Stress, Anxiety, Depression among Nurses Caring for COVID-19 Patients in Babol, Iran: A logestic Regression

Introduction: Nurses who are involved in the caring of COVID-19 patients, are at risk of mental distress. The present study was conducted with the purpose of demonstrating the relationship between stress, anxiety, depression and characteristics of nurses who provide care to COVID-19 patients. Methods: This descriptive cross-sectional study was conducted using the census method from May to June, 2020 on 224 clinical nurses who were working in hospitals aliated with Babol University of Medical Sciences and were involved in caring for COVID-19 patients. The data collection instruments included the Depression, Anxiety and Stress Scale - 21 items (DASS-21) and the individual characteristic questionnaire. Bivariate and multivariate binary logistic regressions were computed to identify the associated factors. Result: The mean and the standard deviation for stress, anxiety and depression scores were 9.47±7.30, 9.29±7.51 and 8.84±7.22 respectively. 17.4% had stress, 54% had anxiety and 43% had a degree of depression. There were signicant relationships between the nurses’ stress level and characteristics including age (OR =3.009, 95%CI 1.46–6.16, P=0.003), having children (OR=0.26, 95%CI 0.11-0.63, P=0.003), work experience(OR=4.50, 95% CI 2.17-9.96, P=0.000) and employment status(OR=0.39, 95% CI 0.16-0.95, P=0.04). Moreover, along with these characteristics, job satisfaction (OR =3.03, 95%CI 5.64– 1.63, P=0.000), level of physical activities (OR =0.26, 95%CI 0.08–0.82, P=0.02), exercising (OR =2.27, 95%CI 1.31-3.90, P=0.003) and violence in the workplace (OR =0.27, 95%CI 0.12– 0.56, P=0.001) also had signicant relationships with the nurses’ anxiety level. Furthermore, the relationships between the nurses’ level of depression and characteristics including age(OR


Study setting, design, population, and sampling technique
This descriptive cross-sectional study was conducted from May to June, 2020 using census method on all clinical nurses working in two hospitals a liated with Babol University of Medical Sciences (Ayatollah Rouhani and Shahid Yahya Nejad) in northern Iran who were involved in direct care of COVID-19 patients.
Among the 302 nurses who directly involved in the care of Covid-19 patients, 224 completed the questionnaire. The inclusion criteria were: having a bachelor's or master's degree in nursing, being directly involved in providing care for COVID-19 patients in one of the wards of the two aforementioned hospitals, at least one year experience of working in the hospital and willingness to participate in the study.

Data collection procedure and tools
The research instruments used in this study included the individual characteristics questionnaire and the short form version of Depression Anxiety Stress Scales (DASS-21). The individual characteristics questionnaire had 18 items comprising age, gender, marital status, level of education, work experience, type of work shift, level of job satisfaction, average amount of overtime, employment status, type of hospital, having children, level of physical activity, exercising, history of chronic diseases, workplace violence, sleep and history of depression in the family. DASS-21 is a self-report instrument that shows the recent mood swings and was rst introduced by Lovibond in 1995 (14). As the name suggests, it has 21 items which are scored on Likert scales. These 21 items contain: 7 questions related to stress including nding it hard to wind down, tendency to over-react to situations, using a lot of nervous energy, getting agitated, nding it di cult to relax, being intolerant of anything that keeps you from getting on with what you were doing, and feeling touchy; 7 questions related to anxiety including dryness of the mouth, breathing di culty, trembling, being worried about situations in which you might panic and make a fool of yourself, feeling close to panic, action of the heart in the absence of physical exertion, and feeling scared without any good reason; and nally 7 questions related to depression including inability to experience any positive feeling at all, nding it di cult to work up the initiative to do things, feeling that you have nothing to look forward to, feeling down-hearted and blue, being unable to become enthusiastic about anything, feeling that you are not worth much as a person, and feeling that life is meaningless. The validity and reliability of this instrument have been reviewed and approved by Mehdipour and Naja (15,16). The internal consistency of the scale was determined in the study of Shamsaie et al. (2018) using Cronbach's alpha and the gures for each subscale were as follows: 85% for stress, 86% for anxiety and 83% for depression (17).
After de ning the purpose of the study and reassuring the participants that all information will remain con dential, the researcher distributed the questionnaires and gave explanation in case any of the items were unclear. The participants had to read each statement and rate the intensity of the symptom mentioned in it on a 4-point Likert scale as it had been experienced over the past week. In this regard, zero meant that the symptom had not been experienced at all, one meant it had been experienced to some extent, two meant it had been experienced rather much, and three meant it had been experienced very much. The total score of each subscale of this instrument is obtained by summing up the scores of the questions related to that subscale.
Since DASS-21 is the short-form version of the main scale (42-item), the total score of each subscale must be doubled (16). Then, the intensity of the signs and symptoms can be determined according to Table1. Data processing and analysis Data analysis was performed using SPSS statistical software version 18. The correlations between independent and dependent variables were tested by binary logistic regression analysis. The variables which had a p-value less than 0.2 were entered into multivariate analysis. In the end, the model was nalized using backward elimination. Finally, a p-value of less than 0.05 was considered statistically signi cant association, and the adjusted odds ratio with 95% CI was calculated to determine the strength of association.

Data quality management
Regular monitoring of data collection was performed by the principal investigator to ensure that all necessary data were properly collected. In every stage of data collection, the lled questioners were cheeked manually rst for completeness and consistency then the collected data were processed timely and enter from a paper onto the computer twice. If a questionnaire is incomplete, it will be returned for completion, if the respondents were not willing to ll the missed information, the questionnaire was discarded from the analysis for gross incompleteness.

Ethical approval and consent to participate
The study protocol was approved by the Research Deputy of Babol University of Medical Sciences. Written informed consent was obtained from each study participant. The involvement of the study participants was voluntarily and participants were informed of the right to withdraw anytime from the study.

Results
Demographic Characteristics: The study sample contained 224 nurses with an average age of 32.9 ± 6.31. The demographic characteristics of the participants were as follows: 167 (74.6%) were female, 148 (68%) were younger than 50 years of age, 85% had bachelor's degree, 80% worked rotating shifts, 58% had 1 to 10 years of work experience, 67% were working under a formal contract, 80% were married, 59% had at least one child, 52% were working 50 to 100 hours of overtime per month, 67% said they were satis ed with their jobs, 90% had physical activities, 53% did not exercise regularly, 87% did not have any chronic diseases, 98% did not have any history of mental disorders, 90% had no history of depression in their families, 68% got 7 to 8 hours of sleep at least 3 to 4 days a week, and 78% did not experience workplace violence ( Table 2).

Level of Stress, Anxiety and Depression
The mean (± standard deviation) stress score calculated by DASS-21 was 4.73 (± 3.65), which was multiplied by two (to make it equal to the result of the main scale) and the result was 9.47 (± 7.30). More than 82% of the nurses had normal stress levels.
The mean (± standard deviation) anxiety score was 4.67 (± 3.75) based on DASS-21 and 9.34 (± 7.51) after multiplying it by two. More than half of the nurses (54%) had degrees of anxiety.
The mean (± standard deviation) depression score was 4.42 (± 3.61) as measured by DASS-21 and 8.84 (± 7.22) after it was doubled. About 43% of the nurses experienced some degrees of depression (Table 3).

Factors Associated with Stress
According to the results of the binary logistic regression variables including age, work experience, having children and employment status had signi cant relationships with the nurses' stress levels. In this regard, the nurses who were older than 35 years had three times greater risk of experiencing stress than the nurses who were younger than 35 years; the nurses who had more than 10 years of experience were 4.65 times as likely to feel stressed as the nurses who had less than 10 years of experience; the nurses who did not have children were 0.26 times less likely to get stressed than the nurses who had children; and the nurses who were employed in non-governmental hospitals were 0.39 times less likely to feel stressed than those employed in governmental hospitals (Table 4).

Factors Associated with Anxiety
According to the results of the binary logistic regression, there were signi cant relationships between the nurses' anxiety levels and the variables including age, work experience, having children, employment status, job satisfaction, level of physical activities, exercising and violence in the workplace. In this regard, the nurses who were older than 35 years had 2.61 times greater risk of suffering from an anxiety disorder than the nurses who were younger than 35 years; the nurses who had more than 10 years of experience were 3.25 times as likely to get anxious as those nurses who had less than 10 years of experience; the nurses who did not have children were less likely to feel anxious (OR = 0.45); the nurses who were employed in non-governmental hospitals or had an unclear employment status were less likely to experience anxiety than the nurses with a clear employment status (OR = 0.46); the nurses who were not satis ed with their jobs felt anxious 3.3 times more than the nurses who were satis ed with their jobs; and the nurses who were active in terms of physical activities were far less likely to get anxious than the nurses who were physically inactive (OR = 0.26). Similar to physical activity, the nurses who did not exercise experienced anxiety 2.27 times more than the nurses that exercised regularly. Moreover, the nurses who did not experience workplace violence were less likely to get anxious (OR = 0.27) ( Table 5).

Factors Associated with Depression
According the correlations between the research variables and the level of depression measured by the binary logistic regression, the variables including age, work experience, job satisfaction, exercising, having chronic diseases, workplace violence and sleep pattern were signi cantly associated with level of depression. In this regard, the nurses who were older than 35 years had 2.07 times greater risk of getting depression than the nurses who were younger than 35 years; the nurses who had more than 10 years of experience had 1.79 times greater risk of experiencing depression than the nurses who had less than 10 years of experience; and the nurses who were not satis ed with their jobs were 1.7 times as likely to get depression than the nurses who were satis ed with their jobs. Exercising was one of the factors that had a signi cant relationship with depression. The nurses who did not exercise had 1.76 times greater risk of getting depression than those who exercised. In addition, the nurses who suffered from a chronic disease were at an increased risk for experiencing depression (OR = 0.35). Moreover, workplace violence and sleep were among the factors associated with depression in nurses who provide care for COVID-19 patients. The nurses who had no violence in their workplace were less likely to get depression than the nurses who experienced workplace violence (OR = 0.39) and the nurses who got 7 hours of sleep less than 3 to 4 days a week were more likely to get depression (OR = 1.77) ( Table 6).

Multivariate Analyses
In the nal model and in the presence of all variables, ve variables including marital status, work experience, having children, job satisfaction and history of depression emerged as the signi cant stressrelated factors. Among these factors, work experience was the strongest variable related to stress in nurses involved with COVID-19, in a way that the nurses who had more than 10 years of experience had a 4 times greater risk of experiencing stress than those nurses who had less than 10 years of experience (OR = 4.5).
Moreover, in the nal model and in the presence of all variables, three variables including work experience, job satisfaction and violence in the workplace emerged as the factors associated with anxiety in nurses involved with COVID-19. Among these factors, again work experience had the strongest relationship with anxiety in nurses, in a way that the nurses who had more than 10 years of experience were 3.87 times as likely to experience anxiety as the nurses who had less than 10 years of experience(AOR = 3.87, 95% CI = 2.09-7.18).
Finally, regarding the factors associated with depression, six variables including gender, history of depression, exercising, sleep pattern, history of chronic diseases and violence in the workplace emerged in the nal model and in the presence of all variables. The nurses who did not have a history of depression had 0.27 times greater risk of depression than the nurses with a history of depression and the nurses who did not exercise were 2.04 times as likely to suffer from depression as the nurses who did exercise. Moreover, the nurses who got 7 to 8 hours of sleep less than 3 to 4 days a week were 2.28 times as likely to get depressed as the nurses who had a normal sleep pattern and the nurses who did not experience violence in their workplaces and had no history of chronic diseases had respectively 0.39 and 0.32 times greater risk of depression than the nurses who experienced workplace violence and had chronic diseases (Table 7). Nepal indicated that 38% of the healthcare workers involved in caring for COVID-19 patients suffered from anxiety or depression (18). A meta-analysis showed that the prevalence of anxiety and depression in healthcare workers who provide care for COVID-19 patients was 23.2% and 22.8% respectively (19). The studies conducted in other parts of the world indicate that the prevalence of anxiety is about 11.3 to 50% (21,4,20). The ndings of the present study report that it is slightly higher than this range. The high prevalence of anxiety in Iranian nurses could be attributed to the lack of protective equipment and fear of infection. Previous studies have also con rmed the high levels of anxiety in those who had direct clinical contact with COVID-19 patients (18,21).
In the present study, prevalence of depression was 43%, which is much higher than what was reported in previous studies in other parts of the world. For instance, the prevalence of depression among healthcare workers was reported to be 8% in a study in Nepal (18) and 28% in a study in China (21). Moreover, in a meta-analysis, the combined prevalence was reported to be 22.8% (19). The high prevalence of depression in Iranian nurses is probably related to risk factors such as the high transmission risk of COVID-19 and the social isolation of nurses due to the fear of transmitting the disease to their family members and friends.
In this study, 17.4% of the nurses experienced some degrees of stress, which is consistent with the results of the study of Wu et al. (22). However, in the study of Mo et al. (2020) the mean stress score of Chinese nurses was reported to be 39.9%. The long work hours during a week as well as fear of getting infected by COVID-19 infection through the respiratory droplets and direct contact (3) increased the Chinese nurses' stress to a relatively higher level compared to Iranian nurses.
Generally, stress in health care workers ghting COVID-19 is attributed to long shifts, unrealistic wages, lack of personal protective equipment, and fear of getting infected or infecting one's family (23,24).
Work experience was the strongest variable associated with stress and anxiety among nurses who were involved with COVID-19 in a way that nurses who had more than 10 years of experience were at an increased risk of experiencing stress and anxiety compared to the nurses with less than 1o years of experience. Long work hours during a week is one of the factors affecting nurses' stress and anxiety (3). The Iranian nurses who were satis ed with their jobs reported less stress, anxiety and depression than those who were not interested in and satis ed with their jobs. In the study of Letavak et al. (2012) it was perceived that the nurses who have a lower level of job satisfaction get more depressed (25). There are other studies that show job satisfaction and professional commitment help to reduce stress and anxiety (26,27). Moreover, a history of depression emerged as a factor associated with stress and depression.
Many of the previous studies also showed that high stress levels in nurses lead to anxiety, frustration, depression and other mental and emotional disorders (28,29).
Marital status and having a child in the family were among factors affecting the nurses' stress. Having children causes stress in nurses because the children need care and also because there is a risk of transmission of COVID-19 from the nurse parent to the child, which is in line with the study of Mo (3).
Moreover, there was a signi cant relationship between marital status and nurses' stress, which is consistent with the study of Yildirim et al. (2020) (30).
The nurses who are older than 35 years are more likely to get stress and anxiety. Although in the study of Yildirim et al. (2020), it was shown that the young healthcare staff had higher anxiety scores compared to older ones (30). In addition, the nurses who are informally employed are less likely to experience stress and anxiety than the nurses who are formally employed. The type of employment is associated with workplace anxiety (31). It seems that informally employed nurses can quit their jobs more easily because they have no obligation to the organization. As a result, they have less stress and anxiety than formally employed nurses.
Violence in the workplace emerged as one of the factors associated with stress and anxiety in nurses involved with COVID-19. In the study of Tong et al. (2019), it was found that workplace violence was followed by symptoms of depression in nurses. Reducing the workplace violence and developing the psychological capital can be helpful in the ght against the symptoms of depression (32).
The nurses who exercise and have higher levels of physical activities are less likely to get depression than the nurses who do not exercise and have low levels of physical activities. Physical inactivity and sedentary behavior are signi cantly associated symptoms of depression and anxiety (33).
The risk of depression is higher in nurses who have a chronic disease or an abnormal sleep pattern. The study of Levatak et al. (2012) indicates that the nurses who have health problems are more likely to be depressed. In other words, there is a relationship between nurses' medical problems and their depression (25). The study of Wu showed that more than one third of the medical staff had symptoms of insomnia during the COVID-19 outbreak, which was associated to isolation and the psychological distress caused by the prevalence of the disease. Considering various psychological factors, it is necessary for the medical personnel to receive interventions in this regard (34). The distress caused by symptoms of anxiety affects the medical staff's sleep quality (35).
The nurses who ght against COVID-19 are generally under psychological pressure. The authorities should consider the nurses' individual characteristics (age, work experience, gender, marital status, having children and history of chronic diseases), improve their working conditions and provide opportunity for them to get enough sleep to help them modulate their stress, anxiety and depression. Moreover, the nurses should be taught appropriate sports and physical activities to do in their leisure time as well as relaxation techniques to maintain their mental health.

Conclusion
Prevalence of stress, anxiety and depression is high among nurses involved with caring for COVID-19 patients and it is associated with a number of their individual characteristics including age, work experience, gender, marital status, having children, job satisfaction, workplace violence, sleep, exercising and history of chronic diseases. Therefore, screening services and psychological supports should be provided for nurses and their working conditions should be improved to reduce their anxiety and distress.

Limitations
The present study has several limitations that should be noted. First, this study only investigated the nurses in the city of Babol in North of Iran. Therefore, the results cannot be generalized to all Iranian nurses. Second, since this was a cross-sectional study, it was not possible to observe the subjects in a longer period of time like a longitudinal study and it could only assess stress, anxiety, and depression at the time of the study. Third, only one questionnaire was used in this study because of the time constraints.