Sleep Disorders, Perceived Stress and Family Support Among Nursing Staff During the Pandemic Crisis

Background: The covid-19 pandemic is likely to cause mental health issues, especially for healthcare professionals. The aim of this study is to investigate levels of perceived stress, sleep disturbances and sense of family support among nurses in pandemic conditions. Methods: Athens Insomnia Scale (AIS), Perceived Stress Scale (PSS) and Family Support Scale (FSS) were administered in a sample of 150 nurses from different hospital departments during the COVID-19 pandemic. Individual and demographic data were recorded. Results: 120 women and 30 men completed the study questionnaires. Almost half of participants (49.7%) reported the presence of sleep diculties and more than half (50.3%) experienced increased stress levels. Scores on AIS scale correlated positively with PSS scores (P<0.01), and nega tively with FSS scores (P<0.01). A signicantly negative correlation was observed between the PSS scores and the FSS scores (p<0.01). Regression model showed ‘scores on PSS scale’ and ‘years of work experience’ were signicant predictors of ‘scores on AIS scale’, each explaining 43.6% and 2.3% of the variance. Scores on AIS scale’ and ‘scores on FSS scale’ were signicant predictors of PSS explaining 43.7% and 9,2% of the variance . Conclusion: The study conrmed that working with COVID-19 patients has a negative effect on the sleep of nurses, possibly mediated by increased levels of stress. Family support, as a protective factor, appears to moderate the deleterious consequences of stress.

infection during hospitalization, helplessness, fear, anxiety and concern for patients and family members and the need to make sense of the morally challenging decisions being made, can have dramatic effects on their physical and mental well-being. Their resilience can be further compromised by isolation and loss of social support, risk of contagion and infection of friends and relatives, as well as drastic, often annoying changes in working conditions. Previous studies have reported that healthcare staff especially those who work in the frontline during viral epidemic outbreaks are at high risk for developing mental health issues, including stress, anxiety, depressive symptoms, anger, insomnia, fear, and sleep disorders [3,4,5,6].
As the COVID-19 pandemic took hold, nurses were on the front line of health and social care in the most extreme of circumstances. In similar outbreaks, nurses have had the highest levels of occupational stress and resulting distress compared with other groups [7,8,9,]. Having to enter the negative pressure ward to care for the patients after only undergoing a brief training, working in the Intensive Care Units and the Infection Departments, spending hours each day putting on and removing airtight protective equipment, being transferred to other non-anti-epidemic positions and in the meanwhile worrying about becoming infected or infecting family members, at times may compromise safety and wellbeing and can lead to adverse mental health outcomes [10].
Family support refers to the sense of perceived support from the family environment. It is an important element of social support. Evidence on the in uence of family support on chronic illness selfmanagement behaviors shows that perceived family support is positively associated with improved outcomes among individuals with diabetes mellitus [11,12], chronic obstructive pulmonary disease [13], even in lung cancer patients [14]. Also, research highlights the negative correlation between family support and depression [15], while another study reported a negative correlation between family support and burnout [16]. The role of family support is likely to become more in uential in situations where entire societies are placed under tightened quarantine restrictions.
The aim of the study is to investigate the prevalence of sleep disorders and perceived stress and evaluate the effects of the sense of family support in a specialized COVID-19 hospital nursing staff during the pandemic crisis.

Research Design
This cross-sectional study was conducted in one of the largest hospitals in Greece after approval from the Clinical Research Ethics Committee. The hospital received the rst suspected Covid-19 case in February 2020 and remained a specialized hospital treating exclusively COVID-19 patients until May 2020. The study took place in the second half of May.

Study participants
One hundred and fty nurses (out of 679) were randomly selected, and agreed to participate anonymously by completing self-reported questionnaires. Participants worked in departments of respiratory medicine, fever clinics, intensive care units or in other non-anti-epidemic positions, during past month.

Measurement Tools
Demographic and social data from the study participants included age, gender, education, and marital status. Professional and work information included their title, role, department that they worked during last month, and work experience.

Athens Insomnia Scale (AIS)
The severity of sleep disturbances was measured by the Athens Insomnia Scale (AIS) [17]. The scale is a self-assessment psychometric instrument designed for quantifying sleep di culty based on the ICD-10 criteria. It consists of eight items: the rst ve pertain to sleep induction, awakenings during the night, nal awakening, total sleep duration, and sleep quality; while the last three refer to well-being, functioning capacity, and sleepiness during the day. In this scale, the sleep di culty severity is measured based on a 4-point Likert scale, since last month. The scores ranged from 0 (meaning not being a problem) to 3 (meaning more acute sleep di culties). A cut-off score of 6 has been determined to distinguish between insomniacs and healthy subjects [18].

Perceived Stress Scale (PSS)
PSS was used to evaluate the perception of stressful experiences. The 10-item PSS [19] measures global perceived stress experienced, by asking the respondent to rate the frequency of his/her feelings and thoughts related to events and situations that occurred across the past 30 days, on a 5-point scale (0never, 1 = almost never, 2 = once in a while, 3 = often, 4 = very often). Total scores range from 0 to 40, with higher scores indicating higher perceived stress. Scores ranging from 0-13 would be considered low stress, from 14-26 moderate stress and from 27-40 high perceived stress. As the PSS is not a diagnostic instrument, there are no cut-off scores.

Family Support Scale (FSS)
To evaluate perception of family support we used the family support scale [20] which aims to record the sense of support that a subject receives from the members of his/her family (with whom he/she lives). The scale consists of 13 items, which are answered on a Likert scale, ranging from 1 ("I disagree a lot") to 5 ("I agree a lot"). The scale is self-administered and all of the items focus on the interrelations of individuals that live together. High scores correspond to an increased sense of family support. Individuals that live alone did not complete the scale.

Statistical Analysis
SPSS software, version 20, was used for the statistical analysis. All variables were assessed with the use of descriptive statistics and values were expressed as the mean ± standard deviation for continuous variables. The prevalence of insomnia and perceived stress levels were determined in percentages. All distributions of the continuous variables were normal (One -Sample Kolmogorov-Smirnov Test, p > 0.05). Independent t-test, one sample t test, ANOVA, Pearson's correlation and stepwise linear regression analysis were used for the evaluation of data. Statistical signi cance was set at P < 0.05 (two-tailed).

Results
A total of 150 nurses (120 women and 30 men) completed the study questionnaires. The demographic and working data of the study participants are shown in Table 1. No statistically signi cant difference was noted between men and women regarding age and years of work experience (Independent t-test, p > 0.05). Also, no statistically signi cant gender differences were observed in mean scores on AIS, PSS and FSS scales ( Table 2, p > 0.05).  Males stated increased sense of family support with a mean value of 54.5 on FSS scale, a result that differs signi cantly from the reference value of 50.8 [20] in the standardized version of the scale in a sample of Greek healthcare professionals (one sample t test p < 0.05) and similarly mean FSS value in female participants was signi cantly different from the corresponding reference value of 47 [20] in the standardized version of the scale (52.16 versus 47, one sample t test p < 0.01).
Scores on AIS scale correlated positively with increasing age and years of work experience (p < 0.05, Table 3). There was a signi cant positive correlation between the AIS scores and the PSS scores (P < 0.01), and a negative correlation between the AIS scores and the FSS scores (P < 0.01). Also, a signi cantly negative correlation was observed between the PSS scores and the FSS scores (p < 0.01, Table 3).  On the basis of the results of the bivariate analyses, a stepwise multiple regression test was performed to identify the best predictors of the dependent variable 'scores on PSS scale' among the independent variables that showed signi cant associations (scores on FSS scale, scores on AIS scale) and to examine their contribution to the variation (expressed as R 2 ) in the dependent variable. The nal regression model showed that 'scores on AIS scale' and 'scores on FSS scale' were signi cant predictors of the dependent variable explaining 43.7% and 9,2% of the variance (F 1,116 =65.040, p = 0.000, Table 5).

Discussion
This cross-sectional study revealed a high prevalence of sleep disorders among nurses working in a specialized COVID-19 hospital, in the past month, during the pandemic. A recent study in the Greek population, using the AIS scale to explore sleep di culties during the COVID-19 pandemic, reported sleep problems in 36.7% of the general population [21]. A survey of insomnia among medical staff members in China found that more than one-third of the medical staff suffered insomnia symptoms during the COVID-19 outbreak [22]. In another study [23] scores on AIS scale to evaluate sleep disturbances of Chinese frontline medical workers were not statistically different from the ndings of our study (6.3 ± 4.2 versus 5.95 ± 4.238, p > 0.05).
More than half of nursing staff who responded to questionnaires expressed moderate to high levels of stress. Literature suggests that nursing staff are psychologically more vulnerable to the pressure caused by the pandemic. These results may be partly biased because of the gender effect [24], taking into account that the majority of nurses are females, but could also be attributed to the fact that they may face extreme challenges. A higher risk of exposure to patients with COVID-19 as they spend more time in wards providing nursing care to patients and being responsible for the collection of sputum for the detection of viruses or working continuously and intensively in the isolation wards justify the highest levels of occupational stress and resulting distress compared with other groups. In addition, due to their intimate contact with patients, they may be more exposed to moral injuries associated with suffering, death and ethic dilemmas [25].
Results from linear regression analysis in this study suggested that the effect of the coronavirus outbreak on sleep di culty has partly been mediated by stress. Research indicates that stress is closely related to sleep quality. The HPA axis, central catecholamine systems and sympathetic system play an important role in the regulation of the sleep-wake cycle and there is good evidence that dysregulation of the neural and neuroendocrine mediators of the stress response can lead to sleep disturbances [26]. In turn, sleep disorders may appear to lead to further HPA axis dysregulation, thereby promoting a vicious cycle of stress and insomnia [27]. From another point of view, there is evidence of a bidirectional association between insomnia and loneliness [28], providing an explanation for the positive effects of family support.
The role of perceived family support seems to be important in moderating stress levels possibly by reducing the perception of the threat of stressful events and the physiological response and inappropriate behavior that can result from stress, leading to improved self-e cacy and a sense of professional achievement and acting protectively against the development of burnout [29]. Also, increased sense of family support, reported by our study, in the midst of a pandemic, may be related to the recognition of the absolute need and the signi cant contribution of the nursing profession.
The message from the World Health Organization for healthcare workers forewarned that managing mental health and psychosocial wellbeing during this pandemic is as important as managing physical health. In other words, even before the crisis begins to recede mental health issues need to be addressed in order to avoid long-lasting impacts on healthcare workers.

Conclusion
High rates of sleep disturbances were observed among nurse participants during the pandemic period. Reported stress levels correlate signi cantly with sleep disorders. Sense of family support is negatively associated with perceived stress and insomnia.