Untitled Document




Analysis of salivary cortisol as occupational stress biomarker in nursing workers


Juliana Faria CamposI; Helena Maria Scherlowski Leal DavidII


IPhD in nursing by UERJ. Assistant Professor, Department of Fundamental Nursing, School of Nursing Anna Nery, Federal University of Rio de Janeiro. Brazil. E-mail: jujufariacampos@yahoo.com.br.
IIPhD in Public Health. Associate Professor, Department of Public Health Nursing by the Graduate Nursing Program, College of Nursing at the University of Rio de Janeiro State. Director of the College of Nursing. Rio de Janeiro, Brasil. E-mail: helenalealdavid@gmail.com.

ABSTRACT: Sectional analytic observational epidemiological study conducted from March to April 2012, whose object was the levels of salivary cortisol of 103 nursing workers of a public hospital in Rio de Janeiro. The purpose was to measure and analyze the cortisol levels of these workers. It was collected four saliva samples from each participant in a work day. Cortisol was quantified using indices: average increase (MnInc), excretion in the period after waking (AUCtrab) and area under the curve from zero of the diurnal cycle (AUCCD). The statistical association between covariates and cortisol levels was assessed by bivariate analysis. The average values ​​of cortisol levels were: MnInc 10.78 nmol/L (± 6.99), AUCtrab and AUCCD were 32.51 nmol/L (± 21.99) and 107.99 nmol/L (± 61.63), respectively. It is concluded that these results provide a contribution to understand the biological pathways by which work stress influences health.

Keywords: Occupational stress; salivary cortisol; nursing; worker’s health.


It is admitted that the look on the health of the worker should be modified and updated to account for the complexities of the current work context that occur under new management, and may involve risks and costs for the worker.

In the field of study of worker’s health, there is focus on occupational stress, which arises as a result of the imbalance between the working conditions, the responsiveness of workers to perform their tasks associated with the level of control allowed to meet the demands and the social support received from employees and managers. If this imbalance is maintained, it may result in chronic stress, which may cause distress or illness of worker1.

This research on occupational health aimed to evaluate the levels of salivary cortisol on nursing workers.


When an individual is exposed to stressful situation, neuroendocrine systems are activated and secrete substances such as adrenaline and cortisol, promoting physiological reactions in response to stress. If the possibility of recovery of balance is impaired, the body suffers due to changing levels of catecholamines and cortisol in the bloodstream2.

The study of the ways in which occupational stress produces adverse health effects is an intriguing challenge. Research indicates that the pathways by which work stress influences health are mediated by the hypothalamus-pituitary-adrenal (HPA) axis, which regulates the body's adaptation to stress in a long term perspective. Changes in the functioning of the HPA axis due to acute or chronic stress are usually examined in ambulatory settings, investigating the cortisol awakening response (CAR) and diurnal cortisol secretory activity, allowing a measurement of the physiological response to stress3.

Currently, salivary cortisol is considered the most promising marker to assess the neurobiological response to stress3. It plays an important role in research in occupational health because of its potential use to assess the physiological response in working groups exposed to workload and stress.

High levels of both catecholamines (adrenaline and noradrenaline) and cortisol maintained in the bloodstream can lead to negative effects on health, such as hypertension, musculoskeletal disorders, mental disorders, among others1.

Changes in the functioning of the HPA axis are usually examined by investigating the salivary cortisol awakening response (CAR) and diurnal cortisol secretory activity. This method allows for repeated cortisol sampling during the day at work, and consequently evaluates a part of the physiological response to stress3.

There are numerous variables that can affect the CAR and cortisol levels during the day, such as sociodemographic (age, sex, genetic factors), lifestyle (exposure to stressors, smoking, consumption of food, mood, sleep, diseases, use of drugs, among others), as well as the time and day of sampling and the method used to determine cortisol levels.

Nursing has its work inserted into an economic and political context driven by capitalist accumulation mode, and it was not immune of the pressures introduced into the working world. This work, by its nature and characteristics, proves to be especially susceptible to the phenomenon of occupational stress, since it concentrates high number of risk factors/load to physical and mental health of workers4.


This is a clipping from a cross-sectional observational study conducted in a public hospital specialized in cardiology, located in the municipality of Rio de Janeiro.

The study population consisted of all active nursing professionals in that hospital, temporary or permanent basis, and who met the following inclusion criteria: age greater than 18 years and time on the job less than or equal to six months.

Exclusion criteria were: non-acceptance in participating in the study, pregnancy, contraceptive use, presence of health problems involving the endocrine system, use of corticosteroids by any means, use of sleep medications, antidepressants, or those who were acutely ill at the time of collection, licensed employees, and those who worked all nights prior to its shift at the hospital.

The source population for the study was 392 employees. Of these, 287 were excluded after applying the exclusion criteria. There was loss of three employees by insufficient volume of saliva. The final study sample was of 103 participants.

Data collection occurred from March to April 2012, and was completed in two phases. The first phase consisted in Questionnaire - Part 1 by the interviewer. This, comprising a set of initial questions, which are the factors of exclusion, and two blocks of questions related to the interviewee identification and characterization of the work. The second phase consisted in the collection of samples of salivary cortisol.

The collection of the cortisol was previously scheduled, and to avoid forgetting, participants were reminded the day before the scheduled for collection of saliva sample through a phone call and a text message via cell phone. Four samples were collected in one (01) day job (duty): upon waking, 30 minutes after waking, 12 a.m. and 6 p.m. The first two samples were collected by the participants themselves, the other by the research team. For collection, Salivette® Cortisol device was used.

Since no device to control time was used, it was adopted for evaluation of adherence to protocol for collecting the samples the criterion of a minimum CAR and 2.5 nmol/L, in other words, all individuals who exhibited a difference of <2.5 nmol/L on cortisol levels measured 30 minutes after awakening in comparison with the levels obtained immediately upon awakening were found non-adhering to protocol.

The study was based on the ethical principles of Resolution 466/12 of the National Board of Health. Research institution authorized the implementation of the study and its ethics committee approved the project and the consent under number 2011/46.

Cortisol levels were quantified using three indices: the average increase (MnInc), the excretion of cortisol after waking period (AUCtrab) and area under the curve from zero or base of the diurnal cycle (AUCCD).

The MnInc is related to the CAR, which has been shown to be independent of the level of cortisol in general and thus provide independent information about the functioning of the HPA axis5. It was calculated based on the values ​​of the second sample minus the first.

The excretion of cortisol after waking period was calculated from the area under the curve for the samples collected during the duty (12 a.m. and 6 p.m., respectively) (AUCtrab). Also, calculation of AUCCD was held based on all measurements performed on the collection day. These areas were calculated through the formulas presented by Pruessner6, using the time in hours.

Other measurements were the minimum and maximum values, mean and standard deviation of each measure of salivary cortisol.

To evaluate the statistical association between the covariates and the rates of salivary cortisol, means were compared and proceeded to bivariate analysis using the Mann Whitney and Kruskal-Wallis tests, assuming that cortisol levels did not achieve normal distribution. For all analyzes the level of significance was of 5% (p <0.05).

The process of data analysis used the Statistical Package for the Social Sciences version 18 (SPSS®).


It is observed that nursing workers surveyed are predominantly female (82.5%) with a mean age of 41.8 years old (± 10). Still, 55.3% of the professionals have no partner, and 89.3% do not have children under 6 years old. The income ranged between R$ 2,001.00 and R$ 4,000.00, which represents between 3.2 to 6.4 times the minimum wage (currently R$ 622.00). Regarding education, 65% of them do not have degrees; 47.6% have other training beyond that required by the position, and 59% attended graduate school.

Of the population interviewed, 66% are technicians and nursing assistants. The majority (89.3%) is on duty and 100% of employees meet 24 hours/week.

Considering the health-related habits, we observed that 64.7% of the population is overweight or obese; 88.3% are non-smokers. The pattern of alcohol consumption was considered high risk for 41.7% of the subjects. On physical activity, 66% of respondents report not performing any physical activity and 87.4%, however, claim to perform leisure activities. Domestic overload was assessed as low for 60.2% of workers. With regard to sleep, 51.5% reported having poor quality sleep.

Measures of salivary cortisol were performed at four different times of day: upon waking, 30 minutes after waking, at 12 a.m. and at 6 p.m. The average time for the 103 individuals awakening was 5:16 a.m. ± 4.0 minutes and the average time to collect the second saliva sample was 5:47 a.m. ± 4.0 min.

Of the 103 participants, 27 exhibited CAR less than 2.5 nmol/L of cortisol levels measured 30 minutes after awakening, compared to the levels immediately upon awakening, so they were found non-adhering to the protocol and excluded from all further analysis after the test procedure.

Since samples pertaining to schedule of 12 a.m. had been collected by interviewers, they had no problems with adherence to the protocol, being a total of 103 workers. Regarding the day laborers, whose work schedule ended at 1 p.m., and others who, for some unknown reason, were absent from duty, it was not possible to collect the last sample, referring to 6 p.m. Therefore, the final sample at 6 p.m. was 80.

The mean cortisol observed upon awakening, 30 minutes after, at 12 a.m. and  at 6 p.m. was 5.82 nanomolar per liter (nmo /L) (± 4.86 nmol/L), 16.60 nmol/L (± 8.31 nmol/L), 7.49 nmol/L (± 6.97 nmol/L) and 3.93 nmol/L (± 3.15 nmol / L), respectively. The increase in cortisol between awakening and 30 minutes after was 64% on average. See Figure 1.

Figure 1

Regarding the adopted cortisol rates, it is observed the average of MnInc, AUCtrab and AUCCD of 10.78 nmol/L (± 6.99 nmol/L), 32.51 nmol/L (± 21.99 nmol/L) and 107.99 nmol/L (± 61.63 nmol/L), respectively.

Characterization of salivary cortisol levels, according to sociodemographic characteristics showed that MnInc was associated with marital status and income (p <0.001 and p <0.05, respectively), and professionals without a partner and earning up to R$ 2000.00 have higher averages for MnInc. It was also observed that among professionals with other training beyond that required for the position, this was graduate degree, and when not possessed child under 6 years old, MnInc average was higher, but without statistical significance.

Related to AUCtrab, only sex presented statistically significant association (p <0.01). Men revealed AUCtrab with the highest average. However, there is the interesting behavior of some variables, although without statistical significance, most likely by the high standard deviation: workers aged between 20 and 36 years old, who had post-graduation, as well as the variable for the item other training beyond that required for the position, without a partner, with child under 6 years old and who earn up to R$ 2,000.00 had higher AUCtrab average.

Regarding AUCCD, only marital status showed a statistically significant association (p <0.01). Professionals without partner revealed highest AUCCD average. Other variables draw attention, but without statistically significant associations. The highest AUCCD average was checked on the condition of age between 20 and 36 years old, male, undergraduate, who have no other training, beyond the required for the position, but when possessed, it was post-graduation, without child under six years old and received up to R$ 2,000.00.

The analysis of rates of salivary cortisol (MnInc, AUCtrab and AUCCD) showed no statistically significant associations with any variable related to health habits and occupational characteristics.


Regarding saliva cortisol concentration, the results show that it was high at the beginning of the day with spike 30 minutes after waking, followed by progressive reduction in the afternoon. This pattern is typical of the release of cortisol during the day7.

Few data exist on normal parameters of cortisol levels measured for a particular group of people8. Although there are several studies regarding variations of cortisol in workers exposed to stress, little have been suggested on the normal or expected functioning of the HPA axis in this group.

To review and discuss the results of the rates of salivary cortisol, they were separated in that which  made reference to CAR (MnInc) and that which reflected in the behavior of cortisol during the rest of the day and diurnal cycle (AUCtrab and AUCCD). The rationale for this division is that different factors may influence the two phenomena. A study indicated that cortisol response to the wake is partly under genetic control, whereas cortisol levels of the rest of the day are not9. Mineralocorticoid receptors in the central nervous system seem to be responsible for the regulation of circadian rhythms of activity of the HPA axis, while glucocorticoid receptors mediate responses to stress and other disorders10.

A study of nurses working in a clinic in Germany11, observed mean values ​​of cortisol upon awakening, 30 minutes later and values of CAR rates evaluation quite similar to a German study9. Other research has shown that the average salivary cortisol upon awakening was 19.3 nmol/L, increasing to 28.6 nmol/L after 30 minutes12.

These averages are higher than those found in this study (5.82 upon awakening and 16.60 nmol/L after 30 minutes). Similar results were found in a study of workers in the underground rail system in London13, the mean values ​​of cortisol upon awakening ranged from 4.4 to 7.8 nmol/L, at different conditions of work shifts. Previous studies with the RIA method used in this experiment, reported mean values ​​ranging between 4.7 to 7.3 nmol/L14,15, while studies involving immunofluorescence assay values ranged between 11 to 18 nmol/L6,11,16.

Considering the average increase, a study noted an increase in cortisol between waking and 30 minutes later of 70% on average, which is within the range previously observed, and also consistent with that seen in this study (64%)5. Other research has shown that the typical increase in cortisol between waking and 30 min after in men was 23.5% and for women 21.35%17. Related to the increase of cortisol 30 minutes after waking, a study had average of 114.1%, higher values than the ​​observed in previous studies (average 48.3 to 68.0%)18.

Some participants in this study had absolute values ​​of cortisol upon waking particularly low (under 0.05 nmol/L). These results may be related to the time at which the participants woke up (between 3:30 and 05:00 a.m.), since a similar pattern was previously reported19. In terms of AUCtrab and AUCCD rates, it is noted, in this study, an average of 32.51 (± 21.99) and 107.99 (± 61.63) nmol/L, respectively. In a study with call center operators, the AUCtrab found was 6416.77 and the diurnal cycle of 11684.07 nmol/L (both calculated with time in minutes)17. In a research conducted with 82 nurses in a pediatric critical unit, it was obtained a AUCtrab average of 3,85nmol/L (log transformed, and calculated per hour)20.

The evaluation of AUCtrab aimed to capture the salivary cortisol during the work, in which stressing factors could be experienced, and their impact on the physiological level, could be captured. The cortisol secretion is episodic and pulsed; the amount of cortisol can increase within minutes following a stressful experience. However, few studies have evaluated the variation of cortisol as a biomarker of stress during the day, or when they do it, they assess the stress response before and after a stressful event, which differs from the methodology adopted in this study.

It was clear the discrepancy between the absolute values ​​of salivary cortisol in the studies, as well as the rates used for evaluation. A plausible explanation for these discrepancies may be related to the different methods of analysis: in this study, salivary cortisol samples were analyzed by the RIA method, while a number of other analytical techniques have been used in different studies, complicating comparisons.

The high standard deviations of salivary cortisol indicators reflect a great individual variability and point to the need for studies and research that emphasize the unique aspects of each worker.

In fact, the results presented here, when compared to studies using similar methodology, have lower absolute values ​​of cortisol upon awakening and 30 minutes later, but with similar CAR rate (MnInc), and some cases even higher. This shows a preserved HPA axis functioning and not reduced as observed in situations of chronic stress.

It was found that for the CAR (MnInc) rate; only the variables marital status and income were associated with statistical significance. As for the AUCtrab, the gender variable was significant and for AUCCD, the variable marital status was significant.

The pathways through which lower socioeconomic levels cause these physiological changes are still not well understood21. Among the factors that compose this assessment is the income. In the present study it can be observed that professionals with incomes up to R$ 2,000.00 showed the most pronounced MnInc, in other words, a more pronounced reactivity of the HPA axis.

Research has shown that higher levels of secretion of complete cortisol in the morning and in the post-awakening period (AUCtrab) were positively associated with a higher level of education22. This finding contrasts with previous results21-24 which reported an inverse association between socioeconomic statuses, measured in terms of income, education or occupation, and cortisol levels.

Another study shows that levels of salivary cortisol in individuals with higher education were lower in the morning and evening and was significantly different at 8 a.m. (p = 0.006) and at 10:30 p.m. (p = 0.03), respectively25.

In relation to marital status, it is noted that those with no a partner obtained the highest CAR evaluation rates, as well as the AUCCD. A similar result was observed in a study where the diurnal cycle levels (AUCCD) were significantly associated with marital status (not married individuals excreting more cortisol than married individuals) (average: 13,476 VS 1.0733 nmol/L, p <0.05)26. There is not, however, an adequate explanation for this association, since there are no investigations to explore the relationships between the production of total daily cortisol and this variable. For these researchers, all levels of cortisol excretion showed specific associations for gender: CAR levels were significantly higher in women than in men, and a difference between the sexes was also observed for the other rates of cortisol, even if not significantly.

Maintaining higher levels of cortisol during the working day in men may reflect an increased pattern of responsiveness of the HPA axis before an acute challenge, as described in the experimental literature27.

In the present study, no evidence of activation of cortisol sustained among women was found, even when comparisons were made between women who claimed not having children and those with children and with high domestic workload.

One must consider, however, these results with caution, since the present study, similar to what occurs with the nursing category as a whole, was carried out with a limited number of men. Furthermore, our measurements do not take into account hormonal differences (menstrual cycle) that may have contributed to the observed difference between groups of cortisol reactivity. Recently, researchers have documented a higher CAR in women28, in particular, during ovulation29. In contrast, previous cross-sectional studies found no differences in the CAR profile between the follicular and luteal phases in healthy women30.

Some limitations were considered for this research. It is a cross-sectional study, which provides an instantaneous or short duration image of the relationship that was assessed, restricting inferences about associations between psychometric measures and cortisol levels.

The adoption of numerous exclusion criteria for participation in the study, on the one hand brought greater control of known and potential confounders or effect modification, on the other hand reduced the number of subjects, which also contributed to the limitation of the method.

Another limitation was not having researched caffeine consumption by participants. Coffee is a strongly activating and stimulating drink that may increase cortisol levels3.


The salivary cortisol is an extremely useful biomarker for psychobiological stress research and the advent of this method facilitated several important studies in recent decades, indicating a significant part of the function of the target system, the HPA axis. But obviously the salivary cortisol measurements may not perfectly reflect the remarkable complexity of this whole system. This is a matter of conceptual nature which may explain why studies, including this one, have partially inconsistent results.

The results provide an interesting contribution to understanding the biological pathways by which work stress influences health. It is highlighted the possibility of replicating this study in other units of different sizes and specialties. Results also indicate the need for further research on the same subject, considering the aspects that were not addressed in this study, such as the inclusion of other confounding variables, different methodologies for assessing exposure to stress, using the principle of triangulation, studies of experimental nature.


1. Karasek RA, Theorell T. Healthy work-stress, productivity, and the reconstruction of working life. New York: Basic Books; 1990.

2. Pacak K, McCarty R. Acute stress response: experimental. In: Fink, G. Encyclopedia of Stress. New York: Academic Press; 2000. p. 8-17.

3. Kudielka BM, Gierens A, Hellhammer DH, Wust S, Schlotz W. Salivary cortisol in ambulatory assessment - some dos, some don’t, and some open questions. Psychosom Med. 2012; 74: 418-31.

4. Oliveira EB, Souza NVM, Chagas SCS, Lima LSV, Correa RA. Esforço e recompensa no trabalho do enfermeiro residente em unidades especializadas. Rev enferm UERJ. 2013; 21: 173-8.

5. Clow A, Thorn L, Evans P, Hucklebridge F. The awakening cortisol response: methodological issues and significance. Stress. 2004; 7: 29–37.

6. Pruessner M, Hellhammer DH, Pruessner JC, Lupien SJ. Self reported depressive symptoms and stress levels in Healthy Young Men: associations with the cortisol response to awakening. Psychol Med. 2003; 65: 92–9.

7. Clements AD. Salivary cortisol measurement in developmental research: where do we go from here? Dev Psychobiol. 2013; 55: 205-8.

8. Patel RS, Shaw SR, Macintyre H, McGarry GW, Wallace AM. Production of gender specific morning salivary cortisol reference intervals using internationally accepted procedures. Clin Chem Lab Med. 2004; 42: 1424–9.

9. Wust S, Wolf J, Hellhammer DH, Federenko I, Schommer N, Kirschbaum C. The cortisol awakening response-normal values and confounds. Noise Health. 2000; 2: 79–88.

10. De Kloet ER, Vreugdenhil E, Oitzl MS, Joëls M. Brain corticosteroid receptor balance in health and disease. Endocr Rev. 1998; 19:269 –301.

11. Federenko IS, Wüst S, Hellhammer DH, Dechoux R, Kumsta R, Kirschbaum C. Free cortisol awakening responses are influenced by awakening time. Psychoneuroendocrinology. 2004; 29: 174-84.

12. Steptoe A, Siegrist J, Kirschbaum C, Marmot M.  Effort–reward imbalance, overcommitment and measures of cortisol and blood pressure over the working day. Psychosom Med. 2004; 66: 323–9.

13. Williams E, Magid K, Steptoe A. The impact of time of waking and concurrent subjective stress on the cortisol response to awakening. Psychoneuroendocrinology. 2005; 30: 139-48.

14. Brooke-Wavell K, Perrett LK, Howarth PA, Haslam RA. Influence of the visual environment on the postural stability in healthy older women. Gerontology. 2002; 48: 293-7.

15. Edwards S, Evans P, Hucklebridge F, Clow A. Association between time of awakening and diurnal cortisol secretory activity. Psychoneuroendocrinology. 2001; 26: 613–22.

16. Schmidt-Reinwalds A, Pruessner JC, Hellhammer DH, Federenko I, Rohleder N, Schürmeyer TH et al. The cortisol response to awakening in relation to different challenge tests and a 12- hour cortisol rhythm. Life Sci. 1999; 64: 1653–60.

17. Maina G, Palmas A, Filon FL. Relationship between self-reported mental stressors at the workplace and salivary cortisol. Int Arch Occup Env Health. 2008; 81: 391–400.

18. Il-Young S, Ryun-sup A, Sae-il C, Young-jin L, Min-soo K, Chea-kwan L, Simon S. Cortisol awakening response and nighttime salivary cortisol levels in healthy working korean subjects. Yonsei Med. J. 2011; 52: 435-44.

19. Hucklebridge F, Mellins J, Evans P, Clow A. The awakening cortisol response: no evidence for an influence of body posture. Life Sci. 2002; 71: 639-46.

20. Metzenthi P, Helfricht S, Loerbroks A, Terris DD, Haug HJ, Subramanian SV et al. A one-item subjective work stress assessment tool is associated with cortisol secretion levels in critical care nurses. Prev Med. 2009; 48(5): 462-6.

21. Kunz-Ebrecht SR, Kirschbaum C, Marmot M, Steptoe A. Differences in cortisol awakening response on work days and weekend in women and men from Whitehall II cohort. Psychoneuroendocrinology. 2004; 29: 516–58.

22. Maina G, Bovenzi M, Palmas A, Rossi F, Filon FL. Psychosocial environment and health: methodological variability of the salivary cortisol measurements. Toxicol Lett. 2012; 213: 21– 6.

23. Cohen S, Schwartz JE, Epel E, Kirschbaum C, Sidney S, Seeman T. Socioeconomic status, race and diurnal cortisol decline in the coronary artery risk development in young adults (CARDIA) study. Psychosom Med. 2006; 68: 41–50.

24. Li L, Power C, Kelly S, Kirschbaum C, Hertzman C. Life-time socioeconomic position and cortisol patterns in mid-life. Psychoneuroendocrinology. 2007; 32: 824-33.

25. Hong R.H, Yang YJ, Kim SY, Lee WY, Hong YP. Determination of appropriate sampling time for job stress assessment: the salivary chromogranin A and cortisol in adult females. J Prev Med Public Health. 2009; 42:231-6.

26. Maina G, Palmas A, Bovenzi M, Filon FL. Salivary cortisol and psychosocial hazards at work. Am J Ind Med. 2009; 52: 251–60.

27. Kirschbaum C, Kudielka BM, Gaab J, Schommer NC, Hellhammer DH. Impact of gender, menstrual cycle phase, and oral contraceptives on the activity of the hypothalamic-pituitaryadrenal axis. Psychosom Med. 1999; 61: 154–62.

28. Vargas I, Lopez-Duran N. Dissecting the impact of sleep and stress on the cortisol awakening response in young adults. Psychoneuroendocrinology. 2014; 40: 10-6.

29. Wolfram M, Bellingrath S, Kudielka BM. The cortisol awakening response (CAR) across the femle menstrual cycle. Psychoneuroendocrinology. 2011; 36: 905-12.

30. Bouma EM, Riese H, Ormel J, Verhulst FC, Oldehinkel AJ. Adolescents cortisol responses to awakening and social stress; effects of gender, menstrual phase and oral contraceptives: The TRAILS study. Psychoneuroendocrinology. 2009; 34: 884–93.

Direitos autorais 2014 Juliana Faria Campos, Helena Maria Scherlowski Leal David

Licença Creative Commons
Esta obra está licenciada sob uma licença Creative Commons Atribuição - Não comercial - Sem derivações 4.0 Internacional.