrected for the higher nondriving-related anxiety of the phobics. These anxiety response ratings are used in subsequent analyses and are presented in Figure 2.



View larger version (24K):
[in this window]
[in a new window]
  		Fig. 2. Response scores (means and standard errors) of self-reported anxiety during experimental days minus comparison day data (raw values depicted for the comparison day). Awake = awakening; Pre = 1 hour before; SitA = quiet sitting; Drive1 = driving outbound leg; Drive2 = driving inbound leg; SitB = quiet sitting. Comparison data (nondriving) was collected at the same times as these epochs on the driving days.

 
Driving Days
Before the groups were compared, a separate ANOVA for controls alone showed that their anxiety levels during the driving days were only marginally higher than comparison day levels. (Intercept of response scores different from zero: F(1,12) = 3.48, p = .087). The global ANOVA showed that the two groups differed in their anxiety response to driving during the first two exposure sessions. There were significant main effects of Group (F(1,21) = 69.70, p < .001); Session (F(1,21) = 7.29, p = .013); and Situations (F(5,105) = 11.87, GG p < .001) and a significant Group by Situation interaction (F(5,105) = 10.99, GG p < .001). Components contributing to the significant interaction effect were dissected in follow-up analyses.

We looked first at predriving anxiety responses (awakening and before leaving home for the hospital), across the two groups and two driving sessions. A main effect of Group (F(1,21) = 14.61, p = .001) seems to reflect higher anxiety responses in phobics before coming to the hospital for driving exposure (Figure 2). Both a significant main effect of Time (F(1,21) = 6.28, p = .021) and a significant Group by Time interaction (F(1,21) = 10.91, p = .003) seemed to be due to the patients’ anxiety responses increasing from time of awakening in the morning to the assessment 1 hour before coming to the hospital, whereas these scores remained constant in controls. We then examined anxiety responses during quiet sitting (SitA and SitB). This ANOVA showed no main Group effect but the Group by Time interaction was significant (F(1,21) = 30.70, p < .001). Phobics had higher anxiety responses than controls while sitting quietly before driving exposure (confirmed by t tests at this time point for both Sessions 1 and 2: t = 2.88, p = .011 and t = 3.53, p = .005, respectively; equal variance not assumed). Control subjects showed no change from pre- to postexposure quiet sitting anxiety responses, whereas the elevated responses seen in phobics pre-exposure had declined to control subject levels by the postexposure quiet sitting period.

Finally, we examined anxiety response during driving. This ANOVA confirmed that phobics had higher anxiety response scores during driving than controls (main Group effect: (F(1,21) = 42.41, p < .001)) and that subjects’ scores were higher in Session 1 than Session 2 (F(1,21) = 10.34, p = .004). The Group by Session interaction missed significance (F(1,21) = 2.98, p = .099); but anxiety scores collapsed across legs (Drive1, Drive2) were higher in Session 1 than in Session 2 (t(9) = 2.35, p = .027; one-tailed) for patients, whereas this comparison did not reach significance in controls. Effect sizes for group differences in response scores were d = 3.38 for anxiety at Drive1 and d = 2.29 at Drive2 of Session 1.

To assess changes across all three exposure sessions in phobics, a phobic-session ANOVA was conducted. This confirmed the heightened anxiety on driving days relative to the comparison day in the phobics (intercept different from zero: F(1,9) = 57.35, p < .001). A significant main effect of Situations (F(5,45) = 10.08, GG p = .001) reflected rising anxiety in anticipation of coming to the hospital, an anxiety response peak with the first driving leg, and declining anxiety responses during and following exposure (Figure 2). A significant Session effect (F(2,18) = 7.24, GG p = .007) seemed due to decline in all anxiety responses across the three sessions. We were particularly interested in anxiety responses during driving and examined this in an ANOVA using exposure (Drive1 and Drive2) response scores only. There was a significant Session effect (F(2,18) = 12.45, GG p < .001), due to a significant decline from Session 1 to Session 2 (t test reported above) and a further significant decline from Session 2 to Session 3 (t = 2.61, p = .014; one-tailed). Across all three sessions, anxiety response scores were lower during the second leg (Drive2) than the first leg (Drive1) (F(1,9) = 5.77, p = .04).

Correlation of Cortisol and Anxiety
Across all subjects, the anxiety response scores and cortisol response scores of the first driving session were not correlated at Awake (r = 0.03, p = .88, N = 23); SitA (r = 0.21, p = .33); or SitB (r = -0.01, p = .96). They were significantly correlated at Pre (r = 0.42, p = .049); Drive1 (r = 0.66, p = 0.001); and Drive2 (r = 0.59, p = .003). When correlations were calculated for patients only, the relationship between anxiety response and cortisol response did not reach significance at any time point (Awake: r = -0.41, p = .23; Pre: r = -0.22, p = .54; SitA: r = 0.29, p = .42; Drive1: r = 0.40, p = .26; Drive2: r = 0.57, p = . 0.09; SitB: r = -0.35, p = .32).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
ACKNOWLEDGMENTS
 REFERENCES
 
Despite considerable interest in the psychological activators of the HPA axis, historical assumptions that any sufficiently intense stressor will release cortisol (5), and more recent hypotheses about the specific role of negative affect in HPA axis activation (1, 2), only a few studies in the past three decades have examined the relationship between phobic fear and cortisol release in humans. These studies have produced mixed results, leaving it uncertain whether intense fear or anxiety generated by phobic stimuli does in fact activate the human HPA axis. The data reported here, however, are unambiguous in showing that situational phobics in a naturalistic, in vivo exposure paradigm do secrete more cortisol than nonphobic subjects, not only during the actual exposure but also during anticipation of exposure and after exposure.

Phobics and controls did not differ in cortisol levels on a nonexposure comparison day when they were following their daily routines, both groups showing the same expected elevation on awakening and decline over the course of that day. The two groups also had identical levels on awakening on driving exposure days. Thus, we have no evidence that cortisol secretion was chronically dysregulated in our phobics as it sometimes is in other chronic stress conditions like PTSD (55), "burnout" (56), and panic disorder (13–15).

Many of the anxiety results paralleled the cortisol results. Patients and controls awakened on exposure days with near equal anxiety response scores, but 1 hour before leaving home for the clinic, patients had higher scores. Patients’ anxiety remained elevated during the quiet sitting period immediately before exposure but rose even higher during the initial exposure to driving itself. The latter rise contrasts with the time course of cortisol, where actual exposure to the previously highly avoided experience of driving on the highway did not further increase the response beyond levels seen during the anticipatory period. At postexposure quiet sitting, cortisol responses remained elevated whereas anxiety responses had diminished, perhaps because of the persistence of previously secreted cortisol in the blood. The correlational association between cortisol responses and anxiety responses when both patients and controls were included was moderately high only during anticipation and driving, points at which patients and controls differed significantly in both anxiety and cortisol responses. When patients were examined alone, a direct, significant relationship between the intensity of anxiety experienced and size of cortisol response was not significant.

Although there is little doubt that driving phobics had an enhanced cortisol response to driving exposure, evidence for the expected decline in this response over repeated exposure sessions is equivocal. The strongest evidence for a decline comes from an analysis focusing on cortisol during driving exposure on the 2 experimental days, where a significant Session by Group interaction was seen. Post hoc analyses suggested that this interaction was due mainly to the phobics’ high cortisol responses during the outbound (Drive1) leg in Session 1. However, the analysis of patients’ cortisol responses over all three sessions did not document significant changes across sessions. Whereas anxiety responses decreased stepwise over the three sessions, cortisol responses did not. This lack of decrease might indicate slower than normal habituation or desensitization of fear-induced cortisol responses in phobics, because nonphobics exposed to frightening situations have shown declines in response over two or three exposures (19, 22). The elevated cortisol with first exposure in patients (Leg 1, Session 1) may be a novelty effect, which might be particularly strong in patients because they had been avoiding driving in the settings we ultimately picked. However, the animal work documenting that the HPA axis is sensitive to novelty (26–28) also demonstrates that this effect habituates rapidly. In our study, patients’ elevation in cortisol response did not return to "control" levels by Session 2 and did not habituate in a stepwise fashion over the three sessions. It is possible that there was a novelty effect present at Leg 1 of Session 1 for patients, which did habituate, but there must then be an additional cortisol effect associated with having a phobia that does not seem to disappear in concert with desensitization of anxiety ratings. A limitation of our design for examining time changes is that we scheduled fewer exposure sessions than are clinically recommended for a course of behavior therapy. Thus, we could not determine how much cortisol responses would decline with more complete desensitization.

Our findings contradict prior work suggesting that phobic exposure generating substantial distress can occur without cortisol release (4, 8). Perhaps our use of comparison day samples to correct for individual differences and circadian factors, unrelated to exposure, increased sensitivity for detecting fear-related cortisol release. However, other aspects of our paradigm may have contributed to our positive results, such as carrying out the exposure in the "real world" rather than in a laboratory setting or using a situational phobia instead of an animal phobia. Situational phobias may be more akin to contextual cue-related fear, where the cue-threat link is less explicitly perceived and thus the threat is unpredictable. There is evidence that the HPA axis hormone corticotropin-releasing factor may mediate contextually but not explicitly cued anxiety (57). The contextual, rapidly changing, and unpredictable nature of fear cues encountered in a situational driving phobia may involve different neurocircuitry (57) and lead to a more tonic cortisol effect than the more explicit and specific cue-linked fear of a simple animal phobia (58). Lack of control over a stressor or feared stimulus may also contribute significantly to HPA axis activation. Experiencing control reduces the HPA response to challenge in humans (59) and animals (60). In our own study, patients may have experienced a low level of control because they were asked to drive a fixed course in a specified amount of time and because they were in a driving situation from which escape would be difficult.

The salivary sampling technique we used was an improved methodology compared with prior cortisol work in phobias. Most importantly, salivary sampling made it feasible to collect neutral comparison day data to control for nonspecific variation in cortisol levels and to collect anticipatory samples before patients arrived at the laboratory. Salivary sampling allowed us, literally, to take the laboratory to the streets where the "fear" experience was much more real, prolonged, and intense than it is likely to be, for example, when viewing pictures of phobic objects in a laboratory setting (6). Studying driving phobia with our methods allowed us to examine physiological and endocrine activity in an ecologically valid context with minimal interference from physical exertion and postural changes that could confound results. Theoretically, other aspects of driving might elevate cortisol in normal drivers (41–43), but our use of a nonphobic control group allowed us to document phobia-specific elevations.

Our study had several limitations. The sample size was small, entirely female, and included only patients with a single, specific, situational phobia. Not all of our subjects were tested at the same time of day. Subtraction of comparison day cortisol levels (obtained at the same time of day as exposure day samples) served to minimize the influence of baseline circadian rhythm in our response score analyses, but response magnitude might still vary by time of day. Available evidence suggests that cortisol response to phobic exposure may be less in the morning than later in the day (5) so the inclusion of some morning sessions may have reduced the overall effect size for group differences. Because cortisol rises and falls over many minutes (29), the postdriving samples may have captured HPA axis activity during the preceding fear exposure. Samples taken after the relatively short 7-minute quiet sitting periods could partially reflect secretion before sitting, for example, while accommodating for just having entered the laboratory (for the first quiet sitting period) or for driving exposure (for the second). An additional possible limitation of our study stems from the difficulty in separating phobic cue-related anxiety from a more general anxious response tendency, which perhaps manifested itself in the small but significantly higher anxiety levels of phobics on the comparison day. Hopefully, confounds from this general response tendency were minimized by our use of scores that subtracted the comparison day values from the corresponding exposure day values.

In conclusion, we have shown that phobic exposure and its anticipation lead to a cortisol response that can be reliably measured. Given the importance of cortisol and the abundant attention it has received as the "classical" stress hormone, it is surprising that the HPA literature on phobic fear has been so scanty. Because cortisol and the HPA axis may play an important role in mediating the link between psychosocial stress and a wide range of disease states (61), examination of cortisol release in phobias has wider relevance to human health. Assaying salivary cortisol in clinically relevant natural situations offers a useful method for exploring the specific psychology of HPA axis activation.


   ACKNOWLEDGMENTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
REFERENCES
 
This research was supported by the Department of Veterans Affairs, NIH grant MH 56094, the German Academic Exchange Service, and the Graduate Studies Foundation of the State of Hessen, Germany. The authors gratefully acknowledge Elizabeth Young, MD, for her input on the experimental design and the cortisol assays.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 

  1. Nejtek VA. High and low emotion events influence emotional stress perceptions and are associated with salivary cortisol response changes in a consecutive stress paradigm. Psychoneuroendocrinology 2002; 27: 337–52.[CrossRef][Medline]
  2. Lovallo WR, Thomas TL. Stress hormones in psychophysiological research: emotional, behavioral, and cognitive implications. In: Cacioppo JT, Fassmary LG, Bernston GG, editors. Handbook of psychophysiology. Cambridge: Cambridge University Press; 2000.
  3. Curtis G, Buxton M, Lippman D, Nesse R, Wright J. "Flooding in vivo" during the circadian phase of minimal cortisol secretion: anxiety and therapeutic success without adrenal cortical activation. Biol Psychiatry 1976; 11: 101–7.[Medline]
  4. Curtis GC, Nesse R, Buxton M, Lippman D. Anxiety and plasma cortisol at the crest of the circadian cycle: reappraisal of a classical hypothesis. Psychosom Med 1978; 40: 368–78.[Abstract/Free Full Text]
  5. Nesse RM. Endocrine and cardiovascular responses during phobic anxiety. Psychosom Med 1985; 47: 320–32.[Abstract/Free Full Text]
  6. Fredrikson M, Sundin O, Frankenhaeuser M. Cortisol excretion during the defense reaction in humans. Psychosom Med 1985; 47: 313–9.[Abstract/Free Full Text]
  7. Regan MF, Brown K, Howard RC. Cortisol excretion and mood during predictable phobic stimulation. Med Sci Res 1991; 19: 481–3.
  8. Dallman MF, Akana SF, Scribner KA, Bradbury MJ, Walker C-D, Strack AM, Cascio CS. Stress, feedback, and facilitation in the hypothalamic-pituitary-adrenal axis. J Neuroendocrinol 1991; 4: 517–26.[CrossRef]
  9. Furlan P, DeMartinis N, Schweizer E, Rickels K, Lucki I. Abnormal salivary cortisol levels in social phobic patients in response to acute psychological but not physical stress. Biol Psychiatry 2001; 50: 254–9.[CrossRef][Medline]
  10. Bandelow B, Wedekind D, Pauls J, Broocks A, Hajak G, Ruther E. Salivary cortisol in panic attacks. Am J Psychiatry 2000; 157: 454–6.[Abstract/Free Full Text]
  11. Cameron OG, Lee MA, Curtis GC, McCann DS. Endocrine and physiological changes during "spontaneous" panic attacks. Psychoneuroendocrinology 1987; 12: 321–31.[CrossRef][Medline]
  12. Woods SW, Charney DS, McPherson CA, Gradman AH, Heninger GR. Situational panic attacks. Behavioral, physiologic, and biochemical characterization. Arch Gen Psychiatry 1987; 44: 365–75.[Abstract/Free Full Text]
  13. Abelson JL, Curtis GC. Hypothalamic-pituitary-adrenal axis activity in panic disorder. 24-hour secretion of corticotropin and cortisol. Arch Gen Psychiatry 1996; 53: 323–31.[Abstract/Free Full Text]
  14. Bandelow B, Wedekind D, Sandvoss V, Broocks A, Hajak G, Pauls J, Peter H, Ruther E. Diurnal variation of cortisol in panic disorder. Psychiatry Res 2000; 95: 245–50.[CrossRef][Medline]
  15. Goldstein S, Halbreich U, Asnis G, Endicott J, Alvir J. The hypothalamic-pituitary-adrenal system in panic disorder. Am J Psychiatry 1987; 144: 1320–3.[Abstract/Free Full Text]
  16. Uhde TW, Joffe RT, Jimerson DC, Post RM. Normal urinary free cortisol and plasma MHPG in panic disorder: clinical and theoretical implications. Biol Psychiatry 1988; 23: 575–85.[CrossRef][Medline]
  17. Hennig J, Laschefski U, Opper C. Biopsychological changes after bungee jumping: beta-endorphin immunoreactivity as a mediator of euphoria? Neuropsychobiology 1994; 29: 28–32.[Medline]
  18. Deinzer R, Kirschbaum C, Gresele C, Hellhammer DH. Adrenocortical responses to repeated parachute jumping and subsequent h-CRH challenge in inexperienced healthy subjects. Physiol Behav 1997; 61: 507–11.[CrossRef][Medline]
  19. Levine S. Cortisol changes following repeated experiences with parachute training. In: Ursin H, Baade E, Levine S, editors. The psychobiology of stress. A study of coping in men. New York: Academic Press; 1978. p. 51–6.
  20. Ursin H, Baade E, Levine S. The psychobiology of stress. A study of coping in men. New York: Academic Press; 1978.
  21. Kirschbaum C, Pirke KM, Hellhammer DH. The "Trier Social Stress Test" – a tool for investigating psychobiological stress responses in a laboratory setting. Neuropsychobiology 1993; 28: 76–81.[Medline]
  22. 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]
  23. Al’Absi M, Bongard S, Buchanan T, Pincomb GA, Licinio J, Lovallo WR. Cardiovascular and neuroendocrine adjustment to public speaking and mental arithmetic stressors. Psychophysiology 1997; 34: 266–75.[Medline]
  24. Abelson JL, Curtis GC. Cardiac and neuroendocrine responses to exposure therapy in height phobics: desynchrony within the "physiological response system". Behav Res Ther 1989; 27: 561–7.[CrossRef][Medline]
  25. Erdmann G, Voigt K-H. Vegetative und endokrine Reaktionen im Paradigma "Öffentliches Sprechen": Was indizieren sie? In: Debus G, Erdmann G, Kallus KW, editors. Biopsychologie von Stress und emotionalen Reaktionen. Göttingen: Hogrefe; 1995. p. 113–28.
  26. Armario A, Montero J, Balasch J. Sensitivity of corticosterone and some metabolic variables to graded levels of low intensity stresses in adult male rats. Physiol Behav 1986; 37: 559–61.[CrossRef][Medline]
  27. Hennessy M, Levine S. Sensitive pituitary-adrenal responsiveness to varying intensities of psychological stimulation. Physiol Behav 1978; 21: 295–7.[CrossRef][Medline]
  28. Hennessy M, Mendoza S, Mason W, Moberg G. Endocrine sensitivity to novelty in squirrel monkeys and titi monkeys: species differences in characteristic modes of responding to the environment. Physiol Behav 1995; 57: 331–8.[CrossRef][Medline]
  29. Kirschbaum C, Hellhammer D. Salivary cortisol. In: Fink G, editor. Encyclopedia of stress. Vol 3. San Diego: Academic Press; 2000. p. 379–83.
  30. Smyth J, Stone AA, Kirschbaum C. Moving the laboratory into the field: methodological issues in the measurement of salivary cortisol in the natural environment [abstract]. Psychosom Med 1998; 60: 101.
  31. Smyth J, Ockenfels MC, Porter L, Kirschbaum C, Hellhammer DH, Stone AA. Stressors and mood measured on a momentary basis are associated with salivary cortisol secretion. Psychoneuroendocrinology 1998; 23: 353–70.[CrossRef][Medline]
  32. de Vries WR, Bernards NTM, de Rooij MH, Koppeschaar HPF. Dynamic exercise discloses different time-related responses in stress hormones. Psychosom Med 2000; 62: 866–72.[Abstract/Free Full Text]
  33. Hennig J, Friebe J, Ryl I, Kramer B, Bottcher J, Netter P. Upright posture influences salivary cortisol. Psychoneuroendocrionology 2000; 25: 69–83.
  34. Deleted in proof.
  35. Barlow DH, Esler JL, Vitali AE. Psychosocial treatments for panic disorders, phobias, and generalized anxiety disorder. In: Nathan PE, Gorman JM, editors. A guide to treatments that work. New York: Oxford University Press; 1998. p. 288–318.
  36. DeRubeis RJ, Crits-Christoph P. Empirically supported individual and group psychological treatments for adult mental disorders. J Consult Clin Psychol 1998; 66: 37–52.[CrossRef][Medline]
  37. Edwards S, Evans P, Hucklebridge F, Clow A. Association between time of awakening and diurnal cortisol secretory activity. Psychoneuroendocrinology 2001; 26: 613–22.[CrossRef][Medline]
  38. Walker F. Salivary Cortisol determinations in the assessment of adrenal activity. Front Oral Physiol 1984; 5: 33–50.
  39. Kahn JP, Rubinow DR, Davis CL, Kling M, Post RM. Salivary cortisol: a practical method for evaluation of adrenal function. Biol Psychiatry 1988; 23: 335–49.[CrossRef][Medline]
  40. Kirschbaum C, Kudielka BM, Gaab J, Schommer NC, Hellhammer DH. Impact of gender, menstrual cycle phase, and oral contraceptives on the activity of the hypothalamus-pituitary-adrenal axis. Psychosom Med 1999; 61: 154–62.[Abstract/Free Full Text]
  41. Aronsson G, Rissler A. Psychophysiological stress reactions in female and male urban bus drivers. J Occup Health Psychol 1998; 3: 122–9.[CrossRef][Medline]
  42. Sluiter JK, van der Beek AJ, Frings-Dresen MH. Work stress and recovery measured by urinary catecholamines and cortisol excretion in long-distance coach drivers. Occup Environ Med 1998; 55: 407–13.[Abstract/Free Full Text]
  43. Bellet S, Roman L, Kostis J. The effect of automobile driving on catecholamine and adrenocortical excretion. Am J Cardiol 1969; 24: 365–8.[CrossRef][Medline]
  44. Kelsey RM, Blascovich J, Tomaka J, Leitten CL, Schneider TR, Wiens S. Cardiovascular reactivity and adaptation to recurrent psychological stress: effects of prior task exposure. Psychophysiology 1999; 36: 818–31.[CrossRef][Medline]
  45. Galley N. Saccadic eye movements and blink during long-term driving on the autobahn with minimal alcohol ingestion. In: Gale AG, Brown ID, Haselgrave CM, Taylor SP, editors. Vision in vehicles-V. Amsterdam: Elsevier; 1996. p. 381–8.
  46. DSM-IV Diagnostic and statistical manual of mental disorders. 4th ed. Washington DC: American Psychiatric Association; 1994.
  47. First MB, Spitzer RL, Gibbon M, Williams JBW. Structured clinical interview for DSM-IV Axis I disorders. New York: Biometrics Research Department, New York State Psychiatric Institute; 1995.
  48. Derogatis LR. SCL-90-R administration, scoring, and procedures manual II. Towson (MD): Clinical Psychometric Research; 1983.
  49. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory for measuring depression. Arch Gen Psychiatry 1961; 4: 561–71.
  50. Spielberger CD, Gorsuch RL, Lushene RE. Stait-Trait Anxiety Inventory. Palo Alto (CA): Consulting Psychologists Press; 1970.
  51. Marks I, Mathews AM. Brief standard self-rating for phobic patients. Behav Res Ther 1979; 17: 263–7.[CrossRef][Medline]
  52. Alpers GW. Driving Self-Efficacy Inventory. Würzburg, Germany: Psychscience; 2001.
  53. Gibson EL, Checkley S, Papadopoulos A, Poon L, Daley S, Wardle J. Increased salivary cortisol reliably induced by a protein-rich midday meal. Psychosom Med 1999; 61: 214–24.[Abstract/Free Full Text]
  54. Cohen J. Statistical power analysis for the behavioral sciences. Hillsdale (NJ): Erlbaum; 1988.
  55. Yehuda R, Giller EL, Southwick SM, Lowy MT, Mason JW. Hypothalamic-pituitary-adrenal dysfunction in posttraumatic stress disorder. Biol Psychiatry 1991; 30: 1031–48.[CrossRef][Medline]
  56. Pruessner JC, Hellhammer DH, Kirschbaum C. Burnout, perceived stress, and cortisol responses to awakening. Psychosom Med 1999; 61: 197–204.[Abstract/Free Full Text]
  57. Lee Y, Davis M. Role of the hippocampus, the bed nucleus of the stria terminalis, and the amygdala in the excitatory effect of corticotropin-releasing hormone on the acoustic startle reflex. J Neurosci 1997; 17: 6434–46.[Abstract/Free Full Text]
  58. Grillon C. Associative learning deficits increase symptoms of anxiety in humans. Biol Psychiatry 2002; 51: 851–8.[CrossRef][Medline]
  59. Breier A, Albus M, Pickar D, Zahn TP, Wolkowitz OM, Paul SM. Controllable and uncontrollable stress in humans: alterations in mood and neuroendocrine and psychophysiological function. Am J Psychiatry 1987; 144: 1419–25.[Abstract/Free Full Text]
  60. Weiss J. The effects of coping on stress. J Comp Physiol Psychol 1968; 65: 251–60.[CrossRef][Medline]
  61. McEwen BS. Protective and damaging effects of stress mediators. N Engl J Med 1998; 338: 171–9.[Free Full Text]



This article has been cited by other articles:


Home page
 Psychosom. Med.Home page
W. Schlotz, R. Kumsta, I. Layes, S. Entringer, A. Jones, and S. Wust
Covariance Between Psychological and Endocrine Responses to Pharmacological Challenge and Psychosocial Stress: A Question of Timing
Psychosom Med, September 1, 2008; 70(7): 787 - 796.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
L. M. Soravia, M. Heinrichs, A. Aerni, C. Maroni, G. Schelling, U. Ehlert, B. Roozendaal, and D. J.-F. de Quervain
Glucocorticoids reduce phobic fear in humans
PNAS, April 4, 2006; 103(14): 5585 - 5590.
[Abstract] [Full Text] [PDF]

Home page
 Psychosom. Med.Home page
T. L. Gruenewald, M. E. Kemeny, N. Aziz, and J. L. Fahey
Acute Threat to the Social Self: Shame, Social Self-esteem, and Cortisol Activity
Psychosom Med, November 1, 2004; 66(6): 915 - 924.
[Abstract] [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
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Alpers, G. W.
Right arrow Articles by Roth, W. T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Alpers, G. W.
Right arrow Articles by Roth, W. T.
Related Collections
Right arrow Neuroendocrine
Right arrow Neuropsychology
Right arrow Other Psychiatric Disorders
Right arrow Stress and Coping

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS