Stress is defined as ''a state of mental or emotional strain or tension resulting from adverse or very demanding circumstances'' [1]. In today’s world, we are hearing the word “stress” quite frequently in our daily lives and conversations. Stress comes out in different forms that can change from one person/society to another. The physiological results of stress are regulated by the central nervous system (CNS) through the stimulation of the hypothalamic-pituitary-adrenal (HPA) axis [2]. As a reaction to the source of stress, there happens an increase in the secretion of particularly cortisol and also of glucocorticoids [3]. Symptoms related to stress can be classified into four main groups such as physical, emotional, mental, and social [4]. The Adrenocortex Stress Profile (ASP) ensures the evaluation of the hypothalamus-pituitary-adrenal (HPA) axis employing saliva samples that are meticulously and timely taken and clinically best represent the cortisol and DHEA hormones [5]. Salivary testing is an easy and non-invasive option to measure unbound and biologically active parent/main hormone levels [6]. Daily problems, chronic pains, blood sugar irregularities, stress factors in workplaces, and weak social relations can change the HPA axis. Stress can be assessed through behavioral scales. Biological measurements are also used in the evaluation of stress. The most commonly used biological measurement is the examination of the salivary cortisol level obtained in a non-invasive manner. The secretion of cortisol in stress-free circumstances shows diurnal characteristics [7]. Cortisol levels peak during the early hours of the morning and drop to the lowest concentration at nighttime [8]. In case of stress, it is observed that the cortisol level rises independently of its diurnal rhythm [9]. Cortisol is defined as the biological indicator of psychological stress [10]. HCW’s occupation is the one that requires quite an important and heavy responsibility given that any mistake can lead up to the loss of human life. Therefore, healthcare is one of the occupational sectors, in which professionals are most likely faced with stress. It is reported that anxiety levels among HCW are higher than the general population due, among other things, to frequent night shifts, only a few hours of sleeping and heavy workload [11]. We think that upon that the World Health Organization (WHO) on March 11, 2020, declared the novel coronavirus (COVID-19) outbreak a global pandemic [12], stress rate increased in all people, first and foremost HCW to some extent. This study aims: (1) to determine whether there is a correlation between the level of anxiety detected by using “BAI” for HCW, who work at hospital at least 24 hours uninterruptedly, and their level of anxiety specified through examining the cortisol level in saliva, (2) to compare the level of stress in HCW at emergency departments, where the COVID 19 cases are come across by far the most, and HCW, who work at other medical departments. Through the groups formed within the scope of this study, we come up with the results regarding developing the ways to deal with stress employing determining the stress level.

For saliva samples, “Salivatte” branded “specific saliva samples collection tubes” produced by Sarstedtcompany (SarstedtAG@Co D 51582 Nümbrect) were used. Samples were collected from HCW, who work at hospital 24 hours uninterruptedly. They were informed to gargle their mouth with enough water 20 minutes before collecting samples and then not to eat, drink, or smoke. And, they were asked to take the cotton from the specific tubes on their own between 10:00 and 11.00 in the evening and between 07.00 and 08.00 in the morning, to chew it for about 2-3 minutes until they make sure that it gets adequately saturated with saliva and then to put it back to the tube. The tubes were immediately transferred to the laboratory. Following that the samples were centrifugated at 1000 g'' for 3 minutes, saliva samples at the bottom of the tubes were carefully portioned with Eppendorf tubes and they were frosted at minus 80 degrees Celsius till the day they were examined. After collecting all the samples belonging to groups that were covered in the study, samples were defrosted and they were studied by using Roche Diagnostic’s “Cortisol Kit” (Roche Diagnostic GmBH Sandhofer Strasse 116 D 68305 Manheim lot 6687733) and through “Electrochemiluminescence (ECLIA)” method in “Cobas601” branded “hormone analyzer device” produced by Roche Diagnostic company. Simultaneously, HCW’s anxiety levels for the last week were measured through BAI, which is a self-report inventory developed by Beck et al. used to determine the frequency of anxiety symptoms experienced by individuals. It is a Likert type scale composed of 21 questions, each of which has a point between 0 and 3. There are four choices in each of the 21 symptom categories. There are four anxiety levels indicated by BAI depending on certain thresholds: normal anxiety: 0-7 points, mild anxiety: 8-15 points, moderate anxiety: 16-25 points, and severe anxiety: 25-63 points. The higher is the BAI score, the more severe is the level of anxiety experienced by the individual. Its validity and reliability study was done in Turkey [13-14].

Ethics statement

To protect the respondents' privacy, the survey was conducted anonymously. Ethical approval for the research was obtained from Taksim Education and Research Hospital (Certificate e 2020-62).

 Statistical analysis

The values that were used in descriptive statistics of data set are mean/average, standard deviation, median (minimum and maximum), frequency, and ratio. Kolmogorov-Smirnov test was utilized to analyze whether the data were distributed normally or not. Mann-Whitney test was used for the analysis of independent qualitative data. Chi-Square test was employed in analyzing independent quantitative data. In case that conditions were not met for the Chi-Square test, the Fischer test was utilized. Spearman correlation analysis was used in the correlation analyses. SPSS 26.0 program was utilized in all analyses. The limit of significance was taken as a p-value <0.05 and in dual direction. P-value <0.05 was considered significant. Data were analyzed by using IBM SPSS (Statistical Package for Social Science) Version 26. Regarding the size of samples, the 95% confidence interval (CI) was calculated as 76 persons by the G-Power 3.1.7 Package Program (Heinrich-Heine-Universitat, Düsseldorf, Germany).

BAI was applied to HCW consisting of doctors and nurses. Through looking at such variables as occupational groups of HCW, several years spent in the profession, their habits and the correlation between stress and anxiety measured by BAI, we envisaged that (1) HCW, who either work at emergency departments or have relatively less professional experience, can be more stressful and (2) BAI score will be low for those HCW whose morning saliva cortisol values are lower.

98 persons composed of doctors and nurses working at Gaziosmanpasa Education and Research Hospital participated in this study. However, 8 persons were taken out of the study because 6 of them couldn’t give the samples properly, one of them had a chronic disease that could affect the cortisol level and was taking medicine, and one of them didn’t complete BAI. As a result, 90 volunteer HCW, who work on 24-hour shifts, were included in the study through their consent. While 26.7% (n=24) of HCW were smoking, 73.3% (n=66) were not smoking (Table 1).

There are two groups (Group A and Group B) covered in this study. While Group A refers to the group whose evening cortisol level is higher than or equal to that of morning, Group B consists of the people whose evening cortisol level is lower than that of morning.

Group A and Group B did not show a significant difference in terms of HCW’s age and gender distribution(p>0.05), BMI value, professions and professional durations, chronic disease rates, and BAI scores. The rate of smoking in Group A was significantly higher than that of Group B. The rate of HCW working in the emergency department was significantly higher in Group A than in Group B (p<0.05) (Table 2).

HCW’s morning and evening cortisol values varied from 0.05 to 1.68 mg/dl (0.30 ± 0.28 on average) and from 0.05 to 0.28 mg/dl (0.09 ± 0.05 on average), respectively. BAI score was measured 0-46 (11.2 ± 10.3 on average). Morning and evening cortisol values of females and males did not show a significant difference (p>0.05). The BAI score of females was significantly higher than that of males (p<0.05). The morning cortisol value of HCW working at the emergency department was significantly lower than that of HCW working at other departments (p<0.05). Evening cortisol value of HCW working at the emergency department was significantly higher than that of HCW working at other departments (p<0.05). BAI score did not show a significant difference for HCW working either in the emergency department or not (p>0.05). The morning cortisol value of doctors was significantly higher than that of nurses (p<0.05). Evening cortisol value did not display significant difference in terms of professional groups (p>0.05). BAI scores did not differ significantly in terms of professional groups (p>0.05) (Table 3). Morning and evening cortisol levels did not show a significant difference in terms of having a chronic disease or not (p>0.05). BAI scores of HCW with chronic disease were significantly higher than that of those not having a chronic disease (p<0.05). Non-smokers’ morning cortisol level was significantly lower than that of smokers (p>0.05). Evening cortisol level did not show a significant difference between smokers and non-smokers (p>0.05). BAI scores of smokers and non-smokers did not differ significantly (p>0.05) (Table 3).

51.1% (n=46) of the HCW, who participated in the study, were female, and morning and evening cortisol levels did not show a significant difference between females and males (p>0.05). However, BAI score of females was significantly higher than that of males (p<0.05). The higher is the BAI score, the more severe is the level of anxiety experienced by the individual. In our study, there wasn't found any significant difference between a change in cortisol level and BAI score (p>0.05). The cortisol value in saliva is more objective than the BAI score. In our study, Salivary cortisol value can be evaluated more reliably. This contrasts with the high stress coefficient in female gender in previous studies.

In the study conducted by González-Sanguino et al. it was found, concerning the variables related to psychological impact, that the female gender is associated with greater depressive symptoms, anxiety, and post-trauma stress disorder (PTSD) [15]. Likewise, given that the health of family and children is given much more importance in traumatic incidents such as the COVID 19, the finding of a high level of stress among females in our study should be perceived as normal.

As regards the relationship between saliva cortisol and BMI, while the study by Abraham et al. showed that saliva cortisol is inclined to rise in case of high BMI, the other study carried out in the same year by Champaneri S et al. revealed that there is not a significant correlation between cortisol and BMI [16-,17]. In parallel to the finding of the latter study, any significant correlation was not found between BMI and the increase in cortisol level in our study.

Working environment and conditions can also influence the stress level. Compared to the general population, it is three times more likely for HCW working at the emergency department to experience major depression (MD), PTSD, and physical health risks related to these factors [18]. An increase in the level of stress together with the lack of sleep and long working hours can put the proper treatment of patients in jeopardy [19]. In the study carried out by Pereira-Lima and Loureiro, it was stated that "anxiety levels of HCW are higher than the general population because, among other things, of frequent night shifts, only a few hours of sleep and heavy workload" [11]. This finding is in line with what was found in our study. The study carried out by Seunghyeon et al. showed that DHEAS levels of nurses at the emergency department were higher than nurses working at other departments [20]. In our study, morning cortisol value of HCW working at the emergency department was significantly lower than that of HCW working at other departments; however, it did not show any correlation with BAI score.

 Gaziosmanpasa Education and Research Hospital’s Emergency Department receives 1000-1200 patients on average per day. It can be said that the emergency department's HCW are used to and prepared for a heavy workload and they have even moved away from stress thanks to a decrease in their workload. A study by Cai et al., which is supportive of the findings of our study, showed that “people without public health emergency treatment experience showed worse performance in mental health, resilience and social support, and tended to suffer from a psychological abnormality on interpersonal sensitivity and photic anxiety. This finding suggested that high levels of training and professional experience, resilience, and social support were necessary for HCW who are first taking part in public health emergence” [21]. A study by Backé et al. revealed that the ambulance crew appears to be used to critical situations. It also showed, as regards the physiological stress response, that there is a little awareness indicating that stress is probably not perceived in work situations characterized by routine [22].

The salivary cortisol normal reference range provided by the Roche laboratory manufacturer and determined in samples from 154 healthy individual swabs: 8 AM - 10 AM: <0.69 μg/dL (<19.1 nmol/L); 2:30 PM - 3:30 PM: <0.43 μg/dL (<11.9 nmol/L).[23] Another study that covered 127 subjects, among which 57 were healthy volunteers, 39 were persons suspected of Cushing’s syndrome and 31 were persons with proven Cushing’s syndrome, showed that 2.5^th-97.5^th percentile of the late-night salivary cortisol concentrations in normal subjects was 0.054 to 0.1827 μg/dL [24]. The saliva samples of 90 HCW covered in our study were processed through the devices produced by the same company, it was seen that 07.00-08.00 AM cortisol values were 0.05-1.68 μg/dL, 10.00-11.00 evening cortisol values were 0.05-0.28 μg/dL and they were higher than the normal/reference values.

As for the relationship between stress and occupational groups, it was found in a study that among healthcare professional groups with the highest exposure to stress, nurses are the most affected by occupational stress. A study by Huang et al. revealed that "in the COVID-19 epidemic, the incidence of anxiety and stress disorder is high among medical staff. In the "Post-Traumatic Stress Disorder Self-Rating Scale (PTSD-SS), the incidence of anxiety in nurses was higher than in doctors and the score of Self-rating Anxiety Scale(SAS) of nurses was higher than that of doctors” [25]. In our study, although doctors, whose morning cortisol values were measured significantly higher than that of nurses (p<0.05), seemed to be more stressed than nurses during COVID-19 outbreak, this was not confirmed/supported by the BAI scores of doctors and nurses because they did not differ significantly regarding the occupational groups (p>0.05). Therefore, depending on these findings, it cannot be asserted for sure that doctors were more stressed than nurses or vice versa.

 In our study, morning and evening cortisol values did not show a significant difference for those having a chronic disease or not (p>0.05). BAI score of those having chronic disease was significantly higher than that of those not having a chronic disease (p<0.05). The study by Yahia et al. showed that patients with cardiovascular disease were at greater risk of developing COVID-19, especially in its severe form. These patients were five to ten times more at risk of death. These occurred in 7 to 17% of hospitalized patients. The presence of a new heart lesion in patients with COVID-19 was consistently associated with a poor prognosis [26]. It was found in other studies that individuals with diabetes mellitus (DM), hypertension, and severe obesity (BMI ≥ 40 kg/m²) are more likely to be infected and are at a higher risk for complications and death from COVID-19 [27,28]. As also put forward in these studies, it can be said that HCW with chronic disease are under more stress given that they bear more health risks during the COVID-19 outbreak.

In the study by Lyu et al. “no significant correlation was found between housing ownership, smoking, sleep duration, and perceived stress" [29]. In our study, morning cortisol value of smokers was significantly lower than that of non-smokers (p<0.05). BAI scores did not differ significantly for smokers and non-smokers (p>0.05). In this vein, it can be put forward that it is wrong to see the cigarette as a means to deal with stress; indeed, it affects the quality of life negatively because of the morbidity it causes.

We envisaged in our hypothesis that HCW, who either work at emergency departments or have relatively less professional experience, can be more stressful and the BAI score will be low for those HCW whose morning saliva cortisol values are lower. Because they have a higher ability to cope with intensity, emergency department’s HCW were less affected by stress. BAI scores of HCW in our study varied from 0 to 46 (11.2 ± 10.3 on average). This showed that even if there was not a difference regarding the BAI score between the emergency department and other departments, all HCW had a mild level of stress. It was seen that stress did not decrease as the professional experience increased; rather, it showed parallelism with the new-comers in the profession. This led us to think that it could be because experienced HCW assume more responsibility during the COVID-19 outbreak.

This study showed us that doctors and nurses working at such relatively busier departments as emergency departments etc. gained experience towards stress and they were much more prepared than those working at other departments to cope with stress during the COVID 19 outbreak. While some HCW were less affected by stress, some others were impacted by stress at a higher degree. Stress can vary from one individual/occupation to another. Particularly HCW have a high level of stress and they need more psychological support when they get under the psychological effect of COVID-19. HCW should be prepared for such catastrophes as the COVID-19 through simulation-intensive in-service training.

We thank Dr. Cihan Ekinci for his contribution in collecting saliva samples.

MC designed this study and made additional contributions to its design. MC, NC, and NU conceived and conducted statistical analyses, with additional advice regarding analyses. MC drafted the manuscript and approved the final manuscript.

This study was funded by MC, NA, and NU.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.