Advanced search
Publication Cover
The World Journal of Biological Psychiatry Volume 24, 2023 - Issue 5
Submit an article Journal homepage
1,076
Views
1
CrossRef citations to date
0
Altmetric
Original Investigations

The relationship between cortisol awakening response and trait resilience in two patient cohorts and one population-based cohort

Patrick Linnemanna Institute of Epidemiology and Social Medicine, University of Münster, Munster, GermanyCorrespondencePatrick.Linnemann@uni-muenster.de
ORCID Iconhttps://orcid.org/0000-0003-1682-9633View further author information
ORCID Icon,
Nele Friedrichb Institute of Clinical Chemistry and Laboratory Medicine, University Medicine Greifswald, Greifswald, GermanyORCID Iconhttps://orcid.org/0000-0002-1017-7622View further author information
ORCID Icon,
Matthias Nauckb Institute of Clinical Chemistry and Laboratory Medicine, University Medicine Greifswald, Greifswald, GermanyORCID Iconhttps://orcid.org/0000-0002-6678-7964View further author information
ORCID Icon,
Henning Teismanna Institute of Epidemiology and Social Medicine, University of Münster, Munster, GermanyORCID Iconhttps://orcid.org/0000-0001-5807-5572View further author information
ORCID Icon &
Klaus Bergera Institute of Epidemiology and Social Medicine, University of Münster, Munster, GermanyORCID Iconhttps://orcid.org/0000-0001-8966-3684View further author information
ORCID Icon
Pages 429-438 | Received 26 Jun 2022, Accepted 24 Sep 2022, Published online: 14 Oct 2022

Abstract

Objectives

We examined (1) the proportion of cortisol awakening non-responders, (2) the association between cortisol awakening response (CAR) and trait resilience, and (3) the association between CAR increase and trait resilience in two patient cohorts (depression and myocardial infarction [CVD]) and one population-based cohort.

Methods

Eight hundred and eighty study participants delivered CAR scores (response and increase) based on three self-collected saliva samples and a trait resilience score. Descriptive data of CAR non-responders were reported and calculated. Associations between CAR response/increase and trait resilience, sociodemographic and compliance variables were evaluated using multiple logistic and multiple linear regression analyses stratified by cohort.

Results

The proportion of CAR non-responders was high in all cohorts (57% depression cohort, 53.4% CVD cohort, 51.6% control cohort). In the depression cohort age was associated with CAR response and increase. In the CVD cohort salivary collection on a weekday was associated with CAR response and awakening time with CAR increase. In the control cohort age was associated with CAR response and sex with CAR increase.

Conclusions

We observed many CAR non-responders and significant associations between CAR response and CAR increase with single sociodemographic and compliance variables. We did not find significant relationships between CAR response or increase and trait resilience.

Introduction

Activation of the hypothalamic-pituitary-adrenal axis (HPA) and release of the steroid hormone cortisol is an important mechanism of the body to deal with stress (Rosmalen et al. Citation2005). Cortisol is released in a diurnal rhythm – levels rise to a peak in the morning and fall to the lowest level at night (De Weerth et al. Citation2003). The most significant increase in this diurnal rhythm happens in the morning, immediately after waking up and for the following 30 to 45 min period (Pruessner et al. Citation1997; Clow et al. Citation2004) – a phenomenon known as the cortisol awakening response (CAR) (Stalder et al. Citation2016). CAR is widely seen as an indicator of the functioning of the HPA axis (Adam and Kumari Citation2009; Steptoe and Serwinski Citation2016). Deviations from the typical CAR are associated with mental and physical diseases (Stalder et al. Citation2016; Steptoe and Serwinski Citation2016). However, the studies are inconclusive as to the direction of deviation from typical CAR, if it increases or decreases (Steptoe and Serwinski Citation2016). Different deviations from the typical CAR result were reported for depression (Chida and Steptoe Citation2009; Dedovic and Ngiam Citation2015). Prospective studies with adolescents support diverging results on the relationship between CAR and depression. The study by Adam et al. (Citation2010) revealed that a higher CAR was associated with increased odds of developing MDD one year later. In contrast, a study by Carnegie et al. (Citation2014) detected no association between CAR and depression. Nederhof et al. (Citation2015) observed that a lower CAR was related to an increased risk of developing psychopathology, including depression, over three years. The relationship between CAR and a physical disease, such as cardiovascular disease (CVD), has also already been studied (Iob and Steptoe Citation2019). For example, the link between psychosocial stress and CVD is well established in the epidemiological literature (Iob and Steptoe Citation2019). It has previously been postulated that dysregulation of the HPA axis underlies the adverse physiological effects of stress on CVD (Kivimäki and Steptoe Citation2018; Iob and Steptoe Citation2019). The review by Iob and Steptoe (Citation2019) determined that cortisol dysregulation is associated with CVD incidence, cardio-metabolic risk factors, prognosis, and mortality, among others. For example, it was recently found that dysregulated diurnal cortisol patterns are associated with cardiovascular mortality and, at the same time, that large diurnal cortisol fluctuations, such as a large morning CAR increase, appear to have a protective effect on cardiovascular mortality (Karl et al. Citation2022).

In healthy adults, the range of a CAR increase 30 min after awakening varies between 50 and 160% (Clow et al. Citation2004). However, some individuals, so-called non-responders, show a blunted increase, no increase at all, or even a CAR decrease (Stalder et al. Citation2016). It is consensus that the occurrence of non-responders is a real phenomenon that is not explained by putatively inaccurate handling of biosamples or analysis (Stalder et al. Citation2016). In a general population study, 25% of the participants were non-responders (Wüst et al. Citation2000), while in a sample recruited for suspected coronary artery disease, the proportion of non-responders was as low as 14.7% (Dockray et al. Citation2008). However, among patients suffering from neurotic and personality disorders (Dembińska et al. Citation2020), the percentage of non-responders was 43.1%. These findings suggest that the proportion of non-responders varies according to the specific study group (Stalder et al. Citation2016). Regarding the time stability of the CAR: CAR is regarded as the least stable among cortisol output indices (Laceulle et al. Citation2015). However, it is relatively stable (Wüst et al. Citation2000; Platje et al. Citation2013) but not comparable with classic traits like extroversion (Laceulle et al. Citation2015).

The ability to adapt to adverse circumstances or to cope with them is called resilience (Southwick et al. Citation2014). One way of conceptualising resilience is the trait approach, which considers resilience as a relatively time-stable personality trait (Hu et al. Citation2015; Linnemann et al. Citation2020). The relationship between diurnal cortisol rhythm and trait resilience has already been investigated (Lai et al. Citation2020). Lai et al. (Citation2020) argue that higher trait resilience is associated with better health outcomes. Thus, they conclude that resilient persons should show a similar diurnal cortisol rhythm as in healthy persons. They examined the association between diurnal cortisol rhythm and trait resilience in undergraduate students. The study included CAR among other analyses of diurnal cortisol rhythm. They found that higher trait resilience was associated with a higher CAR and a subsequently stronger cortisol level decline over the day. In addition, it has been shown that in children with parents diagnosed with HIV, trait resilience was positively associated with higher salivary cortisol levels upon awakening and steeper cortisol slopes over the day, but not with CAR (Chi et al. Citation2015). Lai et al. (Citation2020) concluded that the diverging results on the association between diurnal cortisol rhythm and resilience could be due to the examination of different study populations and/or the use of different survey instruments, and thus to different conceptualisations of resilience.

The aim of the present study was to analyse (1) the proportion of CAR responders, (2) the association between CAR response and trait resilience, and (3) the association between CAR increase and trait resilience. Analyses were carried out separately in two patient (depression vs. cardiovascular disease) and one population-based cohort to account for the very own characteristics of each cohort.

Methods

Study design

We used data from the BiDirect study (Wersching and Berger Citation2012; Teismann et al. Citation2014). The BiDirect study was a large prospective cohort study investigating the bidirectional relationship between depression and (subclinical) arteriosclerosis. BiDirect is conducted in and around the university city of Münster, Germany. Participants were recruited simultaneously into three distinct cohorts: (1) Patients hospitalised with an acute episode of depression (‘depression cohort’). (2) Patients 3–4 months after an acute coronary event (‘CVD cohort’). (3) Individuals randomly invited via the registry of the city of Münster (‘population-based cohort’). The BiDirect study consists of one baseline and three follow-up examinations and included several investigations that could contribute to explain the bidirectional relationship between (subclinical) atherosclerosis and depression. The investigations included, for example, (1) a computer-assisted interview on sociodemographic and medical histories, (2) diagnostic tests such as MRI and blood sampling, (3) questionnaires, for example, on childhood trauma experiences or pain perception. The ethics committee of the University of Münster and the Westphalian Chamber of Physicians in Münster, Germany, approved the study. All participants gave written informed consent. More details on the BiDirect Study can be found in (Teismann et al. Citation2014).

Procedure

The participants collected saliva samples at home. Cortisol levels in humans can be determined by various means such as blood, hair, urine, or saliva (Kobayashi and Miyazaki Citation2015). The determination of cortisol levels in saliva is advantageous over other methods in that it is simple to perform, non-invasive, and does not cause pain (Kalman and Grahn Citation2004). Study assistants explained the exact procedure of taking saliva samples to the participants during their visit at the study centre; moreover, additionally to the saliva sample tubes (Salivette®, Sarstedt, Nümbrecht, Germany), the participants took home a detailed information sheet including these instructions. Participants were instructed to collect their first saliva sample in the late evening (s1). The second sample (s2) was supposed to be taken immediately after waking up the following day while still lying in bed. The third sample (s3) was to be taken exactly 30 min after the second sample. The participants were instructed not to have eaten, drunk, smoked, or brushed their teeth before obtaining the samples. In order to control whether the participants followed the instructions, they were asked to fill in a standardised compliance questionnaire (paper-based). The three saliva samples were sent back by pre-stamped mail envelopes to the study centre, where the samples were frozen at −80 °C. Saliva cortisol analyses were undertaken in a batch at the Institute of Clinical Chemistry and Laboratory Medicine, University Medicine Greifswald, between November 2019 and January 2020. For the measurements an automated assay based on chemiluminesence technology was used (IDS-iSYS Salivary Cortisol on the IDS-iSYS Multi-Discipline Automated System, Immunodiagnostic Systems, Boldon, UK). The coefficients of variance (CV) were 15.8%, 13.3%, and 13.0%, respectively, for low, median, or high levels.

Measures

Cortisol awakening response (CAR)

In BiDirect, CAR was operationalised as the increase in total cortisol level over a time interval of roughly 30 min, starting from the salivary cortisol output level on awakening (s2). Therefore, the area under the curve with respect to the increase (AUCi) formula by Pruessner et al. (Citation2003; formula 5) was applied A U C i = ∑ i = 1 n − 1 ( m i + 1 − m i ) t i 2 − m i ∑ i = 1 n − 1 t i where m i denotes the first morning measurement and t i the time distances between measurements. Notably, in BiDirect, only two measurements were made (at s2, and 30 min later), so that the formula simplifies to A U C i = ( m 2 + m 1 ) t 1 2 . We abbreviated the area under the curve associated with the cortisol awakening response as CARAUCi.

Following Wüst et al. (Citation2000) and the recommendation of an expert’s consensus (Stalder et al. Citation2016), we used a dichotomous CAR response variable (CARresp; 0 = responder, 1 = non-responder). This variable categorises individuals with an absolute CAR increase higher than 2.5 nmol/L within 30 min after awakening as responders, and individuals with a CAR increase less than 2.5 nmol/L as non-responders.

Before creating the two CAR variables (CARAUCi & CARresp), we converted the salivary cortisol raw values, initially reported in micrograms/decilitre in the BiDirect study, to nmol/L by a factor of 27.59.

Covariates

As sociodemographic variables, we considered age, sex, and years of full-time education. Furthermore, a dichotomous compliance value (0: non-compliant; 1: compliant) was calculated for each of the three salivary cortisol sampling points to control if a person complied with the instructions; in addition, we considered whether the saliva samples were collected during the week or at the weekend (0: weekday; 1: weekend) and the exact time of a participant’s awakening (expressed in minutes after midnight [00:00 time] of the relevant day). These covariates should be considered and controlled for when analysing salivary cortisol data (Stalder et al. Citation2016).

Trait resilience

Resilience, conceptualised as a relatively stable personality trait, was assessed using the Resilienzskala 11 (RS-11; Schumacher et al. Citation2005), a German short version of Wagnild and Young (Citation1993) Resilience Scale 25 (RS-25). The RS 11 captures in 11 items traits that should help to deal with difficult life circumstances, such as self-confidence and perseverance (example item 1: ‘When I have plans, I follow them’). The response options for the 11 items range from 1 = ‘I don’t agree’ through 7 = ‘I completely agree’. Thus, the RS-11 sum score ranges from 11 to 77, with more items corresponding to a more pronounced resilience. The RS 11 is a reliable (internal consistency with a Cronbach´s Alpha of .91), valid (convergent validity: correlation between resilience and self-efficacy of r = .70), and economical way (correlation between RS-11 and RS-25 of r = .95) to assess resilience (Schumacher et al. Citation2005).

Depressive symptoms

Depressive symptoms were assessed using the Centre for Epidemiologic Studies Depression Scale (CES-D; Radloff Citation1977). The CES-D scale is a self-report scale that measures depressed mood within the past week. The items capture different facets of depression, such as depressed mood or loss of appetite. The psychometric properties of the CES-D scale are very satisfying (for a detailed description, see Siddaway et al. Citation2017).

Sampling

BiDirect participants who participated and answered the questionnaires (e.g. RS-11; CES-D) in the third face-to-face examination, which took place about five years after the baseline examination, obtained saliva samples independently at home, and returned them by mail to the study centre for analysis. We included participants in the present study with two available saliva samples at awakening and 30 min after awakening. Consistent with Stalder et al.’ (2016) proposal, participants were excluded from the CAR analysis if their sample deviated more than 5 min from the desired 30 min window after awakening.

Data analysis

The data were analysed using R (R Core Team Citation2021) and R Studio (R Studio Team Citation2020). For categorical variables, descriptive statistics are given as counts and percentages; group differences were tested for significance by Chi square test. For numerical variables, we report means and standard deviations; group differences were tested for significance by oneway ANOVA.

Data analysis was carried out in several steps. All analyses were conducted separately for the three cohorts. First, multiple logistic regression analysis was employed to determine the relationship between the dependent variable CAR response (CARresp) and the independent variable trait resilience. The basic model included the independent variables, age, sex, education years, compliance with saliva sampling, sampling during weekend vs. weekday, and awakening time. The full model additionally included the summary score of the trait resilience scale.

Second, multiple linear regression analysis was employed to analyse the relationship between the dependent variable CARAUCi and the independent variable trait resilience. Modelling was analogous to the multiple logistic regression analyses computed in step 1.

Third, we re-ran the multiple logistic and linear regression analyses (steps 1 and 2) after replacing the independent variable trait resilience with the summary score of the CES-D (Centre for Epidemiologic Studies Depression Scale sum score (CES-D; Radloff Citation1977)). These analyses were intended to generate insights into the relationship between CAR and another mental health indicator.

Fourth, we conducted sensitivity analyses (1) to map a person’s diurnal cortisol rhythm as comprehensively as possible to have additional indicators of the diurnal cortisol output of an individual and (2) to detect a potential relationship between the diurnal cortisol rhythm and trait resilience. Therefore, the multiple linear regression analyses from step 2 were re-run after exchanging the dependent variable CARAUCi by the total increase in cortisol level between late evening and the highest level in the morning (Pruessner et al. Citation2003; formula 5). Moreover, we re-calculated the absolute CAR increase (CARi) as the difference between the raw s3 cortisol value and the raw s2 cortisol value.

For all regression analyses reported here, the assumptions regarding the distributions of residuals, multicollinearity, and heteroscedasticity were not violated. The reduction in the number of subjects examined in the respective regression analyses resulted from missing values in the respective relevant outcomes and predictors. We performed a case-wise deletion for these individuals and assumed that the missing values were completely at random.

Results

A total of 880 participants were included in the analyses. Descriptive statistics are shown in . The proportion of women was low in the CVD cohort, high in the depression cohort, and balanced in the control cohort. Age and education years differed significantly between cohorts. The raw measured salivary cortisol levels were lowest at all three measurement time-points in the CVD cohort, but differences were significant between cohorts only at awakening (s2). Participants in the depression cohort were least compliant in saliva sampling. shows the individual cortisol levels at awakening (s2) and 30 min after awakening (s3). The CARresp and the CARAUCi differed non-significantly between the cohorts and were highest, or respectively lowest, in the depression cohort. Participants in the control cohort had the highest resilience score, followed by the CVD and depression cohort members.

Figure 1. Individual raw salivary cortisol levels (nmol/L) change from awakening to 30 min after being stratified.

Figure 1. Individual raw salivary cortisol levels (nmol/L) change from awakening to 30 min after being stratified.

Table 1. Participants’ characteristics.

In logistic regression models, determinants of CAR response were analysed (). The full model that included the resilience score explained a higher proportion of the variance in CAR response in the depression cohort and in the CVD cohort, and the same amount of variance in the control cohort compared to the basic model. There was a significant effect of age in the depression cohort. For a one year increase in age, the odds of being a CAR non-responder increased by 4% (OR = 1.04). In the CVD cohort, the odds of being a non-responder decreased by 58% if saliva collection was carried out on weekends.

Table 2. Results of the full model of the multiple logistic regression analysis with CARresp as outcome stratified by cohort.

There was a significant effect of sex and age in the control cohort: women´s odds of being a non-responder were smaller. In contrast to the depression cohort, a one-year increase in age decreased the odds of cortisol non-response by 3%.

shows the results of the multiple linear regression analysis employing the same independent variables, but with the increase in cortisol levels as the dependent variable. The full model explained more variance in the cortisol increase (CARAUCi) in the depression and CVD cohorts and the same amount in the control cohort compared to the basic model. Higher age in the depression cohort was significantly associated with a decrease in CARAUCi. In the CVD cohort, a significant effect of awakening time (estimate = −0.80, 95% CI = −1.37 to –0.23) was observed. An earlier awakening was associated with a decrease in CARAUCi. In the control cohort, female sex was positively associated with an increase of CARAUCi.

Table 3. Results of the full model of the multiple linear regression analysis with CARAUCi as outcome stratified by cohort.

The sensitivity analyses revealed no significant relationships between depressive symptoms and CARresp or CARAUCi, respectively, and no significant relationship between the evening to highest morning increase of the diurnal cortisol output and trait resilience. In addition, the CARi analyses did not differ from the CARAUCi analyses (data not shown).

Discussion

In the BiDirect study, the proportion of CAR non-responders, characterised by a CAR increase of less than 2.5 nmol/L (Wüst et al. Citation2000), exceeded the proportion of CAR responders in all three cohorts, but did not differ significantly between cohorts. The following section discusses the large number of non-responders in their cohort affiliation.

The high proportion of CAR non-responders in the depression cohort is in line with previous studies that have examined the proportion of CAR non-responders in populations with mental disorders. For instance, the proportion of non-responders was high in patients receiving psychotherapy due to a depressive disorder (43.8%; Huber et al. Citation2006) and patients suffering from neurotic and personality disorders (43.1%; Dembińska et al. Citation2020). The proportion of non-responders in the two other cohorts was considerably higher than in prior studies (Wüst et al. Citation2000; Dockray et al. Citation2008). One problem with CAR studies is the high degree of heterogeneity in its measurement (Stalder et al. Citation2016). The studies described above differed in the number of measurement time-points used to assess CAR. CAR was analysed either on one day at two time- points (Dembińska et al. Citation2020), on one day at three timepoints (Huber et al. Citation2006), or even on two days at three timepoints (Wüst et al. Citation2000).

The equally high numbers of non-responders in all three BiDirect cohorts may have several explanations. First, recruitment into the study including categorisation into a specific cohort took place on average five years before the salivary cortisol level measurements. Thus, the disease event as the recruitment reason may no longer have the same importance as during the recruitment. However, the individuals in the three cohorts still differed significantly in their psychological health variables, trait resilience and depressive symptoms. Second, storage at −80 °C for around two and a half years might have affected cortisol levels, but this holds true for all three salivary samples of an individual and not just the morning ones. Another explanation for the large number of non-responders could be that the CAR in the present study was recorded only once 30 min after awakening and not a second time 45 min later. Thus, it is possible that the CAR peak between 30 and 45 min after wakening (Pruessner et al. Citation1997) was not recorded and that individuals would otherwise have had a more pronounced CAR.

The main aim of the present study was to investigate the relationship between CAR and trait resilience. What is especially appealing about exploring the relationship between diurnal cortisol release and resilience is that it might provide an objective biomarker that could complement self-questionnaires to assess resilience, for example, to investigate the relationship of resilience with potential protective factors (Jang et al. Citation2022). However, we did not find a significant relationship in one of the three BiDirect cohorts. This finding is at odds with recent findings on the relationship between CAR and trait resilience. For example, O’Connor et al. (Citation2021) found that lower levels of trait resilience are significantly associated with lower CAR. In their study, in order to calculate the CAR, raw cortisol levels were assessed over seven days and over three measurement timepoints (awakening time, 30 min later, 45 min later) in participants who were recruited based on a history of suicidal attempt or ideation or as control participants without a psychological condition. Trait resilience was assessed using the Brief Resilience Scale (BRS; Smith et al. Citation2008). The differences in timepoints and scale might explain their different result. Another recent study revealed a significant relationship between a stronger CAR and higher trait resilience values in undergraduate students (Lai et al. Citation2020). In this case, CAR was calculated, like our study, as the Area Under the Curve with respect to increase (AUCi; Pruessner et al. Citation2003; formula 5), but trait resilience was also assessed by means of a Chinese version of the BRS (Lai and Yue Citation2014). In contrast, no significant relationship between CAR was detected in a study (Chi et al. Citation2015) conducted with children with parents suffering from HIV. How can the discrepancies between studies be understood? An overall heterogenous pattern in studies on the relationship between CAR and psychosocial functioning is regarded as ‘the rule rather than the exception’ (Chida and Steptoe Citation2009; Boggero et al. Citation2017, p. 207). This may apply to the relationship between CAR and trait resilience as well. The cited studies considerably differ in how raw cortisol levels were collected and examined, how CAR was calculated, in the type and size of the study population, and how trait resilience was measured and operationalised. Thus, results on the relationship between CAR and trait resilience should only be compared taking the operationalisation of constructs and the methods used into account.

Despite the negative results for trait resilience, some significant determinants of CAR response or CAR increase were found in our study. We think it is important to highlight that we followed the recommendation to include state and compliance variables in our study (Stalder et al. Citation2016). We discuss the significant determinants of CAR response or CAR increase, beginning with the depression cohort and followed by the CVD and control cohorts.

In the depression cohort, the probability of being a CAR non-responder increased with age, while it decreased with age in the control cohort. Regarding the relationship between age and CARAUCi, there was a negative relationship in the depression cohort, i.e. a decrease in CARAUCi with ageing. The finding in the depression cohort that CAR decreases with age is in line with previous results (Heaney et al. Citation2010). However, also the present ambiguous findings on CAR response and age reflect very well the state of research on age and CAR. For instance, an increased CAR increase with age was found in adolescent and adult females suffering from borderline personality disorder, but not in adolescent and adult females of a healthy control (Rausch et al. Citation2021).

There was a significant association between CARresp and sex in the control cohort, or CARAUCi and sex, respectively. The probability of being a female non-responder was lower than being a male non-responder. In addition, being a woman was positively associated with a stronger CARAUCi. This result aligns with the findings that women tend to have a more substantial CAR increase than men (Stalder et al. Citation2016).

In the CVD cohort, conducting salivary cortisol measurement on the weekend was associated with a lower probability of being a non-responder. It is already known from previous CAR studies that it indeed matters whether the samples are taken on a weekday or the weekend (Stalder et al. Citation2016). For instance, a higher CAR was found on work days than on weekends (Kunz-Ebrecht et al. Citation2004). To the best of our knowledge, we found no studies that reported on CAR non-response influenced by weekday. This might again contribute to non-comparability between studies.

In the CVD cohort, earlier awakening was associated with a lower CARAUCi. The literature on the relationship between CAR and awakening seems to be unambiguous. For instance, studies conducted with healthy individuals have shown that an earlier awakening is associated with a more significant CAR increase (Law et al. Citation2013). However, Elder et al. (Citation2014) argue that earlier awakening could be an artefact for shorter sleep duration. Moreover, some studies did not detect any significant relationship between CAR and wakening time (Souza-Talarico et al. Citation2022). Thus, the present finding adds a little more to the existing heterogeneous literature.

Strengths and limitations

In the present study, we were able to control for important known covariates in CAR studies. In their meta-analysis, Boggero et al. (Citation2017) identified this as a quality feature of a CAR study. In the BiDirect study, the compliance, e.g. food intake, before saliva sampling was monitored; however, these monitorings were self-reported. Another strength of the present study is the large number of participants recruited into three distinct cohorts.

A particular shortcoming is that CAR was determined using only two measurement time-points on one day. Thus, the possible CAR peak, which lies between 30 and 45 min after awakening, could have been missed (Clow et al. Citation2004). This disadvantage was due to logistics and cost. However, we were able to include a rather large sample size. Recruitment into the cohorts happened roughly five years before the salivary cortisol sampling. Another shortcoming is that the present analysis is cross-sectional, and it cannot be ruled out that trait and especially state variables change over time. Thus, causal inferences cannot be drawn based on cross-sectional studies.

Conclusions

In the BiDirect Study, a high number of CAR non-responders was observed in two patient and one population-based cohort. Notably, significant associations were found between two CAR measures and single sociodemographic and compliance variables. No significant relationship between CAR and trait resilience was found in any cohort. This result is part of a heterogeneous body of research on CAR and trait resilience.

Acknowledgment

We want to express our sincere gratitude to all participants and staff of the BiDirect study.

Statement of interest

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The German Federal Ministry of Education and Research [grants 01ER0816 and 01ER1506] funded the BiDirect Study.

References

  • Adam EK, Doane LD, Zinbarg RE, Mineka S, Craske MG, Griffith JW. 2010. Prospective prediction of major depressive disorder from cortisol awakening responses in adolescence. Psychoneuroendocrinol. 35(6):921–931.
  • Adam EK, Kumari M. 2009. Assessing salivary cortisol in large-scale, epidemiological research. Psychoneuroendocrinol. 34(10):1423–1436.
  • Boggero IA, Hostinar CE, Haak EA, Murphy ML, Segerstrom SC. 2017. Psychosocial functioning and the cortisol awakening response: meta-analysis, P-curve analysis, and evaluation of the evidential value in existing studies. Biol Psychol. 129:207–230.
  • Carnegie R, Araya R, Ben-Shlomo Y, Glover V, O’Connor TG, O’Donnell KJ, Pearson R, Lewis G. 2014. Cortisol awakening response and subsequent depression: prospective longitudinal study. Br J Psychiatry. 204(2):137–143.
  • Chi P, Slatcher RB, Li X, Zhao J, Zhao G, Ren X, Zhu J, Stanton B. 2015. Perceived stigmatization, resilience, and diurnal cortisol rhythm among children of parents living with HIV. Psychol Sci. 26(6):843–852.
  • Chida Y, Steptoe A. 2009. Cortisol awakening response and psychosocial factors: a systematic review and meta-analysis. Biol Psychol. 80(3):265–278.
  • Clow A, Thorn L, Evans P, Hucklebridge F. 2004. The awakening cortisol response: methodological issues and significance. Stress. 7(1):29–37.
  • De Weerth C, Zijl RH, Buitelaar JK. 2003. Development of cortisol circadian rhythm in infancy. Early Hum Dev. 73(1–2):39–52.
  • Dedovic K, Ngiam J. 2015. The cortisol awakening response and major depression: examining the evidence. Neuropsychiatr Dis Treat. 11:1181–1189.
  • Dembińska E, Rutkowski K, Sobański J, Mielimąka M, Citkowska-Kisielewska A, Klasa K, Konietzka M. 2020. Abnormal cortisol awakening responses in patients with neurotic and personality disorders admitted for psychotherapy in day hospital. J Psychiatr Res. 130:207–214.
  • Dockray S, Bhattacharyya MR, Molloy GJ, Steptoe A. 2008. The cortisol awakening response in relation to objective and subjective measures of waking in the morning. Psychoneuroendocrinol. 33(1):77–82.
  • Elder GJ, Wetherell MA, Barclay NL, Ellis JG. 2014. The cortisol awakening response–applications and implications for sleep medicine. Sleep Med Rev. 18(3):215–224.
  • Heaney JL, Phillips AC, Carroll D. 2010. Ageing, depression, anxiety, social support and the diurnal rhythm and awakening response of salivary cortisol. Int J Psychophysiol. 78(3):201–208.
  • Hu T, Zhang D, Wang J. 2015. A meta-analysis of the trait resilience and mental health. Pers Individ Differ. 76:18–27.
  • Huber TJ, Issa K, Schik G, Wolf OT. 2006. The cortisol awakening response is blunted in psychotherapy inpatients suffering from depression. Psychoneuroendocrinol. 31(7):900–904.
  • Iob E, Steptoe A. 2019. Cardiovascular disease and hair cortisol: a novel biomarker of chronic stress. Curr Cardiol Rep. 21(10):1–11.
  • Jang S, Choi SW, Ahn R, Lee JY, Kim J, Seok JH. 2022. Relationship of resilience factors with bio-psycho-social markers using a comprehensive home evaluation kit for depression and suicide risk: a real-world data analysis. Front Psychiatry. 13:847498.
  • Kalman BA, Grahn RE. 2004. Measuring salivary cortisol in the behavioral neuroscience laboratory. J Undergrad Neurosci Educ. 2(2):A41–A49.
  • Karl S, Johar H, Ladwig KH, Peters A, Lederbogen F. 2022. Dysregulated diurnal cortisol patterns are associated with cardiovascular mortality: findings from the KORA-F3 study. Psychoneuroendocrinol. 141:105753.
  • Kivimäki M, Steptoe A. 2018. Effects of stress on the development and progression of cardiovascular disease. Nat Rev Cardiol. 15(4):215–229.
  • Kobayashi H, Miyazaki Y. 2015. Distribution characteristics of salivary cortisol measurements in a healthy young male population. J Physiol Anthropol. 34(1):1–4.
  • Kunz-Ebrecht SR, Kirschbaum C, Marmot M, Steptoe A. 2004. Differences in cortisol awakening response on work days and weekends in women and men from the Whitehall II cohort. Psychoneuroendocrinol. 29(4):516–528.
  • Laceulle OM, Nederhof E, van Aken MA, Ormel J. 2015. Adolescent personality: associations with basal, awakening, and stress‐induced cortisol responses. J Pers. 83(3):262–273.
  • Lai JC, Leung MO, Lee DY, Lam YW, Berning K. 2020. Biomarking trait resilience with salivary cortisol in Chinese undergraduates. Front Psychol. 11:536510.
  • Lai JC, Yue X. 2014. Using the brief resilience scale to assess Chinese people’s ability to bounce back from stress. Sage Open. 4(4):215824401455438.
  • Law R, Hucklebridge F, Thorn L, Evans P, Clow A. 2013. State variation in the cortisol awakening response. Stress. 16(5):483–492.
  • Linnemann P, Wellmann J, Berger K, Teismann H. 2020. Effects of age on trait resilience in a population-based cohort and two patient cohorts. J Psychosom Res. 136:110170.
  • Nederhof E, Van Oort FVA, Bouma EMC, Laceulle OM, Oldehinkel AJ, Ormel J. 2015. Predicting mental disorders from hypothalamic-pituitary-adrenal axis functioning: a 3-year follow-up in the TRAILS study. Psychol Med. 45(11):2403–2412.
  • O’Connor DB, Branley-Bell D, Green JA, Ferguson E, O’Carroll RE, O’Connor RC. 2021. Resilience and vulnerability factors influence the cortisol awakening response in individuals vulnerable to suicide. J Psychiatr Res. 142:312–320.
  • Platje E, Vermeiren RR, Branje SJ, Doreleijers TA, Meeus WH, Koot HM, Frijns T, van Lier PAC, Jansen LM. 2013. Long-term stability of the cortisol awakening response over adolescence. Psychoneuroendocrinology. 38(2):271–280.
  • Pruessner JC, Kirschbaum C, Meinlschmid G, Hellhammer DH. 2003. Two formulas for computation of the area under the curve represent measures of total hormone concentration versus time-dependent change. Psychoneuroendocrinology. 28(7):916–931.
  • Pruessner JC, Wolf OT, Hellhammer DH, Buske-Kirschbaum A, Von Auer K, Jobst S, Kaspers C, Kirschbaum C. 1997. Free cortisol levels after awakening: a reliable biological marker for the assessment of adrenocortical activity. Life Sci. 61(26):2539–2549.
  • R Core Team. 2021. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. URL https://www.R-project.org/.
  • Radloff LS. 1977. The CES-D scale: a self-report depression scale for research in the general population. Appl Psychol Meas. 1(3):385–401.
  • Rausch J, Flach E, Panizza A, Brunner R, Herpertz SC, Kaess M, Bertsch K. 2021. Associations between age and cortisol awakening response in patients with borderline personality disorder. J Neural Transm. 128(9):1425–1432.
  • Rosmalen JGM, Oldehinkel AJ, Ormel J, De Winter AF, Buitelaar JK, Verhulst FC. 2005. Determinants of salivary cortisol levels in 10–12 year old children; a population-based study of individual differences. Psychoneuroendocrinology. 30(5):483–495.
  • R Studio Team 2020. RStudio: Integrated Development for R. RStudio, Boston, MA: PBC. http://www.rstudio.com/.
  • Schumacher J, Leppert K, Gunzelmann T, Strauß B, Brähler E. 2005. Die Resilienzskala – Ein Fragebogen zur Erfassung der psychischen Widerstandsfähigkeit als Personmerkmal. Z Klin Psychol Psychiatr Psychother. 53(1):16–39.
  • Siddaway AP, Wood AM, Taylor PJ. 2017. The Center for Epidemiologic Studies-Depression (CES-D) scale measures a continuum from well-being to depression: testing two key predictions of positive clinical psychology. J Affect Disord. 213:180–186.
  • Smith BW, Dalen J, Wiggins K, Tooley E, Christopher P, Bernard J. 2008. The brief resilience scale: assessing the ability to bounce back. Int J Behav Med. 15(3):194–200.
  • Southwick SM, Bonanno GA, Masten AS, Panter-Brick C, Yehuda R. 2014. Resilience definitions, theory, and challenges: interdisciplinary perspectives. Eur J Psychotraumatology. 5(1):25338.
  • Souza‐Talarico JN, Santos S, Trevisan Zanetta DM, Marcourakis T, Paula Santos U. 2022. Association between awakening cortisol levels and cardiovascular performance in sugarcane rural workers. Stress Health. 38(1):102–110.
  • Stalder T, Kirschbaum C, Kudielka BM, Adam EK, Pruessner JC, Wüst S, Dockray S, Smyth N, Evans P, Hellhammer DH, et al. 2016. Assessment of the cortisol awakening response: expert consensus guidelines. Psychoneuroendocrinology. 63:414–432.
  • Steptoe A, Serwinski B. 2016. Cortisol awakening response. In: Fink G., editor. Stress: concepts, cognition, emotion, and behavior. London, Oxford, Boston, New York, San Diego: Elsevier Academic Press; p. 277–283.
  • Teismann H, Wersching H, Nagel M, Arolt V, Heindel W, Baune BT, Wellmann J, Hense HW, Berger K. 2014. Establishing the bidirectional relationship between depression and subclinical arteriosclerosis–rationale, design, and characteristics of the BiDirect Study. BMC Psychiatry. 14(1):174–179.
  • Wagnild GM, Young HM. 1993. Development and psychometric. J Nurs Mea. 1(2):165–17847.
  • Wersching H, Berger K. 2012. New cohorts. The BiDirect study. Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz. 55(6–7):822–823.
  • Wüst S, Federenko I, Hellhammer DH, Kirschbaum C. 2000. Genetic factors, perceived chronic stress, and the free cortisol response to awakening. Psychoneuroendocrinology. 25(7):707–720.
Download PDF

Related research

People also read lists articles that other readers of this article have read.

Recommended articles lists articles that we recommend and is powered by our AI driven recommendation engine.

Cited by lists all citing articles based on Crossref citations.
Articles with the Crossref icon will open in a new tab.